AU2024202145B1 - Detection device and detection method of machine-picked seed cotton defects based on deep learning - Google Patents
Detection device and detection method of machine-picked seed cotton defects based on deep learning Download PDFInfo
- Publication number
- AU2024202145B1 AU2024202145B1 AU2024202145A AU2024202145A AU2024202145B1 AU 2024202145 B1 AU2024202145 B1 AU 2024202145B1 AU 2024202145 A AU2024202145 A AU 2024202145A AU 2024202145 A AU2024202145 A AU 2024202145A AU 2024202145 B1 AU2024202145 B1 AU 2024202145B1
- Authority
- AU
- Australia
- Prior art keywords
- component
- machine
- seed cotton
- detection
- rod
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
- B65G15/00—Conveyors having endless load-conveying surfaces, i.e. belts and like continuous members, to which tractive effort is transmitted by means other than endless driving elements of similar configuration
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
- B65G43/00—Control devices, e.g. for safety, warning or fault-correcting
- B65G43/08—Control devices operated by article or material being fed, conveyed or discharged
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
- B65G47/00—Article or material-handling devices associated with conveyors; Methods employing such devices
- B65G47/02—Devices for feeding articles or materials to conveyors
- B65G47/16—Devices for feeding articles or materials to conveyors for feeding materials in bulk
- B65G47/18—Arrangements or applications of hoppers or chutes
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8806—Specially adapted optical and illumination features
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8806—Specially adapted optical and illumination features
- G01N2021/8835—Adjustable illumination, e.g. software adjustable screen
Landscapes
- Analytical Chemistry (AREA)
- Immunology (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Biochemistry (AREA)
- Pathology (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Treatment Of Fiber Materials (AREA)
- Preliminary Treatment Of Fibers (AREA)
Abstract
A detection device of machine-picked seed cotton defects based on deep learning is
provided, and a technical field of cotton detection is related, including a frame, a material
delivery component, a feeding component, a blanking component and a detection component. A
detection method of machine-picked seed cotton defects based on deep learning is also provided,
including the following steps: putting in raw materials, defect detecting, blanking and adaptive
debugging. The detection component is located above the material delivery component, when the
machine-picked seed cotton follows the material delivery component and moves to the below the
detection component, the collection of images by the detection component can be completed, and
defects in the machine-picked seed cotton can be detected by analyzing and calculating the
images. The light guide block and the lamp tube are arranged inside the material delivery
component, the light rays can be supplemented during collecting images of machine-picked seed
cotton by cooperating the light guide block and the lamp tube, to ensure an accuracy of the
collection of images, and thus a detection accuracy of machine-picked seed cotton is improved
while ensuring the detection efficiency of machine-picked seed cotton.
5
FIG. I
Description
TECHNICAL FIELD The present disclosure relates to a technical field of cotton detection, and in particular to a detection device and detection method of machine-picked seed cotton defects based on deep learning.
BACKGROUNDTECHNOLOGY Lint cotton defects refer to two categories: impurities with fibers and fibers that hinder spinning, including seven types: broken seeds, infertile seeds, cords, soft seed skins, stiff flakes, fibrous seed scraps and cotton neps. Since defects are difficult to remove during a spinning process, residual impurities and defects are wrapped in a yam or attached to a surface of the yarn, and a consequence is a deterioration of dryness, an increase of knots and impurities, and an increase of fluffy yarn. A surface of a gray cloth made from yam with a large amount of fiber seeds shows a large number of neps and impurities, after dyeing, sterile color spots and hair particles appear on a cloth surface, and stiff and rough to touch, and thus affecting a development of a cotton textile industry and causing serious losses to a national economy. Therefore, rapid detection of lint cotton defects content is crucial for determining a quality grade of raw cotton, and a quality of cotton ginning is divided into three grades: good, medium and poor according to the defect content. At present, most domestic enterprises generally use a visual inspection method to identify defects. A test accuracy of the visual inspection method is often affected by human factors, moreover, for small defects, the false detection rate and missed detection rate of detection results are relatively high, and it is labor-consuming and time-consuming. At the same time, a large amount of lint and dust will be generated during a detection process, so detection personnel are often affected by an external environment, and human eyes are easily fatigued after working for a long time, resulting in a decrease in accuracy. A cotton defect detection and identification method based on multispectral technology with a publication number CN104751443B, including the steps of: acquisiting image; preprocessing the image; preliminary detecting defects: using an improved morphological gradient algorithm to perform edge detection, to obtain a comprehensive edge detection intensity of each defect, and obtain a contrast-enhanced image of each defect; precise locating of defects: converting an image into a connected domain image after processed by improved iterative thresholding and morphology, precisely locating of defects by selecting a main defect algorithm, and extracting defect areas by marking; judging the quality of cotton ginning: counting the number of each defect, calculating a total number of cotton defects, and obtaining a grade of the quality of cotton ginning. The present disclosure has the advantages of simple principle, fast detection speed and high detection accuracy.
There are some problems in a practical application of the above technical solutions. A
detection of machine-picked seed cotton is carried out through image recognition, in order to
ensure accuracy of image recognition, the machine-picked seed cotton needs to be transported
separately and quantitatively, making it difficult to detect defects in large quantities of seed
cotton. Moreover, an equipment requires continuous debugging by staff during use, making it
difficult to achieve autonomous learning and debugging.
Therefore, it is necessary to invent a detection device and detection method of
machine-picked seed cotton defects based on deep learning to solve the above problems.
SUMMARY OF THE PRESENT DISCLOSURE A purpose of the present disclosure is to provide a detection device and detection method of machine-picked seed cotton defects based on deep learning, to solve the problems raised in the background technology. In order to achieve the above purpose, the present disclosure provides a detection device of machine-picked seed cotton defects based on deep learning, including a frame, wherein a feeding component, a material delivery component and a blanking component is located inside the frame in sequence from right to left, and a detection component fixedly connected to the frame is located above the material delivery component. The detection component includes a multispectral camera. The material delivery component includes a fixed cylinder, a plurality offixing brackets are arranged and distributed in an array around the outer side of the fixed cylinder, a limit frame is located at both ends of the fixed cylinder, a light guide block is located between two adjacent fixing brackets, and a plurality of fixed blocks distributed in a rectangular array are located above the light guide block. The feeding component includes a mounting frame located on the right side of the fixed cylinder, and an adjustment component is fixedly located on the inner top of the mounting frame. The adjustment component includes a fixed rod fixed to the inner top of the mounting frame, a plurality of sliding seats corresponding to the plurality of fixed blocks and distributed in an array are slidingly connected to the fixed rod, a movable rod parallel to the side of the fixed cylinder is located at the bottom of the sliding seat, an elastic cloth is fixed between the movable rods on both sides and the mounting frame, an expansion piece is located at both ends of each of the plurality of the sliding seats, an air pump fixedly connected to the top of the mounting frame is located above the plurality of the sliding seats, and the air pump is connected to the expansion piece through a conduit.
A detection roller is elastically and slidingly connected to the inner top of the mounting frame, the detection roller is not in contact with the fixed rod, and both ends of the detection roller are in contact with the side of the limiting frame, and a plurality of pressure-sensitive elements distributed in an array are arranged on the outer peripheral side wall of the detection roller. The plurality of fixed blocks above the light guide plate are used as identification points by the detection component, an area above the light guide plate are divided into multiple intervals distributed in a rectangular array, and seed cotton inside the multiple intervals is detected and identified by the detection component; at the same time, since the movable rod corresponds to a lateral interval above the light guide plate, a problem with a light transmittance in a certain lateral interval is detected by the detection component, and at this time, the expansion piece corresponding to the lateral interval is controlled to expand, and the sliding seat is pushed to move left or right by the expansion piece, and thus a spacing between the movable rods corresponding to the lateral interval is adjusted. In one embodiment, the movable rod is a hollow structure, a through hole corresponding to the expansion piece is located on the movable rod, and a solenoid valve is located in each through hole. When the spacing between the movable rods is adjusted, the solenoid valve corresponding to the movable rod requiring adjustment is controlled to open, and the remaining solenoid valves are controlled to close, to avoid an incorrect movement of the movable rod not requiring adjustment. In one embodiment, the expansion piece includes a corrugated pipe fixedly connected to the movable rod, an elastic member is arranged inside the corrugated pipe, one end of the elastic member is fixedly connected to the sliding seat, and the other end of the elastic member is fixedly connected to the corrugated pipe. After all raw materials are finished unloading, all the solenoid valves are opened, and at this time, the gas in the corrugated pipe is discharged under a restoring force of the elastic member, to allow the corrugated pipe to be restored to its original state.
In one embodiment, the material delivery component further includes a lamp tube with
serpentine cross-section fixedly connected under the light guide block, a light-transmitting plate
is fixedly above the light guide block, the outer side of the light-transmitting plate is fixedly
connected to the fixed blocks, and both sides of thefixed blocks are concave arc-shaped
structures.
When the machine-picked seed cotton passes through the feeding component and reaches
the surface of the light-transmitting plate, the machine-picked seed cotton can be attached to the
surface of the light-transmitting plate under an action of the plurality of fixed blocks, to realize
the grabbing and conveying of machine-picked seed cotton by the material delivery component.
In one embodiment, a rotating rod is disposed throughout the middle part of the fixed
cylinder, a rotating disk is fixedly located at both ends of the rotating rod, and a plurality of
chutes are arranged around the outer side of the rotating disk.
A dual-axis motor is located below the fixed cylinder, and a transmission rod is fixedly
located at both ends of the dual-axis motor, a connecting seat fixedly connected to the frame is
movably located at one end of the transmission rod by a bearing, a transmission disk is movably
located at the top of the connecting seat by a bearing, a belt pulley is fixedly located on each of
the outer side of the transmission disk and one end of the transmission rod, and a transmission
L0 belt is located between two belt pulleys, a rocker arm is fixedly located on the outer side of the
transmission disk, and a sliding rod is fixedly located on the top of the rocker arm.
The transmission rod is driven to rotate by the dual-axis motor, the transmission disk is
driven to rotate by the transmission rod through the belt pulley, the rocker arm is driven to rotate
by the transmission disk, and the sliding rod is driven to rotate by the rocker arm, to make the
sliding rod squeeze the inner wall of the chute, thereby causing the rotating disk to rotate. The
fixed cylinder can be driven to rotate by the rotating disk through the rotating rod, thereby, the
light-transmitting plate can be driven to rotate to realize a transportation of raw materials. Since
the sliding rod and the chute are intermittently matched, the light-transmitting plate can be driven
to rotate intermittently by the fixed cylinder, and thus the raw materials can stay briefly when
moving below the detection component.
In one embodiment, the feeding component further comprises a feeding cover fixedly located inside of the mounting frame, wherein a movable board is elastically and slidingly connected to a right inner wall of the feeding cover, and a combing board is located on a top of the movable board; and the bottom of the movable board is extended out of the feeding cover and is in contact with the outer peripheral side wall of the limiting frame.
Since an end of the movable board away from the combing plate is always in contact with
the outer peripheral side wall of the limiting frame, when the limiting frame is driven to rotate by
the fixed cylinder, the movable board is pushed by the limiting frame, the comb plate is driven
by the movable board to stir the seed cotton raw materials in the feeding cover, and to make a
distribution of the seed cotton raw materials in the feeding cover in a balanced state.
In one embodiment, the blanking component includes a blanking conveyor belt located on
the left side of the fixed cylinder, wherein a mounting rod is located at the right end of the
blanking conveyor belt, a mounting seat is movably located at both ends of the mounting rod
through a bearing, a torsion spring is located between the mounting seat and the mounting rod, a
plurality of blanking blocks are fixedly located on the outer side of the mounting rod, an
extension seat is fixedly located on the outer sides of the blanking blocks located on both sides, a
pulley is movably located on the outer side of the extension seat through a bearing, and the
pulley is adhered to the outer wall of the limiting frame.
The mounting rod is subjected to a counterclockwise rotational force under an action of the
torsion spring, the blanking block and the pulley can be driven to rotate by the mounting rod,
causing the pulley to remain adhered to the limit frame. At this time, the blanking block remains
adhered to the surface of a light guide plate, and at this time, the raw materials on the surface of
the light-transmitting plate can be peeled off by the blanking block, and the peeled raw materials
can be unloaded under the action of the blanking conveyor belt.
In one embodiment, one end of the blanking block close to the fixed cylinder is an inclined
wedge structure.
The wedge structure ensures that the machine-picked seed cotton can be peeled off at an
inclined angle by the blanking block. In one embodiment, a control system is located on the outside of the frame, and the control system is configured to control electric components and pneumatic components inside the frame.
A detection method of machine-picked seed cotton defects, applicable to any one of the above detection device of machine-picked seed cotton defects based on deep learning, and the detection method comprising:
step 1, putting in raw materials, putting the raw materials of machine-picked seed cotton into the feeding component, and transporting the raw materials in the feeding component to the bottom of the multispectral camera through the material delivery component; step 2, defect detecting, transporting the machine-picked seed cotton to the bottom of the multispectral camera by the material delivery component, and remaining a position fixed, at this time, collecting images of machine-picked seed cotton above the material delivery component by the multispectral camera, and then analyzing the collected images to realize defect detection; step 3, blanking, the raw materials after detecting continue following the movement of the material delivery component, until the raw materials move to the top of the blanking component, and then blanking the raw materials under the action of the blanking component; step 4, adaptive debugging, in the step 2, determining a situation of light rays passing through the machine-picked seed cotton above the material delivery component by the multispectral camera based on a light transmittance degree of machine-picked seed cotton in the collected images, and controlling the feeding component according to a passage of light rays, to control a release process of the raw materials of the feeding component, and to make both an efficiency of machine-picked seed cotton transferred by the material delivery component and an accuracy of defect detection reach an optimal value. Technical effects and advantages of the present disclosure: (1) The detection device of machine-picked seed cotton defects based on deep learning provided by the present invention includes the feeding component, the material delivery component and the blanking component, and through a cooperation of the feeding component, the material delivery component and the blanking component, a cyclic transportation of machine-picked seed cotton can be realized. The detection component is located above the material delivery component, when the machine-picked seed cotton follows the material delivery component and moves to the below the detection component, the collection of images by the detection component can be completed, and defects in the machine-picked seed cotton can be detected by analyzing and calculating the images. The light guide block and the lamp tube are arranged inside the material delivery component, the light rays can be supplemented during collecting images of machine-picked seed cotton by cooperating the light guide block and the lamp tube, to ensure the accuracy of the collection of images, and thus the detection accuracy of machine-picked seed cotton is improved while ensuring the detection efficiency of machine-picked seed cotton.
(2) The feeding component of the present disclosure includes a mounting frame, a plurality
of sliding seats that can slide horizontally is located inside the mounting frame, a movable rod is
located at the bottom of each of the plurality of sliding seats, and an amount of the feeding
component can be adjusted and controlled by adjusting the spacing between multiple movable
rods. Therefore, an adjustment of a conveying capacity of machine-picked seed cotton by the
material delivery component can be realized; and a thickness of machine-picked seed cotton
when the detection component acquires images can be adjusted, and thus the accuracy of defect
detection in machine-picked seed cotton can be ensured.
(3) The detection device of machine-picked seed cotton defects based on deep learning
provided by the present invention includes the material delivery component, the feeding
component and the detection component, and through a cooperation of the material delivery
component, the feeding component and the detection component, a transportation of
machine-picked seed cotton can be realized. The defects in machine-picked seed cotton can be
detected by the detection component; and the situation of the light rays above the material
delivery component passing through the machine-picked seed cotton can be detected by the
LO detection component, and the feeding efficiency of the machine-picked seed cotton of the
feeding component can be adjusted according to the penetration conditions of the light rays,
thereby ensuring both an efficiency of machine-picked seed cotton transferred by the material
delivery component and an accuracy of defect detection reach an optimal value.
(4) The detection device of machine-picked seed cotton defects based on deep learning
provided by the present invention includes a detection roller, when the light-transmitting plate
attached with seed cotton raw materials comes into contact with the detection roller, on one hand,
the seed cotton raw materials attached to the light-transmitting plate are squeezed by the
detection roller, to make the seed cotton raw materials compacted, whether there are cotton seeds
in the seed cotton raw materials can be detected by the pressure-sensitive elements on the outer
peripheral side wall of the detection roller, and the positions of the cotton seeds can be determined according to a torque of the fixed cylinder; on the other hand, compacted cotton has fewer internal voids, and light rays travel in a relatively straight path when passing through the cotton, and the light rays will not suffer too much scattering, therefore, the detection accuracy of the detection component when taking photos of seed cotton raw materials is improved, and misjudgment will not be caused when performing a transmittance analysis. In a further aspect, there is provided a detection device of machine-picked seed cotton defects based on deep learning, comprising a frame, wherein a feeding component, a material delivery component and a blanking component are located inside the frame in sequence from right to left; and a detection component is fixedly connected to the frame and located above the material delivery component, the material delivery component comprises a fixed cylinder and a lamp tube with serpentine cross-section, wherein a plurality of fixing brackets are arranged and distributed in an array around an outer side of the fixed cylinder; a limit frame is located at both ends of the fixed cylinder; a light guide block is located between two adjacent fixing brackets; and a plurality of fixed blocks are distributed in a rectangular array and located above the light guide block, and the lamp tube is fixedly connected under the light guide block; the feeding component comprises a mounting frame located on a right side of the fixed cylinder; and an adjustment component is fixedly located on an inner top of the mounting frame; the detection component comprises a multispectral camera, the multispectral camera is L0 fixedly located at a top of an inside of the frame, the feeding component is used to transport raw materials of machine-picked seed cotton to the bottom of the multispectral camera through the material delivery component, the multispectral camera is used to collect images of the machine-picked seed cotton above the material delivery component, and then analyze collected images to realize defect detection in the machine-picked seed cotton; the adjustment component comprises a fixed rod fixed to the inner top of the mounting frame, wherein a plurality of sliding seats corresponding to the plurality of fixed blocks and distributed in an array are slidingly connected to the fixed rod; a movable rod parallel to a side of the fixed cylinder is located at a bottom of each of the plurality of sliding seats; an elastic cloth is fixed between the movable rod on both sides and the mounting frame; an expansion piece is located at both ends of each of the plurality of sliding seats; and an air pump isfixedly connected to the top of the mounting frame and located above the plurality of sliding seat, and the air pump is connected to the expansion piece through a conduit; a detection roller is elastically and slidingly connected to the inner top of the mounting frame, the detection roller is not in contact with the fixed rod, and both ends of the detection roller are in contact with a side of the limiting frame; and a plurality of pressure-sensitive elements distributed in an array are arranged on an outer peripheral side wall of the detection roller; the movable rod is a hollow structure, a through hole corresponding to the expansion piece is located on the movable rod, and a solenoid valve is located in each through hole; the expansion piece comprises a corrugated pipe fixedly connected to the movable rod, wherein an elastic member is arranged inside the corrugated pipe, one end of the elastic member is fixedly connected to the sliding seat, and the other end of the elastic member is fixedly connected to the corrugated pipe; a rotating rod is disposed throughout a middle part of the fixed cylinder; a rotating disk is fixedly located at both ends of the rotating rod; and a plurality of chutes are arranged around an outer side of the rotating disk and distributed in an array; a dual-axis motor is located below the fixed cylinder; a transmission rod isfixedly located at both ends of the dual-axis motor; a connecting seat fixedly connected to the frame is movably located at one end of the transmission rod by a first bearing; a transmission disk is movably L0 located at the top of the connecting seat by a second bearing; a belt pulley isfixedly located on each of an outer side of the transmission disk and one end of the transmission rod; a transmission belt is located between two belt pulleys; a rocker arm is fixedly located on the outer side of the transmission disk; and a sliding rod adapted to each of the plurality of chutes arefixedly located on the top of the rocker arm; the blanking component comprises a blanking conveyor belt located on a left side of the fixed cylinder, wherein a mounting rod is located at a right end of the blanking conveyor belt; a mounting seat movably is located at both ends of the mounting rod through a third bearing; a torsion spring is located between the mounting seat and the mounting rod; a plurality of blanking blocks are fixedly located on an outer side of the mounting rod; an extension seat is fixedly located on an outer side of the blanking blocks located on both sides; and a pulley movably is located on an outer side of the extension seat through a fourth bearing, wherein the pulley is adhered to an outer wall of the limiting frame; a control system is located on the outside of the frame, and the control system is configured to control electric components and pneumatic components inside the frame; and the multispectral camera is also used to determine a situation of light rays passing through the machine-picked seed cotton above the material delivery component based on a light transmittance degree of machine-picked seed cotton in the collected images, and control the feeding component according to a passage of light rays, to control a release process of the raw materials of the feeding component.
BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic diagram of an overall structure according to an embodiment of the present disclosure; FIG. 2 is a structural diagram of a material delivery component, a feeding component and a blanking component according to an embodiment of the present disclosure; FIG. 3 is a structural diagram of a rotating rod according to an embodiment of the present disclosure; FIG. 4 is a structural diagram of the material delivery component according to an embodiment of the present disclosure; FIG. 5 is a structural diagram of a light guide block according to an embodiment of the present disclosure; FIG. 6 is a structural diagram of a light-transmitting plate according to an embodiment of the present disclosure; FIG. 7 is an enlarged schematic diagram of the structure at A of FIG.1; FIG. 8 is a structural diagram of the blanking component according to an embodiment of the present disclosure; FIG. 9 is a structural diagram of the feeding component according to an embodiment of the present disclosure; FIG. 10 is a structural diagram of a mounting frame according to an embodiment of the present disclosure; FIG. 11 is a structural diagram of a sliding seat according to an embodiment of the present disclosure; FIG. 12 is a side diagram of the overall structure according to an embodiment of the present disclosure; FIG. 13 is a schematic diagram of an overall structure according to embodiment 2 of the present disclosure; FIG. 14 is an enlarged schematic diagram of the structure at B of FIG.13.
In the drawings: 1, frame ; 2, material delivery component; 3, feeding component; 4, blanking component; 5,
detection component; 6, adjustment component; 7, detection roller; 201, fixed cylinder; 202, fixing bracket; 203, limiting frame; 204, light guide block; 205, lamp tube; 206, light-transmitting plate; 207, fixed block; 208, rotating rod; 209, rotating disk; 210, chute; 211, dual-axis motor; 212, transmission rod; 213, connecting seat; 214, transmission disk; 215, belt pulley; 216, rocker arm; 217, sliding rod; 301, mounting frame; 302, feeding cover; 303, movable board; 304, combing board; 401, blanking conveyor belt; 402, mounting rod; 403, mounting seat; 404, torsion spring; 405, blanking block; 406, extension seat; 407, pulley; 601, fixed rod; 602, sliding seat ; 603, movable rod; 604, elastic cloth; 605, expansion piece; 606, air
pump; 607, conduit.
DETAILED DESCRIPTION Technical solutions in embodiments of the present disclosure will be clearly and completely described below, in conjunction with the accompanying drawings in the embodiments of the present disclosure. Obviously, the embodiments described herein are just some of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by a person of ordinary skill in the art without inventive labor fall within a scope of protection of the present disclosure.
The present disclosure provides a detection device of machine-picked seed cotton defects
based on deep learning as shown from FIG. I to FIG. 12, including a frame 1, a material delivery
component 2, a feeding component 3, a blanking component 4 and a detection component 5. The
blanking component 4, the material delivery component 2 and the feeding component 3 are
arranged in sequence from left to right. The detection component 5 is located above the material
delivery component 2; and the material delivery component 2, the feeding component 3, the
blanking component 4 and the detection component 5 are located inside the frame 1.
The material delivery component 2 includes a fixed cylinder 201, and the fixed cylinder 201
is located inside the frame 1.
Specifically, a plurality of fixing brackets 202 are arranged around the outer side of the
fixed cylinder 201, a light guide block 204 is located between two adjacent fixing brackets 202,
and the light guide block 204 has a tapered structure. A lamp tube 205 with a snake-shaped
structure is fixedly located in the middle part of one side of the light guide block 204 close to the
fixed cylinder 201, and a light-transmitting plate 206 is fixedly located on the other side of the
light guide block 204. A plurality of fixed blocks 207 are fixedly located on the outer surface of
the light-transmitting plate 206, both sides of the fixing block 207 are concave curved structures,
and the plurality of fixed blocks 207 are distributed in a rectangular array.
When machine-picked seed cotton passes through the feeding component 3 and reaches the
surface of the light-transmitting plate 206, the machine-picked seed cotton can be attached to the
surface of the light-transmitting plate 206 under an action of the plurality of fixed blocks 207,
therefore, grabbing and conveying of the machine-picked seed cotton by the material delivery
component 2 can be realized.
More specifically, a limiting frame 203 is fixedly located at both ends of the fixed cylinder
201, a rotating rod 208 is disposed throughout the middle part of the fixed cylinder 201, a
rotating disk 209 is fixedly located at both ends of the rotating rod 208, and a plurality of chutes
210 are arranged around the outer side of the rotating disk 209. A dual-axis motor 211 is located
below the fixed cylinder 201, and a transmission rod 212 is fixedly located at both ends of the
dual-axis motor 211, a connecting seat 213 is movably located at one end of the transmission rod
L0 212 by a bearing, and the connecting seat 213 is fixedly located inside the frame 1. A
transmission disk 214 is movably located at the top of the connecting seat 213 by a bearing, a
belt pulley 215 is fixedly located on each of the outer side of the transmission disk 214 and one
end of the transmission rod 212, and a transmission belt is located between two belt pulleys 215.
A rocker arm 216 is fixedly located on the outer side of the transmission disk 214, and a sliding
rod 217 is fixedly located on the top of the rocker arm 216. The sliding rod 217 is adapted to the
chutes 210.
The transmission rod 212 is driven to rotate by the dual-axis motor 211, the transmission
disk 214 is driven to rotate by the transmission rod 212 through the belt pulley 215, the rocker
arm 216 is driven to rotate by the transmission disk 214, and the sliding rod 217 is driven to
rotate by the rocker arm 216, to make the sliding rod 217 squeeze the inner wall of the chute 210, thereby causing the rotating disk 209 to rotate. The fixed cylinder 201 can be driven to rotate by the rotating disk 209 through the rotating rod 208, thereby, the light-transmitting plate 206 can be driven to rotate to realize a transportation of raw materials. Since the sliding rod 217 and the chute 210 are intermittently matched, the light-transmitting plate 206 can be driven to rotate intermittently by the fixed cylinder 201, so that the raw materials can stay briefly when moving below the detection component 5. The feeding component 3 includes a mounting frame 301, the mounting frame 301 is located on one side of the fixed cylinder 201, and the mounting frame 301 is fixedly located inside the frame 1. Specifically, a slide rail is fixedly located on the inner top of the mounting frame 301, a plurality of slide seats 602 are disposed throughout the inside of the slide rail, a limit seat is fixedly located on both ends of the sliding seat 602, and a movable rod 603 is fixedly located at a bottom end of the sliding seat 602. A passing efficiency of machine-picked seed cotton can be adjusted by a cooperation of a plurality of movable rods 603. More specifically, a positioning block is fixedly located on the top of the sliding seat 602, and a positioning rod is fixedly located on the top of the positioning block. A positioning plate is located above the slide rail, and a plurality of positioning grooves are disposed throughout the middle part of the positioning plate. The plurality of positioning grooves are inclined in sequence L0 from the middle to both sides, and the positioning grooves located on the outside has a larger inclination angle. The greater the inclination angle of the positioning groove, the greater a horizontal movement distance of the corresponding positioning rod, therefore, an uniform adjustment of a spacing between multiple positioning rods can be realized. Moreover, a threaded sleeve is fixedly located in middle part of the top of the positioning plate, an adjusting screw is disposed throughout the inside of the threaded sleeve, and the top of the adjusting screw is disposed throughout the middle part of the top of the mounting frame 301 by a bearing. An adjusting motor is fixedly located on the top of the adjusting screw, and the adjusting motor is fixedly located on the middle part of the top of the mounting frame 301. The adjusting motor can carry out control according to a detection situation of the detection component 5, the adjusting screw can be driven to rotate by the adjusting motor, and the adjusting screw cooperates with the threaded sleeve to make the positioning plate move up and down. When the positioning plate moves up and down, the positioning rod can be squeezed by the inclined positioning groove, so that a horizontal position of the positioning rod can be adjusted, and a spacing between the plurality of movable rods 603 can be adjusted.
Moreover, a feeding cover 302 is fixedly located inside of the mounting frame 301, and an
elastic cloth 604 is fixedly located between the movable rods 603 on both sides and the mounting
frame 301.
The elastic cloth 604 ensures that the machine-picked seed cotton will not pass through both
sides of the mounting frame 301 when adjusting the position and spacing of the movable rods
603. The blanking component 4 includes a blanking conveyor belt 401, the blanking conveyor
belt 401 is located on the other side of the fixed cylinder 201, and the blanking conveyor belt 401
is fixedly located inside the frame 1.
Specifically, a mounting rod 402 is located at the upper end of the blanking conveyor belt
401, a mounting seat 403 is movably located at both ends of the mounting rod 402 through a
bearing, the mounting seat 403 is fixedly located inside the frame 1, and a torsion spring 404 is
located between the mounting seat 403 and the mounting rod 402. A plurality of blanking blocks
405 are fixedly located on the outer side of the mounting rod 402, an extension seat 406 is
L0 fixedly located on the outer side of the blanking blocks 405 located on both sides, a pulley 407 is
movably located on the outer side of the extension seat 406 through a bearing, and the pulley 407
is adhered to the outer wall of the limiting frame 203.
The mounting rod 402 is subjected to a counterclockwise rotational force under an action of
the torsion spring 404, the blanking block 405 and the pulley 407 can be driven to rotate by the
mounting rod 402, causing the pulley 407 to remain adhered to the limit frame 203. At this time,
the blanking block 405 remains adhered to the surface of a light guide plate, and at this time, the
raw materials on the surface of the light-transmitting plate 206 can be peeled off by the blanking
block 405, and the peeled raw materials can be unloaded under the action of the blanking
conveyor belt 401.
More specifically, one end of the blanking block 405 close to the fixed cylinder 201 is an inclined wedge structure.
The wedge structure ensures that the machine-picked seed cotton can be peeled off at an
inclined angle by the blanking block 405.
The detection component 5 includes a multispectral camera, and the multispectral camera is
fixedly located at the top of the inside of the frame 1. A collection of images can be completed
by the multispectral camera of the detection component 5, and defects in the machine-picked
seed cotton can be detected by analyzing and calculating the images.
During use, raw materials of the machine-picked seed cotton are put into the feeding
component 3, and the raw materials entering the feeding cover 302 slide throughout the movable
rods 603 along an inclined plane inside the feeding cover 302, and is attached to the fixed block
207 on the surface of the light-transmitting plate 206, and at this time, the grabbing of the raw
materials by the material delivery component 2 can be completed.
After the raw materials are grabbed by the material delivery component 2 ,the transmission
rod 212 is driven to rotate by the dual-axis motor 211, the transmission plate 214 is driven to
rotate by the transmission rod 212 through the pulley 215, the rocker arm 216 is driven to rotate
by the transmission plate 214, and the sliding rod 217 is driven to rotate by the rocker arm 216,
to make the sliding rod 217 squeeze the inner wall of the chute 210, and thereby causing the
rotating disk 209 to rotate. The fixed cylinder 201 can be driven to rotate by the rotating disk 209
through the rotating rod 208, therefore the light-transmitting plate 206 can be driven to rotate to
L0 realize the transportation of raw materials. Since the sliding rod 217 and the chute 210 are
intermittently matched, the light-transmitting plate 206 can be driven to rotate intermittently by
the fixed cylinder 201, so that the raw materials can stay briefly when moving below the
detection component 5. When the raw materials follow the material delivery component 2 and
move below the detection component 5, the lamp tube 205 is energized to emit light, the light
passes through the light guide block 204 and reaches the surface of the light guide plate, and thus
the bottom of the machine-picked seed cotton is illuminated, and at this time, the collection of
images by the multispectral camera of the detection component 5 can be completed, and defects
in the machine-picked seed cotton can be detected by analyzing and calculating the images.
When the defects in the machine-picked seed cotton are detected, a situation of the light
above the material delivery component 2 passing through the machine-picked seed cotton is determined according to a light transmittance degree of machine-picked seed cotton in the collected images, and the spacing between the movable rods 603 is controlled according to the passage of light rays, to control a release process of the raw materials of the feeding component 3, and thereby ensuring that both an efficiency of machine-picked seed cotton transferred by the material delivery component 2 and an accuracy of defect detection reach an optimal value. After the raw materials are detected, the raw materials are driven to continue move by the material delivery component 2. During a rotation process of the fixed cylinder 201, the limiting frame 203 rotates following the fixed cylinder 201, during this process, the mounting rod 402 is subjected to a counterclockwise rotation force under the action of the torsion spring 404; and the blanking block 405 and the pulley 407 can be driven to rotate by the mounting rod 402, thereby causing the pulley 407 to remain adhered to the limit frame 203. At this time, the blanking block 405 remains adhered to the surface of the light guide plate, and at this time, the raw materials on the surface of the light guide plate can be peeled off by the blanking block 405, and the peeled raw materials can be unloaded under the action of the blanking conveyor belt 401. Embodiment 2 Based on the above embodiment, although the above embodiment can control the feeding component 3 according to the transmittance detected by the detection component 5, to ensure the efficiency of transferring machine-picked seed cotton by the material delivery component 2 and an accuracy of defect detection; however, since the spacing between the movable rods 603 is L0 adjusted by using a motor to push the positioning plate to move, so that the spacing between the movable rods 603 can not be controlled individually, and the efficiency of transferring machine-picked seed cotton by the material delivery component 2 and the accuracy of defect detection can not be completely guaranteed. Moreover, the spacing between the movable rods 603 is adjusted according to the light transmittance of the detection component 5, if the cotton on the light-transmitting plate 206 is in a fluffy state when detected by the detection component 5, the light transmittance is reduced, and the detection component 5 misjudges, and thereby affecting the detection accuracy of the detection component 5. In view of this, improvements are made on the basis of Embodiment 1, and an improved technical solution is as follows: Referring to FIG. 13 and FIG. 14, a detection device of machine-picked seed cotton defects based on deep learning, including a frame 1. A feeding component 3, a material delivery component 2 and a blanking component 4 are located inside of the frame 1 and in sequence from right to left, and a detection component 5 fixedly connected to the frame 1 is located above the material delivery component 2.
The material delivery component 2 includes a fixed cylinder 201. A plurality of fixing
brackets 202 are arranged around the outer side of the fixed cylinder 201, a limiting frame 203 is
fixedly located at both ends of the fixed cylinder 201, and a light guide block 204 is located
between two adjacent fixing brackets 202.
The feeding component 3 includes a mounting frame 301 located on the right side of the
fixed cylinder 201, and an adjustment component 6 is fixedly located on the inner top of the
mounting frame 301.
The adjustment component 6 includes a fixed rod 601 fixed to the inner top of the mounting
frame 301, a plurality of sliding seats 602 corresponding to the plurality of fixed blocks 207 and
distributed in an array are slidingly connected to the fixed rod 601, a movable rod 603 parallel to
the side of the fixed cylinder 201 is located at the bottom of the sliding seat 602, an elastic cloth
604 is fixed between the movable rods 603 on both sides and the mounting frame 301, an
expansion piece 605 is located at both ends of the sliding seat 602, an air pump 606 fixedly
connected to the top of the mounting frame 301 is located above the sliding seat 602, and the air
pump 606 is connected to the expansion piece 605 through a conduit 607.
A detection roller 7 is elastically and slidingly connected to the inner top of the mounting
L0 frame 301, the detection roller 7 is not in contact with the fixed rod 601, and both ends of the
detection roller 7 are in contact with the sides of the limiting frame 203; and a plurality of
pressure-sensitive elements distributed in an array are arranged on the outer peripheral side wall
of the detection roller 7.
The fixed blocks 207 above the light guide plate are used as identification points by the
detection component 5, and an area above the light guide plate are divided into multiple intervals
distributed in a rectangular array, and the seed cotton inside the intervals is detected and
identified by the detection component 5. At the same time, since the movable rod 603
corresponds to a lateral interval above the light guide plate, a problem with the light
transmittance in a certain lateral interval is detected by the detection component 5, and at this
time, the expansion piece 605 corresponding to the lateral interval is controlled to expand, and the sliding seat 602 is pushed to move left or right by the expansion piece 605, and thus the spacing between the movable rods 603 corresponding to the lateral interval is adjusted. Specifically, the movable rod 603 is a hollow structure, a through hole corresponding to the expansion piece 605 is located on the movable rod 603, and a solenoid valve is located in each through hole. When the spacing between the movable rods 603 is adjusted, the solenoid valve corresponding to the movable rod 603 requiring adjustment is controlled to open, and the remaining solenoid valves are controlled to close, so as to avoid an incorrect movement of the movable lever rod 603 not requiring adjustment. Specifically, the expansion piece 605 includes a corrugated pipe fixedly connected to the movable rod 603, an elastic member is arranged inside the corrugated pipe, one end of the elastic member is fixedly connected to the sliding seat 602, and the other end of the elastic member is fixedly connected to the corrugated pipe. After all raw materials are finished unloading, all the solenoid valves are opened, and at this time, the gas in the corrugated pipe is discharged under a restoring force of the elastic member, so that the corrugated pipe can return to its original state. Specifically, the feeding component 3 further includes a feeding cover 302 arranged inside the mounting frame 301. A movable board 303 is elastically and slidingly connected to the right inner wall of the feeding cover 302, and a combing board 304 is located on the top of the L0 movable board 303. The bottom of the movable board 303 is extended out of the feeding cover 302 and is in contact with the outer peripheral side wall of the limiting frame 203. Since the end of the movable board 303 away from the combing plate 304 is always in contact with the outer peripheral side wall of the limiting frame 203, when the limiting frame 203 is driven to rotate by the fixed cylinder 201, the movable board 303 is pushed by the limiting frame 203, the comb plate 304 is driven by the movable board 303 to stir the seed cotton raw materials in the feeding cover 302, and to make the distribution of the seed cotton raw materials in the feeding cover 302 in a balanced state. In the initial state, the solenoid valves in the through holes on the movable rod 603 are all closed, the expansion piece 605 is in a contracted state, and the spacing between the movable rods 603 is all equal.
When using, the dual-axis motor 211 is first started, the rocker arm 216 is driven to rotate by the dual-axis motor 211 through the transmission plate 214, the sliding rod 217 is driven to squeeze the inner wall of the chute 210 by the rocker arm 216 to cause the rotating disk 209 to rotate, the limiting frame 203 and the light-transmitting plate 206 are driven to rotate by the rotating disk 209 through the fixed cylinder 201; and at this time, the seed cotton raw materials are put into the feeding cover 302, the seed cotton raw materials slide along the sloping surface inside the feeding cover 302 to the bottom of the feeding cover 302. Since the end of the movable board 303 away from the combing plate 304 is always in contact with the outer peripheral side wall of the limiting frame 203, when the limiting frame 203 is driven to rotate by the fixed cylinder 201, the movable board 303 is pushed back and forth by the limiting frame 203, and the comb plate 304 is driven by the movable board 303 to stir the seed cotton raw materials in the feeding cover 302, to make the distribution of the seed cotton raw materials in the feeding cover 302 in a balanced state. While the limiting frame 203 is driven to rotate by the fixed cylinder 201, the seed cotton raw materials flow out from the gap between the movable rods 603 and adheres to the surface of the light-transmitting plate 206 (the bottom of the feeding cover 302 is on the same plane as the center of the fixed cylinder 201), the seed cotton raw materials are fixed by the fixed blocks 207 on the light-transmitting plate 206, to avoid the seed cotton raw materials from falling off the light-transmitting plate 206 during rotation with the fixed cylinder 201. L0 As the fixed cylinder 201 rotates, the light-transmitting plate 206 attached with the seed cotton raw materials gradually comes into contact with the detection roller 7. Since both ends of the detection roller 7 are elastically and slidingly connected to the inner top of the mounting frame 301, and the diameters of the two ends of the detection roller 7 are smaller than the diameter of the middle part of the detection roller 7, after the light-transmitting plate 206 attached with the seed cotton raw materials is in contact with the middle part of the detection roller 7, a value can be detected by the pressure-sensitive element on the outer peripheral side wall of the detection roller 7, and at this time, a control system can determine that the light-transmitting plate 206 has been loaded, and then the detection component 5 can prepare for work. At the same time, when the light-transmitting plate 206 attached with the seed cotton raw materials is in contact with the middle part of the detection roller 7, the seed cotton raw materials attached to the light-transmitting plate 206 are squeezed by the detection roller 7, to make the seed cotton raw materials compacted, tight cotton has fewer voids, and light rays travel in a relatively straight path when passing through the cotton, and during a process of passing through cotton, the light rays will not suffer too much scattering and be absorbed, so that the light rays can reach a longer distance, and therefore, when the detection component 5 is used to take photos and detect the seed cotton raw materials, the images that can be recognized are sufficiently clear. At the same time, since the seed cotton inside an horizontal area on the light-transmitting plate 206 is compacted, and a height of the seed cotton inside the horizontal area on the light-transmitting plate 206 is certain, therefore, when the light transmittance is detected, the factor that affects the light transmittance is only a content of the seed cotton inside the transverse area, thereby reducing misjudgments caused by the detection component 7 when detecting light transmittance. At the same time, since the plurality of pressure-sensitive elements distributed in an array are located on the outer peripheral side wall of the detection roller 7, whether there are cotton seeds in the seed cotton raw materials within an interval formed by four fixed blocks 207 can be detected, and the positions of the cotton seeds in the interval can be determined according to a torque of the fixed cylinder 201.
When the light-transmitting plate 206 attached with the seed cotton raw materials moves to
the right below the light-transmitting plate 206, the light-transmitting plate 206 attached with
seed cotton raw materials is photographed by the detection component 5, and the plurality of
L0 fixed block 207 on the light-transmitting plate 206 is used as the identification point by the
detection component 5, the light-transmitting plate 206 is divided into a plurality of intervals
distributed in a rectangular array, and the seed cotton inside the intervals is detected and
identified by the detection component 5. Since before the detection is performed by the detection
component 5, the cotton seeds in the seed cotton within the intervals have been determined,
when the seed cotton inside the intervals is detected by the detection component 5, the position
of the cotton seeds can be directly avoided. At the same time, after the shooting is completed by
the detection component 5, the ideas of adaptive threshold segmentation and contour extraction
are adopted, relevant pre-processing operations on the captured images are performed, an image
segmentation technology is used to obtain target impurity images, and experiments on various
contour extraction methods are combined with, to find a suitable extraction method for the outer contours of cotton defects and impurities; and then the extracted feature values are analyzed and calculated, to complete the extraction and location of defects, and the characterization of the type, quantity and impurity content of raw cotton. When the cotton defects in the captured image are detected by the detection component 5, the transmittances of multiple groups of lateral intervals on the light-transmitting plate 206 are simultaneously compared, and whether the difference among the transmittances in each group of the multiple groups of lateral intervals exceeds the set value is detected. Since the movable rods 603 corresponds to the lateral intervals on the light-transmitting plate 206, after the transmittance in a certain lateral interval exceeds the set value is detected by the detection component 5 detects (that is, it indicates that there is too much seed cotton in the lateral interval, after being compacted by the detection roller 7, too much seed cotton affects the light penetration effect), the expansion member 605 corresponding to the the lateral interval is controlled to expand by the control system, and the sliding seat 602 is pushed to move left or right by the expansion piece 605, thereby adjusting a spacing between the movable rods 203 corresponding to the lateral interval, reducing the seed cotton feed amount in the lateral interval, and making the seed cotton content in the lateral interval and other the lateral intervals at a relatively stable value, that is, the light transmittance in each group of lateral intervals on the light-transmitting plate 206 is maintained within a stable numerical interval.
The present disclosure further provides a detection method of machine-picked seed cotton
defects based on deep learning, and the detection method is applicable to the detection device of
L0 machine-picked seed cotton defects based on deep learning. The detection method includes the
following steps:
step 1, putting in raw materials, putting the raw materials of machine-picked seed cotton
into the feeding component, and transporting the raw materials in the feeding component to the
bottom of the multispectral camera through the material delivery component;
step 2, defect detecting, transporting the machine-picked seed cotton to the bottom of the
multispectral camera by the material delivery component, and remaining a position fixed, at this
time, collecting images of the machine-picked seed cotton above the material delivery
component by the multispectral camera, and then analyzing the collected images to realize defect
detection;
step 3, blanking, the detected materials continue following the movement of the material delivery component 2, until the raw materials move to the top of the blanking component, and then blanking the raw materials under the action of the blanking component; step 4, adaptive debugging, in the step 2, the multispectral camera determining the situation of the light rays above the material delivery component passing through the machine-picked seed cotton according to a light transmittance degree of machine-picked seed cotton in the collected images, and controlling the feeding component according to the passage of light rays, to control a release process of the raw materials of the feeding component, and thereby ensuring that both an efficiency of machine-picked seed cotton transferred by the material delivery component 2 and an accuracy of defect detection reach an an optimal value.
The reference to any prior art in this specification is not, and should not be taken as, an
acknowledgement or any form of suggestion that such prior art forms part of the common
general knowledge.
It will be understood that the terms "comprise" and "include" and any of their derivatives
(e.g. comprises, comprising, includes, including) as used in this specification, and the claims that
follow, is to be taken to be inclusive of features to which the term refers, and is not meant to
exclude the presence of any additional features unless otherwise stated or implied.
In some cases, a single embodiment may, for succinctness and/or to assist in understanding
the scope of the disclosure, combine multiple features. It is to be understood that in such a case,
these multiple features may be provided separately (in separate embodiments), or in any other
L0 suitable combination. Alternatively, where separate features are described in separate
embodiments, these separate features may be combined into a single embodiment unless
otherwise stated or implied. This also applies to the claims which can be recombined in any
combination. That is a claim may be amended to include a feature defined in any other claim.
Further a phrase referring to "at least one of' a list of items refers to any combination of those
items, including single members. As an example, "at least one of: a, b, or c" is intended to
cover: a, b, c, a-b, a-c, b-c, and a-b-c.
It will be appreciated by those skilled in the art that the disclosure is not restricted in its use
to the particular application or applications described. Neither is the present disclosure restricted
in its preferred embodiment with regard to the particular elements and/or features described or
depicted herein. It will be appreciated that the disclosure is not limited to the embodiment or embodiments disclosed, but is capable of numerous rearrangements, modifications and substitutions without departing from the scope as set forth and defined by the following claims.
Claims (5)
1. A detection device of machine-picked seed cotton defects based on deep learning,
comprising a frame, wherein a feeding component, a material delivery component and a blanking
component are located inside the frame in sequence from right to left; and a detection component
is fixedly connected to the frame and located above the material delivery component,
the material delivery component comprises a fixed cylinder and a lamp tube with
serpentine cross-section, wherein a plurality of fixing brackets are arranged and distributed in an
array around an outer side of the fixed cylinder; a limit frame is located at both ends of the fixed
cylinder; a light guide block is located between two adjacent fixing brackets; and a plurality of
fixed blocks are distributed in a rectangular array and located above the light guide block, and
the lamp tube is fixedly connected under the light guide block;
the feeding component comprises a mounting frame located on a right side of the fixed
cylinder; and an adjustment component is fixedly located on an inner top of the mounting frame;
the detection component comprises a multispectral camera, the multispectral camera is
fixedly located at a top of an inside of the frame, the feeding component is used to transport raw
materials of machine-picked seed cotton to the bottom of the multispectral camera through the
material delivery component, the multispectral camera is used to collect images of the
machine-picked seed cotton above the material delivery component, and then analyze collected
images to realize defect detection in the machine-picked seed cotton;
the adjustment component comprises a fixed rod fixed to the inner top of the mounting
frame, wherein a plurality of sliding seats corresponding to the plurality of fixed blocks and
distributed in an array are slidingly connected to the fixed rod; a movable rod parallel to a side of
the fixed cylinder is located at a bottom of each of the plurality of sliding seats; an elastic cloth is
fixed between the movable rod on both sides and the mounting frame; an expansion piece is
located at both ends of each of the plurality of sliding seats; and an air pump isfixedly connected
to the top of the mounting frame and located above the plurality of sliding seat, and the air pump
is connected to the expansion piece through a conduit;
a detection roller is elastically and slidingly connected to the inner top of the mounting
frame, the detection roller is not in contact with the fixed rod, and both ends of the detection roller are in contact with a side of the limiting frame; and a plurality of pressure-sensitive elements distributed in an array are arranged on an outer peripheral side wall of the detection roller; the movable rod is a hollow structure, a through hole corresponding to the expansion piece is located on the movable rod, and a solenoid valve is located in each through hole; the expansion piece comprises a corrugated pipe fixedly connected to the movable rod, wherein an elastic member is arranged inside the corrugated pipe, one end of the elastic member is fixedly connected to the sliding seat, and the other end of the elastic member is fixedly connected to the corrugated pipe; a rotating rod is disposed throughout a middle part of the fixed cylinder; a rotating disk is fixedly located at both ends of the rotating rod; and a plurality of chutes are arranged around an outer side of the rotating disk and distributed in an array; a dual-axis motor is located below the fixed cylinder; a transmission rod is fixedly located at both ends of the dual-axis motor; a connecting seat fixedly connected to the frame is movably located at one end of the transmission rod by a first bearing; a transmission disk is movably located at the top of the connecting seat by a second bearing; a belt pulley is fixedly located on each of an outer side of the transmission disk and one end of the transmission rod; a transmission belt is located between two belt pulleys; a rocker arm is fixedly located on the outer side of the transmission disk; and a sliding rod adapted to each of the plurality of chutes are fixedly located on the top of the rocker arm; the blanking component comprises a blanking conveyor belt located on a left side of the fixed cylinder, wherein a mounting rod is located at a right end of the blanking conveyor belt; a mounting seat movably is located at both ends of the mounting rod through a third bearing; a torsion spring is located between the mounting seat and the mounting rod; a plurality of blanking blocks are fixedly located on an outer side of the mounting rod; an extension seat is fixedly located on an outer side of the blanking blocks located on both sides; and a pulley movably is located on an outer side of the extension seat through a fourth bearing, wherein the pulley is adhered to an outer wall of the limiting frame; a control system is located on the outside of the frame, and the control system is configured to control electric components and pneumatic components inside the frame; and the multispectral camera is also used to determine a situation of light rays passing through the machine-picked seed cotton above the material delivery component based on a light transmittance degree of machine-picked seed cotton in the collected images, and control the feeding component according to a passage of light rays, to control a release process of the raw materials of the feeding component.
2. The detection device of machine-picked seed cotton defects based on deep learning
according to claim 1, wherein a light-transmitting plate is fixedly located above the light guide
block, and an outer side of the light-transmitting plate is fixedly connected to the plurality of
fixed blocks, and both sides of each of the plurality of fixed blocks are concave arc-shaped
structures.
3. The detection device of machine-picked seed cotton defects based on deep learning
according to claim 1 or claim 2, wherein the feeding component further comprises a feeding
cover fixedly located inside of the mounting frame, wherein a movable board is elastically and
slidingly connected to a right inner wall of the feeding cover; and a combing board is located on
a top of the movable board, and a bottom of the movable board is extended out of the feeding
cover and is in contact with the outer peripheral side wall of the limiting frame.
4. The detection device of machine-picked seed cotton defects based on deep learning
according to any one of the preceding claims, wherein one end of each of the plurality of
blanking blocks close to the fixed cylinder is an inclined wedge structure.
5. A detection method of machine-picked seed cotton defects based on deep learning,
applicable to the detection device of machine-picked seed cotton defects based on deep learning
according to any one of claims 1 to 4, and the detection method comprising:
step 1, putting in raw materials, putting the raw materials of machine-picked seed
cotton into the feeding component, and transporting the raw materials in the feeding component to the bottom of a multispectral camera of the detection component through the material delivery component; step 2, defect detecting, transporting the machine-picked seed cotton to the bottom of the multispectral camera by the material delivery component, and remaining a position fixed, at this time, collecting images of the machine-picked seed cotton above the material delivery component by the multispectral camera, and then analyzing collected images to realize defect detection; step 3, blanking, the detected materials continue following the movement of the material delivery component, until the raw materials move to the top of the blanking component, and then blanking the raw materials under the action of the blanking component; step 4, adaptive debugging, in the step 2, determining a situation of light rays passing through the machine-picked seed cotton above the material delivery component by the multispectral camera based on a light transmittance degree of machine-picked seed cotton in the collected images, and controlling the feeding component according to a passage of light rays, to control a release process of the raw materials of the feeding component, and to make both an efficiency of machine-picked seed cotton transferred by the material delivery component and an accuracy of defect detection reach an optimal value.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310656228.5A CN116698870B (en) | 2023-06-05 | 2023-06-05 | A machine-picked seed cotton defect detection device and detection method based on deep learning |
| CN202310656228.5 | 2023-06-05 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| AU2024202145B1 true AU2024202145B1 (en) | 2024-10-31 |
Family
ID=87825084
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| AU2024202145A Active AU2024202145B1 (en) | 2023-06-05 | 2024-04-03 | Detection device and detection method of machine-picked seed cotton defects based on deep learning |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN116698870B (en) |
| AU (1) | AU2024202145B1 (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104751443B (en) * | 2014-12-12 | 2017-10-10 | 郑州轻工业学院 | Based on multispectral technology cotton defect detection and recognition methods |
| US20200281122A1 (en) * | 2017-09-05 | 2020-09-10 | Cottlab Ltd. | Self-propelled robotic harvester for selective picking of high quality agriculture row crops |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103698341B (en) * | 2013-12-31 | 2016-01-20 | 中华全国供销合作总社郑州棉麻工程技术设计研究所 | A kind of quality of roll for cotton detection system and quality of roll for cotton image detecting method |
| CN204422432U (en) * | 2015-01-13 | 2015-06-24 | 河南财政税务高等专科学校 | Cotton defect detection system |
| CN106918594B (en) * | 2015-12-25 | 2023-06-20 | 济南大学 | Method for on-line analysis of seed cotton color characteristics and impurities |
| CN109738436A (en) * | 2018-12-28 | 2019-05-10 | 安徽财经大学 | A kind of seed cotton impurity quantitative rapid detection system and method |
| CN114056975A (en) * | 2021-12-07 | 2022-02-18 | 滁州卷烟材料厂 | Automatic aluminum foil paper conversion device and conversion method thereof |
-
2023
- 2023-06-05 CN CN202310656228.5A patent/CN116698870B/en active Active
-
2024
- 2024-04-03 AU AU2024202145A patent/AU2024202145B1/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104751443B (en) * | 2014-12-12 | 2017-10-10 | 郑州轻工业学院 | Based on multispectral technology cotton defect detection and recognition methods |
| US20200281122A1 (en) * | 2017-09-05 | 2020-09-10 | Cottlab Ltd. | Self-propelled robotic harvester for selective picking of high quality agriculture row crops |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116698870B (en) | 2024-01-23 |
| CN116698870A (en) | 2023-09-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107952696B (en) | A kind of detection grading plant and detection method suitable for fresh tobacco leaf | |
| CN111069094A (en) | Multi-index comprehensive spherical fruit and vegetable quality grading and sorting device | |
| EP0545129B1 (en) | Method to detect manmade fibres and/or defective fibres and/or other foreign materials in the processing of silk waste, and relative apparatus | |
| JPH06316862A (en) | Acquisition, measurement and control of fiber substance in process | |
| CN102253052B (en) | Grain quality on-line detection apparatus based on field programmable gate array (FPGA) | |
| CN108931531A (en) | Method, system and computer readable storage medium for automatically detecting fabric defects | |
| CN104502356A (en) | Automatic detection method for defects of inner surface of sliding bearing on basis of computer vision | |
| CN205732018U (en) | A kind of unginned cotton dirt foreign material sorting system based on machine vision | |
| CN211839115U (en) | A multi-index comprehensive quality grading and sorting device for spherical fruits and vegetables | |
| US5752294A (en) | System and method for detection of cotton stickiness and neps and other lint qualities in real time and removal of sticky deposits from processed cotton in the gin | |
| US20220042212A1 (en) | Fiber preparation machine | |
| CN111672773B (en) | Product surface defect detection system and detection method based on machine vision | |
| WO2018213535A2 (en) | Imaging system for sorting poultry meat with wooden breast myopathy | |
| CN1097102C (en) | Method and appts. for recognition and separation of impurities in ready equipment of spinning (opener) | |
| AU2024202145B1 (en) | Detection device and detection method of machine-picked seed cotton defects based on deep learning | |
| CA1162161A (en) | Conditioning line for steel pipes | |
| CN110716033B (en) | Method and device for rapid evaluation of impurities in seed cotton during the purchasing process | |
| CN203120696U (en) | Online non-destructive testing sorting device for eggs with blood spots | |
| CN112767384A (en) | Compensation method and device of elastic cloth gram weight online detection system | |
| CN201255721Y (en) | Detecting and metering apparatus for isomerism fiber in cotton | |
| CN115018781B (en) | Method and device for detecting yarn evenness of high-speed spinning | |
| Li et al. | A direct measurement method of yarn evenness based on machine vision | |
| CN110201905A (en) | A kind of end surface of bearing ring on-line detection device of defects | |
| Delwiche et al. | A high-speed sorting system for dried prunes | |
| CN115537977B (en) | Foreign Fiber Removal System |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FGA | Letters patent sealed or granted (standard patent) |