MXPA97009032A - Bor inspection - Google Patents

Abstract

The present invention relates to an inspection system for the manufacture of razor blades, characterized in that it comprises: an inspector that detects the defects in the edge of a moving strip of a material for razor blade that moves along a When the inspector is located in a first position upstream along the path, the inspector generates a defect signal when the defect is detected, the inspector includes at least one beam directed perpendicular to the direction of movement of the strip of the material for razor blade, a camera for taking the image of an edge of the moving strip of razor blade material, the camera is independent of the inspector, the camera is located downstream of the inspector in a second position downstream along the of the path along the moving strip, the camera is controlled only to take pictures of particular portions that are less than all the In this section, the camera responds to the defect signal to take a photograph of the strip at a predetermined time subsequent to the defect signal, so that the photograph taken of the defect that was detected in the inspector at that time defect passes the camera to a second downstream position, a monitor, coupled with the camera to display the images of the defects captured by the camera, and a storage system coupled with the camera and the monitor, to store the images of the defects captured by the camera, the monitor is able to display the stored images of the defective

Description

INSPECTION OF EDGES DE? 3CRIPCTQW OF IA m EMPTÓW This invention relates to inspection of edges or edges. The sharp edges of razor blades, for example, are typically inspected after the blades have been cut from a strip of steel that has passed through a sharpening machine. An operator transfers the knives to shave shafts to form a block of knives with sharp edges of the knives all oriented in the same direction. The defects are detected by holding the block of knives with sharp edges at different angles to a light source and by looking for diffuse reflections of light indicating damaged knives. To remove the defective knife from the block of knives, the operator transfers a section of good knives from the block to another set of knives and removes and discards the various knives of the block in the vicinity of the reflection. The operator then transfers the good knives back to the original axes and again checks for defects. In general, in one aspect, the invention describes a method that includes processing a continuous edge of a strip REP: 26238 of material which moves in a direction along the length of the strip. The condition of the continuous edge of the strip and movement is instructed after it has been processed. The strip is cut into pieces. The pieces are classified into groups, in response to the condition of the edge. Implementations of the invention include one or more of the following characteristics. The method can be used with a manufacturing line that produces razors for shaving, where the strip of material is a strip of material for a razor and the pieces are razors for shaving. The razors for shaving can be classified into a first group of razors for shaving in good condition and a second group of razors for shaving defective, based on the detection of defects in the edge or edge. The inspector may include a first laser system having a first projector for projecting a first laser beam at the cutting edge in a direction perpendicular to the direction of movement of the strip and perpendicular to the cutting edge, and a first profile detector for detecting a portion of the first laser beam passing over the cutting edge, and to generate a first signal representing the detected portion of the first laser beam. Another detector can receive light reflected from the edge to detect edge damage. There may also be a second laser system, in close proximity to the first laser system, which includes a second projector, for projecting a second laser beam at the cutting edge in a direction perpendicular to the direction of movement of the strip and perpendicular to the edge of cutting, and a second profile detector for detecting a portion of the second laser beam passing over the cutting edge and for generating a second signal representing the detected portion of the second laser beam. A normalizing circuit may receive the first and second signals from the first and second profile detectors. Artifacts or aberrations associated with the movement of the cutting edge can be eliminated by filtration. An edge discontinuity signal can be generated and processed to detect defects in the cutting edge. A defect signal can be generated in response to the detected defects. The defect detection circuit can detect defects by detecting corresponding peaks of opposite polarity within a predetermined amount of time in the edge discontinuity signal. The predetermined amount of time may depend on the speed with which the strip moves and the distance between the first and second laser beams. In general, in another aspect, the invention describes a manufacturing line that includes the machine that processes a continuous edge of a strip of material; an inspector that determines the condition of the edge; a cutter that cuts the strips into pieces; and a classifier that classifies the pieces into at least two groups in response to the edge condition. In general, in another aspect, the invention describes an apparatus for continuously monitoring an edge of a strip of material, the apparatus has a pair of closely spaced laser beams, for example, as described above. In general, in another aspect the invention describes a display system that includes a camera, a monitor and a storage system. The camera produces images of an edge of a moving strip of material, the monitor, coupled to the camera, displays the images captured by the camera. The storage system, coupled to the camera and monitor, stores images captured by the camera. The monitor is capable of displaying the stored images. The implementations of the invention may include one or more of the following characteristics. The light source can be directed to the edge of the strip. A first camera and lens, in close proximity to the light source on a first side of the strip, can take graphic representations of the edge of the strip. A second camera and lens can be in close proximity to the light source on a second side of the strip, and can take graphic representations of the edge of the strip. Strobe lighting can be used to stop the movement of images. A computer that receives the graphic representations of the first and second cameras can generate digitized images of the graphic representations, and can display the digitized images on a monitor. Graphic representations can be acquired on the computer at the discretion of the operator or, when the computer receives a defect signal, the cameras can take graphic representations of the strip at a predetermined time so that graphic representations of the detected defects are taken. The predetermined time can be adjustable and the computer can include an input device through which an operator can enter synchronization settings. A storage system can store the scanned images in response to instructions from the computer, and an operator using the computer's input device can have the computer retrieve the stored scanned images from the storage system for display on the monitor. The invention may include one or more of the following advantages. Defects in the edge or edge can be detected quickly and accurately and defective parts can be rejected. The operator can observe images of the defects and the edge and observe statistical information about the defects. Other advantages and features will become apparent from the following description and from the claims. Figure 1 is a block diagram of a line for making razors for shaving; Figure 2A shows a block diagram of an inspection system; Figure 2B shows a camera system; Figure 3 shows a perspective view of an inspection system; Figure 4 shows a cross-sectional view of a magnetic guide or positioner; Figure 5 shows a laser detector; Figure 6 shows a perspective view of an edge detector including two laser detectors; Figure 7 shows a graphic representation of the edge profile signals generated by the laser detectors; Figures 8A, 8B and 8C show graphic representations of edge profile signals generated by a defect; Figure 9 shows a block diagram of the controls for detecting and rejecting elements of an inspection system; Figure 10 shows a flow chart showing the operation of an analog circuit (PCB), a single SBC, 1, 2, or 6 computer board and a single-board image synchronization computer; Figure 11 shows a flow chart where the detection of a real defect on a single board computer is observed; Figure 12 shows a flow chart showing the operation of an analog circuit to detect edge damage in a single board computer; Figure 13 shows a flow diagram showing the operation of a single-board computer that analyzes the damage of an edge; Figure 14 shows a flow diagram showing the operation of a single-board computer to synchronize the image; Figure 15 shows a rejection image screen; Figure 16 shows a razor-width trend analysis diagram; Figure 17 shows an image control screen.

With reference to Figure 1, a manufacturing line 10 produces razors for shaving by passing a strip 14 of continuous steel from a supply wire 11 through a sharpener 12 which grinds and polishes the strip 14. Before being cut into knives for shaving individual with a cutter 22, the strip is examined by an inspector 24. The inspector 24 detects defects in the edge, on the sharp edge of the strip. Based on the sensitivity of the inspector, different types of edge defects can be detected. Among the defects detected are those which interrupt (cause gaps in) the continuous sharp edge of the steel strip passing through the inspector. The inspector 24 sends defect information to a programmable logic controller (PLC) 28, a display system 46 and a rejector 26. The PLC 28 dynamically controls the operation of the rejector 26. With the information provided by the inspector 24 and by other equipment 10 of line, the PLC 28 causes the rejector 26 to discard the razors 30 for defective shaving and provides razors 32 for shaving free of defects as a finished product of the line 10. The PLC 28 also keeps the count of the number of razors for shaving good produced and the number of disposable razor blades discarded. The accounts can be used by the PLC to detect when the process limit thresholds are satisfied and stop the machine when there is an excessive amount of defective product. On the other hand, when defective blades are not detected in a large quantity of good product, the detection system may have ceased to function. The PLC will stop the machine for a "fail-safe" detector. With reference to Figure 2A, the inspector 24 includes laser detectors 40 which continuously monitor the edge of the strip 14 and send signals 41 to a high speed analog electronic circuit 42. The analog electronic circuit 42 processes the received signals 41 to detect defects at the edge and send the digitalized defect signals 43 to a real-time digital microprocessor 44. The microprocessors 44 use the digitized signal 43 to determine whether actual defects or a sweep (i.e., noise or movement) 14 of the strip have been detected, and the microprocessors 44 send detected signals of actual defect to the PLC 28 and signals 47 of actual defect detected to the display system 46. The PLC 28 then causes the rejector 26 to discard the defective razors. The display system 46 controls a camera system 48 through which the strip 14 passes downstream of the laser detectors 40. As seen in Figure 2B, two cameras 62, 64 in a camera system 48 take pictures of both sides of the razor strip 14 using a fiber optic strobe illuminator 65. The display system 46 generates digitized images of the graphic representations taken by the camera system 48, records the images with dates and times, and renders them available for display at the operator interface 56 or for storage in a storage system 58. The display system 46 and the storage system 58 can be connected to an entire factory-wide network and one or more operator interfaces 56 provide operators through factory access to images and information about the strip 14 If the microprocessors 44 indicate that an actual defect has been detected in the strip 14, the display system 46 determines, based on the current strip speed, the arrival time downstream of the defect to a particular camera within a system. 48 camera and directs to that camera to take a picture of the defect. A graphic representation of the defect taken on the razor blade before the razor to be rejected can be more reliable than an image of the razor discarded, because the discarded razor may be more damaged in the process of being discarded.

Because the camera system 48 only operates at close video speeds, the frequency at which images are captured is limited. You can capture only one image every fifty milliseconds. Therefore, multiple defects detected within short distances between them can not be represented graphically. As explained below, the display system implements a defect priority system to capture images of the largest type of defect detected. In addition, because the field of view of each image shows only 1.7 mm (0.070 inches) along the edge of the razor (just wider than a typical 100X microscope), the full extent of any damaged section can not be visible . The display system 46 can direct the camera system 48 to take pictures at predetermined intervals even when no defects are detected. The information may be available for display at the operator interface 56 or for storage in the storage system 58. The display system 46 also controls a commercial laser micrometer 50 (Figure 3) which measures the width of the total razor of the strip 14 and can be directed to perform measurements periodically. The visualization system 46 analyzes these measurements and generates process trend diagrams. The system 46 then produces the process trend charts and other information available for display at the operator interface 56 and for storage in the storage system 58 and on the factory network. With reference to Figure 3, the inspector 24 includes a detector housing 60 within which the laser detectors 40 are mounted. The strip 14 passes through the detector housing 60 and, in this way, passes through the laser detectors 40 before passing through the camera system 48. The camera system 48 includes a camera and lens 62, a camera and lens 64 and a light source 65. The light source 65 can be a fiber optic illuminator coupled to a strobe light. The strip 14 then passes through a laser micrometer 50. Although the steel strip 14 passes through the inspector 24, it is mounted on a magnetic guider 69 (FIG. 4) referencing against the bottom edge and one side of the strip. Three lower flat portions 54a, 54b, 54c are scattered on the path through the inspector 24 (approximately 35 centimeters (14 inches)). The flat part 54a is at the beginning of the inspector 24, the flat part 54b is close to the chambers, and the flat part 54c is at the end of the inspector 24. Between the flat parts, the magnetic guider is released to allow the strip to be move. The inspector is mounted in the middle part between the flat parts to ensure a regular vertical movement of the strip. As seen in Figure 5, a laser detector 40a includes a single commercial collimated diode laser projector 70 and a cylindrical lens 71 for focusing the laser beam on a line detected at the upper edge 21 of the strip 14, which is shows traveling in Figure 5. The edge profile detector 72 receives the light that passes over the edge 21, and the edge damage detector 76 receives the light reflected from the edge 21 and collected by the lens 74. The detector Edge damage 76 is located on the opposite side of the vertical line above the edge 21 to prevent entry of the scattered laser beam from the illuminated side of the strip 14. A second laser detector 40b, shown in Figure 6, is similar to the detector 40a and includes a laser projector 70 ', a cylindrical lens 71', an edge profile detector 72 ', a lens 74' and an edge damage detector 76 '. The elements of the laser detector 40b, however, can be placed opposite the elements of the laser detector 40a. In this way, the reflected edge damage light from any direction of the edge 21 can be detected. The edge profile sensors 72 and 72 'together are used to detect defects. Each of the detectors 76, 76 'of damage and their respective lenses 74, 74', are also used to independently detect defects. The two detectors 40a, 40b form a parallel array of laser detectors separated by a known small distance D of 5 mm (0.2 inches) in this example. The distance D is small enough to allow the two detectors to experience the same blade stroke perpendicular to the direction of machine movement and is large enough to be greater than the length of many defects that interrupt the edge . Each edge profile detector 72, 72 'generates a continuous analog profile signal. The profile signals of the detectors are then coupled to AC, can be filtered and subtracted to provide a normalized edge profile signal. The normalized edge profile signal is digitally processed to separate the actual defects from the process conditions, which include knife shifting (eg, noise or strip movement). The edge 21, the magnetic guider 69 and the laser projectors 70, 70 'and the detectors 72, 72' are aligned to utilize the central portion of the collimated laser beam, in which the profile of the Gaussian beam is relatively flat. This provides a reasonably linear change in light with edge shift, as shown in Figure 7. Because laser diode projectors emit elliptical collimated beams, the linear region in the long axis direction of the ellipse is reasonably large compared to the size of the sharp edge of the razor strip. The usable linear range of approximately 0.76 mm (0.014 inches) (0.144 to 0.174) shown in Figure 7 is sufficient to adapt to the movement of the edge due to normal variation of the product and stability of the device in the magnetic guider The signal subtracted from the detectors 72, 72 'of front and rear edge profile are normalized to eliminate vibration in strip 14 because detectors 40a, 40b are close (approximately 5 mm (0.2 inches) and observe the same movement of strip. Similarly, the typical product variation on the edge 21 occurs slowly (with a larger special wavelength) in relation to the detector separator and is also eliminated by subtraction of the combined signal. However, the edge discontinuities that both sensors pass sequentially and appear in the subtracted signal. Figures 8A and 8B respectively show a sample of the trace of a signal of an edge profile signal with an edge discontinuity 80 passing through the front edge profile detector 72 and the same edge discontinuity 80 passing through of detector 72 'of rear edge profile. Figure 8C shows in an enhanced manner the characteristic appearance of a form of normalized discontinuity. Two characteristics 81, 82 are generated in the normalized signal, one positive 81 and the other negative 82. These peaks are detected with window thresholds + W and - in the signal. The size of W can be set appropriately for different types of defects. Excessive strip vibration or edge variations may exceed inspector signal thresholds but not show characteristic peaks 81, 82 inverted. Since both the speed of the strip and the separation of the detector are known, any detected peak may have a corresponding peak of opposite polarity within a certain time window to be an edge defect. An additional defect discrimination of the edge vibration and variations are obtained with domain filtering in the time of the signal before normalization. This reduces any strong random signal component of the filter band passages that does not appear simultaneously in both detectors, and also avoids the high frequency artifacts that are generated if the signals are subtracted in some other way. For our application, window thresholds of 0.15 mm (0.006 inches) are used in the normalized signal without filtration, thresholds of 0.02 mm (0.0008 inches) in the signal for a frequency response higher than 400 Hz, and 0.0076 mm (0.0003 inches) ) in the signal with a frequency response higher than 1 KHz. As shown in Figure 9, the analog electronic circuit 42 includes four channels, each to detect a particular kind of defect. The four channels continuously receive signals from the laser detectors. Some defects can be detected using the profile detectors 72, 72 'at the leading and trailing edge. Accordingly, the detector circuit 98 and the detector circuit 102 receive the signals 90, 90 'of the front and rear edge profile detectors. Other defects can be detected based on detector 76 of the leading edge damage or on detectors 76 'of damage on the trailing edge. As a result, the detector circuit 104 receives signals 94, 96 from edge damage detectors 76, 76 ', respectively. The real-time digital microprocessors 44 of the inspector 24 (FIG. 2A) include four single-board computers (SBC), SBC1 112, SBC2 116, SBC3 122 and SBC6 117 that receive the defect signals from the analog detector channels and determine if The indicated defects are real defects in determining if the defect criteria are met. The detector channel 98 sends signals 108 and 110 that affect a defect class to the SBC1. The detector channel 102 sends signals 114 and 116 indicating a second defect class, to SBC6. The detector channel 103 sends signals 97 and 99 indicating the third defect class to SBC2. Similarly, the detector channel 104 sends the signal 118, which indicates a fourth defect class and the signal 120, which indicates a fifth defect class, to SBC3. When SBC1 determines that a defect exists, it sends defect signals 124, 125 and / or defect signals 126, 127 to an image synchronizer SBC4 130 and PLC 28, respectively. When SBC2 determines that there is a defect, sends defect signals 128, 129 to the image synchronizer SBC4 130 and PLC 28, respectively. When SBC3 determines that a defect exists, it sends defect signals 131, 133 to the image synchronizer SBC4 130 and PLC 28, respectively. When SBC6 determines that a defect exists, it sends defect signals 132, 134 to the image synchronizer SBC4 130 and PLC 28, respectively. The display system 46 includes image synchronization SBC4130. It determines the time at which the defective portions of the strip 14 reach the camera system 48 and cause the camera system 48 to take graphic representations accordingly. The PLC 28 causes the rejector 26 to discard the defective razors. A commercial through-beam photodetector 202 is mounted on the rejector, detects that the knives are actually ejected. This fail-safe signal is monitored by SBC5204, which also receives the original rejection signals. SBC5 204 determines that all defects were actually rejected, and sends signals to the PLC to stop the machine if they are not rejected. The flow chart of Figure 10 shows the operation of each of the large, medium and small defect detection circuits. The leading edge profile signal 90 is generated by the leading edge profile detector, and passes to the gain amplifier 144. The signals are then filtered 145 with time domain for the medium and small defect circuits, the large defect circuits are not filtered, the medium defect circuit allows signals to pass higher than 400 Hz, and the small defect circuit allows pass signals higher than 1 KHz. Then the signals are coupled 150 AC to eliminate any deviation of DC level. The trailing edge profile detector signal 72 'follows identical paths for large, medium and small defect circuits. The leading and trailing edge profile signals are then subtracted 151 to provide the normalized signal 153 for each of the large, medium and small defect circuits. The normalized signal is then compared 159, 161 with the upper and lower window detection thresholds 155, 157, for each of the large, medium and small defect circuits. When the normalized signal exceeds the upper threshold positively, the output 170 to the SBC 163 is energized or activated by the duration of the condition. When the normalized signal exceeds the lower threshold in a negative manner, another output 172 is energized SBC for the duration of the condition. The upper and lower detection thresholds are set at ± 0.15 mm (0.006 inches) (voltage equivalent) for the large defect circuit, ± 0.02 mm (0.0008 inches) for the medium defect circuit, and +0.0076 mm (0.0003 inches) for the small defect circuit. The SBC1 receives the resulting signal from the large fault circuit, the SCB6 receives the signal resulting from the medium fault circuit, and the SBC2 receives the signal from the small fault circuit. As described above, single-board computers determine whether defect signals have real defects in determining if certain defect criteria are met. The single-board computers each receive an input of the line speed 165 of the grinding machine from a commercial counter. Since the defect will pass the front and rear detectors with a time difference that depends on the speed of the line, each defect can generate corresponding defect signals through the upper and lower threshold comparators in a time difference proportional to the speed of the line and detector separation (5 mm (0.2 inches) in this example). If the defect is removed from the edge of the razor blade the light reaching the edge profile detectors 72, 72 'will increase and first generate a higher threshold signal, followed by a corresponding lower threshold signal; similarly, if the defect protrudes from the edge of the razor blade, the light reaching the edge profile detectors will decrease, and first a lower threshold signal will be generated followed by a corresponding upper threshold signal. Any threshold signal that rises by itself without a superior opposing threshold signal at the corresponding time is not from a defect, but rather from a random movement of the razor strip or a displacement. With reference to Figure 11, the upper and lower threshold signals 170, 172 generate interrupts in the SBC1, SBC2 and SBC6, which execute similar programs. A higher threshold signal interruption will cause the program to check 174 in any synchronizer activated by the lower threshold interruption 5 mm (0.2 inches) before. If found, a defect has been detected and that synchronizer will be deactivated 176 and will reject signal output 178 to PLC 28 and SBC4130 image synchronizer. The 5 mm (0.2 inch) synchronizer must be valid within a certain tolerance to generate a rejection decision, ± 15% is a reasonable level. If two synchronizers activated do not give an amount equal to 5 mm (0.2 inches) ± 15%, then the program tries to start 180 a new higher threshold signal synchronizer (four are available in the mode program). If all four synchronizers are in use, then the threshold signals must start at a step too fast, and a rejection signal 178 is output. Otherwise, a new upper distance synchronizer is started 182. The program works similarly for lower threshold signal interruptions. SBC1, SBC2 and SBC6 also have internal synchronization switches 185 for verifying the speed of the tuning line from a commercial counter. The speed is verified 186 and updated several times (for example four) every second, and new distance synchronization limits are calculated for the values of 5 mm (0.2 inches) ± 15% at most of the line speed current.

When any distance synchronizer exceeds 5 mm (0.2 inches) + 15%, it generates a program interrupt 188. Then the program checks 190 if the upper or lower signal that activates that synchronizer remains active continuously over the duration of 5 mm (0.2 inches) + 15%. If this is the case, it is generated by a greater defect than the detector of 5 mm (0.2 inches) of separation on the sharpening line, so that the leading edge of the defect passes both detectors before the trailing edge reaches the front detector. Therefore, 192 a rejection signal is generated. Otherwise, the distance synchronizer 194 is deactivated. Figure 12 shows the operation of the edge damage defect detector circuits. The leading edge damage signal 94 is generated by the leading edge damage detector and passes to the gain amplifier 212. The signal is then coupled 214 AC to eliminate any DC deviation. The signal is then compared 215 to a threshold edge 216 of damage and the edge damage output SBC3 218 is energized or activated by the duration of the condition when it exceeds the threshold. The damage signal on the trailing edge follows an identical path. With reference to Figure 13, front and rear edge damage signals 118, 120 that exceed the thresholds generate interrupts for SBC3122. These signals will cause the program to start with 230 synchronizers, 232 of front leading edge damage. A periodic synchronizer switch 240 causes the program to verify 242 each edge damage synchronizer to determine if the initiating signal has remained active for the rejectable period. If so, a rejection signal is output 244. If the edge damage signal continues to be determined 243, then the reject signal is output repeatedly. However, if the edge damage signal expires before 245 has been satisfied in the rejected period, then the edge damage synchronizer is deactivated 246. As in SBC1, 2 and 6 in Figure 11, this program also has a periodic internal synchronizer interruption to verify the speed on line 250 from a commercial counter. The velocity information is obtained 251 and a rejectable period equivalent to the 250 length of rejectable edge damage is used to calculate 252. The SBC3 122 receives the input of the rejectible length 254 of the user selectable switches (the minimum continuous edge damage length which is considered to be rejected). If the SBC1, SBC2, SBC3 or SBC6 determine that the actual defect has been detected, then they signal the PLC 28 to reject the defective knife and synchronize to SBC4 130 to form images. With reference to Figure 10 and Figure 14, SBC4 130 receives the detected fault signals through a trigger synchronizer (268 being between them) and tilting circuits 272 (which are between them) and a signal notification through of an OR gate 274. Since more than one defect detection channel can be energized by any particular knife strip defect, SBC4 130 uses tilted signal displays to choose the largest defect type for the image production display . This ensures that the defect displayed is categorized with the appropriate rejection type. With reference to Figure 14, the SBC4 130 receives the rejection interrupt signal 290, and then interrogates the rejection types 292 and resets the tilt signals. Since the image formation is restricted to video speeds as explained above, the SBC4 determines 294 whether an image synchronization conflict will occur with a graphic representation of the defect previously in the row. If there is no conflict, then a graphic representation synchronizer is activated 295 and the graphic representation type (large, medium, etc.) is added to the row. If there is a conflict, the program compares the priority of the graphic representation type of the new defect with the previous graphic representation 296, and the larger defects have higher priority. If the new graphic representation is of higher priority, then the previous graphic representation synchronizer is deactivated 297 and a new graphic representation synchronizer 298 is started with the new type of graphic representation placed in the row. Otherwise, if the new graphical representation is of lower priority, 300 is ignored. This process is similar to the logic used to choose the largest rejection image type from the skewed signal representations for SBC4 130. the image representation synchronizer reaches the preset distance equivalent to the placement of the defect in the front of the cameras, an interrupt 302 is generated. Then the program deactivates the synchronizer 304, verifies 306 the type of graphic representation in the row, and output 308 to the information for the display system. The display system 46 then acquires the image using the appropriate camera and strobe lighting, stores the image in a digital memory and labels the image with the information of type of graphic representation, date and time. As in the other SBCs, this program also has a periodic internal synchronizing switch 310 to verify the speed of the line from the commercial counter. The velocity information is obtained 311 and used to calculate 312 the period of graphical representation synchronization equivalent to the distance from the rejection of the sensor to the camera. The SBC4 also receives a user selectable input 402 from the display system to advance or delay image synchronization, and in this way move the centering of the defect in the resulting images. Once a defect is detected, the PLC 28 locates the defect in the knife in the strip immediately in the inspector 24. The whole of the knife is later considered defective. PLC 28 follows the razor down the sharpening line and through the cutter using knife-by-knife pulses from a commercial encoder mounted on the sharpening line. The defective knife is then separated by a device similar to that used to stretch knives of a cutter and sharpen or deposit them. A commercial through beam photoelectric device verifies the presence of rejected blades that are segregated by the rejector. The SBC 204 (FIG. 9) receives the rejection signals of the SBC1, SBC2, SCB3 and SBC6, as well as the presence of the razor rejected from the through beam photodetector. The SBC5 follows the rejection through the sharpening line and the cutter and rejector using knife-by-knife pulses from the sharpening line encoder. The SBC5 acts as a fail-safe system for the PLC and the rejector. If the defective blades are not rejected successfully, the SBC5 will signal the PLC to stop the sharpening machine. The display system 46 may be a personal computer system containing a commercial image forming card, a video camera and lens, and a strobe light. The operator interface screen is by means of a VGA video monitor with a commercial capi screen attached to a personal computer system. The SBC4 130 activates the display system 46 to capture an image when the defect detected in the razor blade has traveled down towards the sharpening line and is within the field of view of the camera (a width of 1.8 mm (0.070 inches) ) in this mode). The movement of the strip is frozen or stopped by the impulse of the strobe light to provide a clear defect image, which is displayed on the operator's interface screen. Up to forty of the most recent defect images can be maintained in a RAM memory of a 16 megabit (megabyte) imaging board. An example of an operator interface rejection image screen is shown in Figure 15. This screen is initialized with the "switching type" button 357 which is set to LAST, which causes the most recent defect image of Any type is updated on the display screen. The "switching type" button can be activated to cycle through different categories of defects, such as large, medium, small defects or edge damage. A row of the most recent images of each type of defect is stored in a RAM memory. The row of images for each type of defect can be moved through the use of the previous buttons 358 or following 360 once the switch type button 357 is used to select the defect type. The selected image will remain on the screen until it is removed from the end of the row of the most recent images for that defect. The activation of the current button 361 will display the most recent image of the selected type. Activation of the four image buttons 364 causes the display system 46 to divide the display screen into four quadrants and display a defect image in each quadrant. Activation of the storage button 366 causes the display system 46 to write the displayed image to a permanent storage 58 (Fig. 2A) on a local hard disk or in a network if the personal computer is connected to a network.

A trend screen can also be displayed on the display system as shown in Figure 16. The width information of the plotted knife is measured by the laser micrometer 50 which samples the knife width data at a selectable speed . The data is then plotted on the graph that is displayed and the trend lines are drawn to connect the data points. The target width 392 is shown in the graph and can be a different color than the width 394 of the actual strip, and warnings can be set automatically if the width of the strip approaches predetermined limits. An image control screen may also be displayed on the display system, as shown in Figure 17. This screen allows the image acquisition timing to be adjusted. When acquiring images a little before or a little later in time, defects can be displaced to the left or right in the displayed images. By moving the image acquisition time, you can find evidence of process conditions that are causing defects (ie, scratches, streaks, etc.). The slide control button 402 can be activated either to advance or delay the synchronization of the arriving images with quarter-field of view increments. A maximum setting is almost two or more fields of vision or about 4.4 mm (0.174 inches). It is noted that in relation to this date, the best method known by the applicant to carry out the aforementioned invention, is the conventional one for the manufacture of the objects to which it relates. Having described the invention as above, property is claimed as contained in the following:

Claims (18)

1. A method, characterized in that it comprises: processing a continuous edge or edge of a strip of material which moves in a direction along the length of the strip; inspect the condition of the continuous edge of the moving strip after it has been processed; cut the strip into pieces; and store the pieces in groups, in response to the condition of the edge.

2. The method according to claim 1, characterized in that it is used with a production line that produces razors for shaving, in which the strip of material is a strip of razor material for shaving and in which the pieces are razors for shaving .

3. The method according to claim 2, characterized in that the inspection of the edge detects defects in the edge, and in which the classification classifies the razors for shaving in a first group of razors for shaving in good condition, and a second group of knives for shaving defective.

4. The method according to claim 1, characterized in that it additionally comprises: a first laser system including a first projector for projecting a first laser beam at the cutting edge in a direction perpendicular to the direction of movement of the strip and perpendicular at the cutting edge, and a first profile detector for detecting a portion of the first laser beam passing over the cutting edge, and for generating a first signal representing the detected portion of the first laser beam; and a second laser system, in close proximity to the first laser system, including a second projector for projecting a second laser beam at the cutting edge in a direction perpendicular to the direction of movement of the strip and perpendicular to the cutting edge, and a second profile detector for detecting a portion of the second laser beam passing over the cutting edge, and for generating a second signal representing the detected portion of the second laser beam.

5. A manufacturing line, characterized in that it comprises: a machine that processes a continuous edge or edge of a strip of material; an inspector that determines the condition of the edge; a cutter that cuts the strip into pieces; and a classifier that classifies the pieces into at least two groups, in response to the condition of the edge.

6. The manufacturing line according to claim 5, characterized in that the manufacturing line produces razors for shaving, and in which the strip of material is a strip of razor material for shaving, and in which the individual pieces are knives for to shave.

7. The production line according to claim 6, characterized in that the condition detected by the inspector indicates a defect in the edge and in which the classifier classifies the razors for shaving in a first group of razors for shaving in good condition and a second group of defective razor blades.

8. The manufacturing line according to claim 5, characterized in that the parallel set of laser systems include: a first laser system including a first projector for projecting a first laser beam at the cutting edge in a direction perpendicular to the direction of movement of the strip and perpendicular to the cutting edge, and a first profile detector, for detecting a portion of the first laser beam passing over the cutting edge and for generating a first signal representing the detected portion of the first laser beam; and a second laser system, in close proximity to the first laser system, including a second projector for projecting a second laser beam at the cutting edge in a direction perpendicular to the direction of movement of the strip and perpendicular to the cutting edge, and a second profile detector for detecting a portion of the second laser beam passing over the cutting edge, and for generating a second signal representing the detected portion of the second laser beam.

9. An apparatus for continuously monitoring an edge of a strip of material, characterized in that it comprises: a pair of parallel laser beams, closely spaced apart.

10. The apparatus according to claim 9, characterized in that the parallel laser beams include: a first laser system including a first projector for projecting a first laser beam at the edge, in a direction perpendicular to the direction of movement of the strip and perpendicular to the edge, and a first profile detector for detecting a portion of the first laser beam passing over the edge, and for generating a first signal representing the detected portion of the first laser beam; and a second laser system, in close proximity to the first laser system, including a second projector for projecting a second laser beam at the edge, in a direction perpendicular to the direction of movement of the strip and perpendicular to the cutting edge, and a second profile detector for detecting a portion of the second laser beam passing over the edge and for generating a second signal representing the detected portion of the second laser beam.

11. The apparatus according to claim 10, characterized in that it additionally comprises: a normalizing circuit that receives the first and second signals from the first and second profile detectors, substantially eliminating filtering the movement of the edge, and generating a discontinuity signal of edge; and a defect detection circuit that receives the edge discontinuity signal, processes the edge discontinuity signal to detect defects at the edge, and generates a defect signal in response to the detected defects.

12. The apparatus according to claim 10, characterized in that the defect detection circuit detects defects by detecting corresponding peaks of opposite polarity within a predetermined amount of time in the edge discontinuity signal.

13. The apparatus according to claim 10, characterized in that the predetermined amount of time is dependent on the speed with which the strip moves and the distance between the first and second laser beams.

14. A display system, characterized in that it comprises: a camera for forming images of an edge of a strip of material in movement, a monitor, coupled to the camera, for displaying images captured by the camera, and a storage system, coupled with the camera and the monitor, to store images captured by the camera, the monitor is able to display the stored images.

15. The display system according to claim 14, characterized in that it comprises: a light source directed to the edge of the strip; a first chamber and a lens, in close proximity to the light source and on a first side of the strip, wherein the first chamber and the lens take graphic representations of the edge of the strip as the strip passes in front of the first chamber and lens; a second camera and lens, in close proximity to the light source and on a second side of the strip, wherein the second camera and lens take graphic representations of the edge of the strip as the strip passes in front of the second camera and lens; and a computer that receives graphic representations of the first and second cameras, generates digitized images of the graphic representations and displays the digitized images on a monitor.

16. The display system according to claim 15, characterized in that it additionally comprises an inspector that detects defects in the edge of the strip and generates a defect signal when a defect is detected, in which the computer receives the defect signal and causes the first camera and the lens, and the second camera and lens to take graphic representations of the strip at a predetermined time so that graphic representations of the detected defects are taken.

17. The display system according to claim 16, characterized in that the predetermined time is adjustable, and in which the computer includes an input device through which an operator can introduce synchronization settings to which the computer responds by causing the first camera and lens, and the second camera and lens take graphic representations of the strip, and a predetermined time set when the defective signals are received.

18. The display system according to claim 15, characterized in that it additionally comprises a storage system that stores the digitized images in response to instructions from the computer, and in which the operator using an input device of the computer can cause the computer to retrieve the stored digitized images from the storage system for display on the monitor. A continuous edge of a strip of material (14) is processed as it moves in a direction along the length of the strip. The condition of a continuous edge of the strip and movement is inspected after it has been processed. The strip is cut into pieces. The pieces are classified into groups, in response to the condition of the edge or edge. The edge is continuously monitored by a pair of laser beams and parallel, closely spaced. A system (25) Display includes a camera, a monitor and a storage system. The camera produces images of an edge of the moving strip of the material. The monitor, coupled to the camera, shows images captured by the camera. The storage system, coupled to the camera and the monitor, stores the images captured by the camera. The monitor is capable of displaying the stored images.