CN106236016A - A kind of in-situ projection method for vein displaying picture - Google Patents

Abstract

The invention relates to an in-situ projection method for vein imaging. The implementation steps are as follows: open the near-infrared module and the visible light module to ensure uniform illumination of the area to be tested, use the image acquisition module to capture images, and simultaneously acquire visible light images and near-infrared images ;According to the contrast enhancement formula, the image processing module enhances the vein information and waits for projection; place the concentric circle test piece at an appropriate distance in front of the image acquisition module, and fine-tune the camera; during projection calibration, the projector sequentially projects a blue visible light rectangular image of a determined size And correct the red visible light rectangular image of the position coordinates, and correct it cyclically until the offset error is less than 2mm; after the calibration is completed, the enhanced image is projected to the detection area in situ in the green light mode. The invention can ensure the in-situ presentation of the virtual image and accurately guide the puncture process.

Description

An in situ projection method for vein imaging

technical field

The invention relates to the field of auxiliary medical treatment, in particular to an in-situ projection method for vein imaging.

Background technique

Venipuncture is a common clinical medical method, such as drug injection, infusion, and blood extraction. At present, in traditional operations, medical staff mainly locate blood vessels visually or empirically. However, for some special patients with thin blood vessels, thick fat or collapsed blood vessels (such as children, obese people, the elderly, etc.), the blood vessels are difficult to distinguish, which seriously affects the success rate of venipuncture. Therefore, there is an urgent need for a positioning-assisted vein imaging method to guide puncture.

At present, the mainstream vein imaging methods mostly use the principle of near-infrared vein imaging to enhance the contrast between veins and surrounding tissues, such as the Chinese patent "A Vein Imaging System" (patent application number: 201320064740.2), "A Infusion Aid Vein Developing Apparatus" (patent application number: 201320007342.7), etc. This type of imager uses a transmissive light source, the detection area is limited, and the imaging result is displayed on a display, which is not convenient for actual operation. Since then, there have been projection imaging devices using direct light sources, such as "A Dual Optical Path System Suitable for Peripheral Vascular Vein Imaging Devices" (patent application number: 201320376695.4) Chinese patent "A Vein Imaging Method and Vein Imaging Device" Image System" (patent application number: 201510606622.3), etc. However, the above solutions all use a hot mirror structure, which is fragile and the installation of supporting optical components is complicated. At the same time, the calibration process is completed by experience, which is not conducive to the in-situ presentation of vein projection.

Contents of the invention

The problem to be solved by the present invention is to provide an in-situ projection method for vein imaging in view of various defects of the projection imaging device in the prior art, which can ensure the in-situ presentation of the virtual image and accurately guide the puncture process.

The present invention solves the problems referred to above and realizes through the following technical solutions:

A: Place the vein to be tested in the designated detection position;

B: Turn on the near-infrared module and visible light module to ensure that the area to be tested is evenly illuminated;

C: Use the image acquisition module to capture images, and simultaneously acquire visible light images and near-infrared images;

D: The image processing module preprocesses the collected images to remove interference and noise in the vein images;

E: According to the contrast enhancement formula, the image processing module enhances the vein information and waits for projection;

F: Place the concentric circle test piece at an appropriate distance in front of the image acquisition module, and fine-tune the camera position to ensure that the collected image is coaxial with the concentric circle and parallel to the edge of the detection surface;

G: Before projection, adjust the optical path structure of the projector to ensure that the projected image and the collected image are completely overlapped;

H: During projection calibration, the projector sequentially projects a visible light rectangular image with a determined size and another visible light rectangular image with corrected position coordinates, and cyclically corrects until the offset error is less than 2mm;

I: There are 3 calibration surfaces for projection calibration, which are the coincident surface in F and G (that is, the clearest image taken by the camera), as well as the minimum depth of field surface of the camera and the maximum depth of field surface of the camera, and repeat step H respectively;

J: After the calibration is completed, the enhanced image in E is projected onto the detection area in situ using the green light mode.

In the step A, the vein to be measured area may be the head, neck, back of the hand, back of the foot, inner side of the elbow, etc.

In the step B, the near-infrared module refers to a high-power near-infrared LED equipped with a uniform light device, which can be frosted glass, a light uniform plate with a light transmittance of more than 70%, or a Fresnel lens. The near-infrared light band can be one or a combination of 760nm, 850nm, and 940nm. The visible light module can be an LED in the 450nm-650nm band, or it can be natural light in the external environment.

In the step C, the image acquisition module mainly automatically separates the near-infrared image and the visible light image through the multi-spectral camera. The multi-spectral camera here has a multi-lens type, a multi-camera type, and a light-speed separation type. The 2CCD camera effect based on the light-speed separation optimal.

In the step D, the image processing module uses a scan window of 5×5 pixels to scan the obtained image to be tested in the order from top to bottom and from left to right, and calculates the mean value and Variance Var, if the variance Var is greater than the set threshold T _D , the fast median filtering method is used to smooth the point to remove the interference and noise in the vein image to be measured.

In the step E, the contrast enhancement formula is f _boost =m(f _nir -nf _vis ). Among them, f _boost is the enhanced image, f _nir is the near-infrared image, f _vis is the visible light image, m is the scaling factor (determined according to the maximum unsaturation of the image pixel value), and n is determined by the formula n=(h _nir / h _vis )[(G _boost -G _nir )/(G _boost -G _vis )] to get. In the calculation formula of n, h _nir and h _vis are the gray average intensity of the near-infrared image and the visible light image respectively, and G _boost is the gray contrast ratio between the vein and other tissues in the desired enhanced image (experimentally adjustable value), G _nir and G _vis are the gray contrast ratios of veins and other tissues in near-infrared images and visible light images, respectively, and are determined by the formula G=|K _vein -K _skin |/(K _vein +K _skin ). In this formula, K _vein is the average gray value of the vein region in the corresponding image, and K _skin is the average gray value of other tissues adjacent to the vein in the corresponding image.

In step F, the concentric circles are kept coaxial with the center of the CCD imaging surface, and both are perpendicular to the ground plane, and the upper and lower edges of the CCD are parallel to the horizon.

In the step G, assuming that the projector is located on the concentric axis of F, and the size of the collected image is known, then the imaging formula _fc = _lcdc / _{Lc ( fc} represents the focal length of the camera, and _lc represents the focal length of the image The horizontal width, L _c represents the horizontal width of the captured image area) can calculate the camera working distance d _c . At the same time, from the camera imaging ratio ρ _c =d _c /L _c and the projector projection ratio ρ _p =d _p /L _p (d _p represents the distance from the projection screen to the projector, L _p represents the horizontal width of the projection screen), if Acquisition image and projected image overlap, need to meet That's it. After completion, fine-tune the projector horizontally for a certain distance to prevent interference with the camera installation position.

In the above steps H and I, there are three times of projection calibration, which are respectively the surface with the best depth of field of the camera, the surface with the minimum depth of field of the camera and the surface with the maximum depth of field of the camera. For each calibration, the projector first projects a blue rectangle of a certain size on the white plane to be measured, and the camera collects the image and submits it to the image processing module, which automatically calculates the coordinates of the four corners. Wherein, the standard position coordinates of the rectangle are known (can be calculated automatically or stored in the registry). Then, calculate the deviation from the standard coordinates at this time, and after correction, project a red rectangle through the projector. If the calculated deviation is less than 2mm again, the calibration is over, otherwise the corrected blue rectangle is thrown again, and the cycle is met until the condition is met. The purpose of casting the blue and red rectangles in sequence here is to prevent the situation where the calibration rectangles cannot be recognized twice before and after the calibration is too fast.

In the steps G, H, I, and J, the projector used is based on the DLP technology, controls the DMD chip through the DSP, and satisfies the requirements of overlapping optical path projection in G to project images.

Compared with prior art, the beneficial effect of the present invention is:

(1) An in-situ projection method for vein imaging proposed by the present invention replaces the traditional hot mirror structure, the optical path is simple and compact, and it is easy to operate and use;

(2) A kind of in-situ projection method for vein imaging proposed by the present invention realizes software automatic calibration and corrects projection deviation in real time;

(3) An in-situ projection method for vein imaging proposed by the present invention presents virtual images in situ and accurately guides puncture operations.

Description of drawings

FIG. 1 is a structural diagram of a preferred embodiment of the in-situ projection method of the present invention.

Fig. 2 is a flowchart of the in-situ projection method of the present invention.

Fig. 3 is a kind of projection lens structural diagram of in-situ projection method of the present invention; (a) is that this lens group has 4 lenses, working distance 261mm, angle of view 16 degrees, actual focal length 25.9mm, relative illuminance more than 90%. (b) is the location distribution of the projection lens, and (c) is four lenses;

Fig. 4 is a schematic diagram of the calibration process of the vein imaging method of the present invention.

detailed description

The present invention will be further described below in combination with examples of implementation (with accompanying drawings). These embodiments are only used to illustrate the present invention and not to limit the scope of the present invention. In addition, it should be understood that after reading the teachings of the present invention, those skilled in the art can make various changes and modifications to the present invention, and these equivalent forms also fall within the scope defined by the appended claims of the present application.

The structural diagram of a preferred implementation of the in-situ projection method of the present invention is shown in FIG. 1 : the implementation includes a near-infrared light source module 101 , an image acquisition module 102 , an image processing module 103 and an in-situ projection module 104 .

The near-infrared light source module 101 includes two Fresnel lenses 111 , a reflective cup 112 and two 3-5W high-power near-infrared LEDs 113 , and is located at the front side of the image acquisition module 102 . Two power near-infrared LEDs are about 70mm apart, and the wavelength bands are 850nm and 960nm respectively. The 940nm LED is located in the reflection cup, which can diffusely reflect the outside natural light to the area to be tested at the same time.

The image acquisition module 102 includes a multispectral camera 121 and an automatic iris lens 122 . Among them, the multispectral camera is a beam splitting 2CCD camera.

The image processing module 103 includes a central control unit 131 and an image processing algorithm. The central control unit 131 is one or a combination of digital signal processor (DSP), programmable gate array (FPGA), microprocessor (ARM) or industrial computer.

The in-situ projection module 104 includes a digital light processing projector (DLP) 141 and an in-situ projection algorithm. The micro projector is a digital light processing projector (DLP), which is composed of a DMD chip, a DLP circuit, an image controller, a color wheel, a converging lens, a bright light source, and a projection lens.

According to this embodiment, the flow chart of the in-situ projection method of the present invention is as shown in Figure 2:

1) Place the vein area to be tested flat on the detection position;

2) Turn on the high-power near-infrared LEDs of 850nm and 940nm;

3) Simultaneously acquire visible light images and near-infrared images through a beam splitting 2CCD camera;

4) Use the image processing algorithm in the industrial computer to enhance the vein information and wait for projection;

5) Place the concentric circle test piece at 260mm in front of the camera, fine-tune the camera until it is clear, and ensure that the collected image is coaxial with the concentric circle and parallel to the edge of the detection surface;

6) Open the projection lens of the installation design, fine-tune to make the projection image and the collected image completely overlap;

7) Projection calibration is performed on the surface with the best depth of field of the camera, the surface with the smallest depth of field of the camera, and the surface with the largest depth of field of the camera;

8) The vein image in 4) is projected onto the back of the hand in situ with green light mode.

According to this implementation example, a projection lens structure diagram of the in-situ projection method of the present invention is shown in (a) in Figure 3: the lens group has 4 lenses in total, the working distance is 261 mm, the field of view is 16 degrees, and the actual focal length is 25.9 mm, relative illuminance over 90%. (b) is the location distribution of projection lenses, and (c) is the actual size of each lens.

Fig. 3 is a kind of projection lens structural diagram of in-situ projection method of the present invention; (a) is that this lens group has 4 lenses, working distance 261mm, angle of view 16 degrees, actual focal length 25.9mm, relative illuminance more than 90%. (b) is the position distribution of the projection lens, (c) is four lenses, and the four lenses are set by optical principles to achieve the function of focusing and setting the size of the field of view.

(c) The upper left lens in (c) is a double lens with a circular cross-section, a diameter of 9.00±0.05mm, a curvature radius of 55.85mm at the front and rear, a lens center thickness of 1.5±0.05mm, and a side thickness of 1.14mm; (c) in Figure 3 ) The upper right lens is a concave-convex lens with a circular cross-section and a diameter of 9.00-0.05mm. The radius of curvature of the front surface is 8.00mm, the radius of curvature of the rear surface is 6.35mm, and the diameter of the latter surface is 7.00mm. The thickness of the lens center is 2.20± 0.05mm, side thickness 1.07mm; the lower left lens in (c) in Figure 3 is a plano-convex lens with a circular cross-section and a diameter of 9.00-0.05mm. The surface diameter is 8.40mm, the lens center thickness is 1.20±0.05mm, and the side thickness is 2.25mm; the lower right lens in (c) in Figure 3 is a double lens with a rectangular cross section, 9-0.05mm high and 7-0.05mm wide , the radius of curvature of the front and rear surfaces is 29.65mm, the thickness of the center of the lens is 2.00±0.05mm, and the thickness of the side is 1.31mm.

According to this implementation case, the schematic diagram of the calibration process of the in-situ projection method of the present invention is shown in Figure 4: there are three calibrations in projection calibration, which are the surface 2 with the best camera depth of field, the surface 3 with the smallest camera depth of field, and the surface 4 with the largest camera depth of field. For each calibration, the projector first projects the first visible light rectangle in step H of 64mm*36mm on the white plane to be tested, and the camera collects the image and submits it to the image processing module, which automatically calculates the values of a, b, c, and d. 4 corner coordinates. Wherein, the standard position coordinates of the rectangle are known (the rectangle abcd in the figure). Then, calculate the deviation from the standard coordinates at this time, and after correction, project the second visible light rectangle in step H through the projector. If the calculated deviation is less than 2mm again, the calibration is over, otherwise the corrected first visible light rectangle is cast again, and the cycle is repeated until the condition is met. Here, the purpose of sequentially projecting the above two different visible light rectangle images is to prevent the calibration from being too fast and cause the two correction rectangles to be unrecognizable.

Claims (10)

1. the in-situ projection method for vein displaying picture, it is characterised in that realize step as follows:

A: region to be measured for vein is positioned over the detecting position specified；

B: open near-infrared module and visible ray module, it is ensured that region to be measured uniform illumination；

C: image capture module absorbs image by camera, obtains visible images and near-infrared image simultaneously；

D: image processing module carries out pretreatment to gathering image, removes the interference in vein image and noise；

E: strengthen formula according to contrast, image processing module strengthens venous information, waits to be projected；

F: concentric circular test block is placed in suitable distance on front side of image capture module, finely tunes camera position, it is ensured that gather image with Concentric circular is coaxial and detection faces top edge is parallel；

G: before projection, adjusts the light channel structure of projector, it is ensured that projection picture and collection image are completely superposed；

H: during Projection surveying, projector projects three kinds of visible ray rectangular images that color that size determines is different, Yi Zhongke successively Seeing light rectangle and the visible ray rectangle of another kind of correction position coordinate, circulation is revised, until offset error is less than 2mm；

Totally 3 demarcation of I: Projection surveying, are that the coincidence face in F and G step, i.e. camera are adopted figure and the most clearly located respectively, and camera Minimum depth of field face and camera the maximal field depth face, each repeat H step；

J: after having demarcated, uses the third visible mode in-situ projection to detection region by the enhancing image in E.

In-situ projection method for vein displaying picture the most according to claim 1, it is characterised in that: in described step B, Near-infrared module refers to high-power near-infrared LED, is furnished with dodging device, can be clouded glass, the even tabula rasa of more than 70% light transmittance, Or Fresnel Lenses；Near infrared light wave band is one or several the set in 760nm, 850nm, 940nm.

In-situ projection method for vein displaying picture the most according to claim 1, it is characterised in that: in described step B, Visible ray module is the LED of 450nm～650nm wave band, it is also possible to be the natural light in external environment.

In-situ projection method for vein displaying picture the most according to claim 1, it is characterised in that: in described step C, Image capture module is automatically separated near-infrared image and visible images by multispectral camera, and described multispectral camera has many mirrors Head dummy, heterogeneous type and light velocity divergence type, the 2CDD camera best results wherein separated based on the light velocity.

In-situ projection method for vein displaying picture the most according to claim 1, it is characterised in that: in described step D, It is suitable that image processing module uses in the scanning window of 5 × 5 pixels testing image to obtaining according to from top to bottom, from left to right Sequence is scanned, average and variance Var in each image in calculating scanning window, if variance Var is more than setting threshold value T_D, the most right This point uses Fast Median Filtering method to be smoothed, and removes the interference in vein image to be measured and noise.

In-situ projection method for vein displaying picture the most according to claim 1, it is characterised in that: in described step E, It is f that contrast strengthens formula_boost=m (f_nir-nf_vis), wherein, f_boostFor image, f after strengthening_nirFor near-infrared image, f_vis For visible images, m is scaling coefficient (determining according to the maximum degree of unsaturation of image pixel value), and n is by formula n= (h_nir/h_vis)[(G_boost-G_nir)/(G_boost-G_vis)] obtain；In the computing formula of n, h_nirAnd h_visIt is respectively near-infrared image With the gray scale mean intensity of visible images, G_boostIntensity contrast for desired enhancing image medium-sized vein Yu other tissue Rate, G_nirFor the intensity contrast rate of near-infrared image Yu other tissue, G_visAsh for visible images medium-sized vein Yu other tissue Degree contrast ratio, by formulaWithDetermine,For the average gray value in near-infrared image medium-sized vein region,Other tissue for neighbouring in near-infrared image vein Average gray value,It is respectively the average gray value in visible images medium-sized vein region and adjacent to vein tissue Average gray value.

In-situ projection method for vein displaying picture the most according to claim 1, it is characterised in that: in described step F, Concentric circular keeps coaxial with CCD imaging surface center, is both perpendicular to ground level simultaneously, and CCD lower edges is parallel to horizon.

In-situ projection method for vein displaying picture the most according to claim 1, it is characterised in that: in described step G, Assume that projector is positioned on the mandrel of F, gather image size it is known that then by imaging formula f_c=l_cd_c/L_cCamera work can be calculated Make distance d_c, f_cRepresent camera focus, l_cRepresent the horizontal width of image, L_cRepresent the horizontal width gathering image-region；With Time, camera shooting compare ρ_c=d_c/L_cρ is compared with projector projects_p=d_p/L_p, d_pRepresent that projected picture is to the distance of projector, L_p Representing the horizontal width of projected picture, if gathering image and projection picture registration, need to meet?；After completing, will Projector WidFin one segment distance, prevents from interfering with camera installation position.

In-situ projection method for vein displaying picture the most according to claim 1, it is characterised in that: described step H and I In, carry out multiple projections demarcation, in the optimal face of the camera depth of field, camera depth of field minimal face and demarcating during camera depth of field largest face；Often Secondary timing signal, first projector projects the first the visible ray rectangular image in step H that size determines in white plane to be measured, Meeting at image processing module after collected by camera image, this module calculates 4 angular coordinates, wherein, the normal place of rectangle automatically Coordinate is known；Then, calculate now with standard coordinate deviation, after correction, in projector launches step H for correction position sit Target visible ray rectangular image；If again calculating deviation to be less than 2mm, demarcation terminates, and the most again launches and sits for correction position Target visible ray rectangular image, is recycled to and meets condition.

In-situ projection method for vein displaying picture the most according to claim 1, it is characterised in that: described step G, In H, I, J, the projector used, based on DLP technology, controls dmd chip by DSP, meets light path in G and projects wanting of coincidence Seek projects images.