Method for Adjusting Symmetry of Optical Power of Cone Component of Dual-Probe Lit Laser Module

Abstract

The present invention belongs to the field of optical engineering, and specifically relates to a method for adjusting the optical power symmetry of a cone assembly of a double-probe illuminated laser module, including the following steps: (1) designing a device for adjusting the symmetry of the laser module cone assembly; (2) collecting characteristic images at the bottom of the laser module; (3) designing a lighting scheme; (4) image processing; (5) target feature extraction; (6) image angle compensation; (7) adjusting the optical power symmetry; The present invention uses the method of energizing the double probes, completely energizing from the bottom of the laser module, so that the laser module with an insulated outer cylinder can be normally illuminated in the improved double-probe JAS-15 chuck; Four groups of optical power acquisition components clamp the directions of two strong lights and two weak lights, which can effectively improve the adjustment accuracy and shorten the adjustment time; The sobel operator is used to extract the target contour, and the mathematical morphology method is used to process the target contour, and the edge integrity of the extracted graphic target features is better.

Description

Technical Field

[0001] The present invention belongs to the field of optical engineering, and particularly relates to a method for adjusting the optical power symmetry of a cone component of a double-probe lit laser module. Background Art

[0002] With the increasing number of application scenarios of laser modules, people are constantly innovating the functions of laser modules, continuously improving the light projection accuracy and service life of laser modules, and maximizing the use value of laser modules. By installing a cone component above the light-emitting surface of the laser module, it is possible to reflect the linear light emitted by the laser module by 90°, so that the light emitted by one laser module can be projected onto four planes in a three-dimensional space, forming a light ring, which is convenient for people's production operations.

[0003] Generally, when powering on and adjusting a laser module, one probe is in contact with the electrode of the bottom chip of the laser module, and the other probe is in contact with the outer sleeve of the laser module, so that the laser module emits light when powered on. As people's requirements for the appearance of laser modules are getting higher and higher, a layer of paint is often applied to the outer sleeve of the laser module to make the laser module more aesthetically pleasing, but this increases the difficulty of powering on the laser module. Therefore, a power-on method needs to be designed so that the laser module has both a beautiful appearance and can be powered on normally.

[0004] In addition, the quality of laser modules on the market is uneven at present, and the accuracy of the light rings projected by laser modules on the four planes in a three-dimensional space is uneven. The main reason is that the position accuracy of the cone components on the laser module is inaccurate, so that the cone center of the cone component does not align with the light source center emitted by the laser module, resulting in the light emitted by the laser module not forming symmetry between the strong power sides and between the weak power sides. As a result, the light rays projected by the laser module through the cone component onto the four planes in a three-dimensional space are not at the same horizontal position, affecting the use accuracy. Therefore, it is necessary to continuously optimize the scheme for adjusting the accurate installation position of the cone component on the laser module to improve the light projection accuracy of the laser module. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides a method for adjusting the optical power symmetry of a cone component of a double-probe lit laser module, which can effectively power on and light up the laser module with an insulating material coated on the outer cylinder from the bottom of the module, and can accurately place the cone component above the light-emitting body of the laser module, so that the optical power of the light emitted by the laser module is similar or even equal in any symmetric direction of 360° in the plane, thereby making the horizontal height of the light rays emitted by the laser module consistent in the plane of 360°.

[0006] The technical solution of the present invention is as follows: A method for adjusting the optical power symmetry of a cone component of a double-probe lit laser module includes the following steps:

[0007] (1) Design a symmetry adjustment device for the laser module cone assembly, and use the symmetry adjustment tooling for the laser module cone assembly to adjust the accurate placement position of the cone assembly on the laser module;

[0008] (2) Collect the bottom feature image of the laser module, find effective and reliable target features, calculate the position information of the current target features, so as to perform position and angle compensation on the laser module;

[0009] (3) Design a lighting scheme to clearly highlight the image information at the bottom of the laser module;

[0010] (4) Image processing: Perform image enhancement processing on the bottom image of the laser module collected by the industrial camera, remove the influence of ambient light, decompose the enhanced target image into color channels, and obtain the grayscale image with the largest contrast between the target features and the background;

[0011] (5) Target feature extraction: After performing histogram equalization processing on a single color channel, use the sobel operator to extract the target contour, and then use mathematical morphology methods for processing to obtain a good target contour;

[0012] (6) Image angle compensation: Compensate the position and angle of the current laser module obtained in step (5) to rotate the laser module to the standard position, so that the laser module can be normally lit by powering on from the bottom of the module;

[0013] (7) Optical power symmetry adjustment

[0014] Automatically place the laser module on the adjustment station. After completing steps (4), (5), and (6), automatically place the cone assembly on the light-emitting body of the module, and adjust the cone assembly forward, backward, left, and right. When the optical power values of the laser emitted after being reflected by the cone assembly in the symmetric direction are within the accuracy error range, keep the cone assembly stationary, apply ultraviolet glue for curing, complete the position installation of the laser module cone assembly, that is, complete the optical power symmetry adjustment of the laser module cone assembly.

[0015] Further, the bottom feature acquisition system of the laser module includes an AOI light source, a Midevision industrial camera MV-USB131GC, an industrial control computer, an independent 24V DC power supply, and automated industrial software; the 24V DC voltage provides power for the AOI light source, the industrial control computer, and the Midevision industrial camera, and the automated industrial software performs bottom image acquisition, target feature extraction, angle compensation, and cone assembly installation position adjustment of the laser module.

[0016] Further, the specific method for designing the symmetry adjustment device of the laser module cone assembly in step (1) is as follows: Place 4 optical power acquisition components at the vertices of a 200mm×200mm square platform. Each optical power acquisition component is standardized and customized using metal materials, and the surface and interior of each optical power acquisition component are blackened. A PCB board is installed behind the head of each optical power acquisition component. The PCB board is evenly divided into 6 vertical strips, each strip having a width of 4mm, and is sequentially labeled as positions 1, 2, 3, 4, 5, and 6. Then, two optical power acquisition chips are installed at positions 1 and 6, and the remaining positions are left vacant. The 8 optical power acquisition chips inside the optical power acquisition components at the vertices of the 200mm×200mm square platform are sequentially labeled clockwise as: 1, 2, 3, 4, 5, 6, 7, 8. Connect the 8 optical power acquisition chips to the single-chip microcomputer in sequence according to the labels. Install a JAS-15 chuck at the center of the 200mm×200mm square platform. The chuck is connected to the single-chip microcomputer, and the laser module is powered on. Transmit the optoelectronic data collected by the 8 optical power acquisition chips to the single-chip microcomputer for real-time data processing to obtain the real-time optical power data collected by each optical power acquisition chip.

[0017] Further, the detailed process of the image processing method for the bottom of the laser module collected in step (4) is as follows: Use the Retinex theory to enhance the image of the bottom of the collected laser module. The Retinex theory assumes that the original image S(x, y) is the product of the illumination image L(x, y) and the reflectance image R(x, y), which is expressed by formula (1):

[0018] S(x, y) = R(x, y)·L(x, y) Formula (1)

[0019] The purpose of image enhancement based on Retinex is to estimate the illumination L(x, y) from the original image S(x, y), thereby decomposing R(x, y), eliminating the influence of uneven illumination, and improving the visual effect of the image. Usually, the image is transferred to the logarithmic domain, that is, s(x, y) = log s(x，y) , l(x, y) = log L(x，y) , r(x, y) = log R(x，y) , and then the product relationship is converted into a sum relationship. The core of the Retinex method is to estimate the illuminance L(x, y), estimate the L(x, y) component from the image S(x, y), solve the reflection component r(x, y) from S(x, y) and I(x, y). The reflection component r(x, y) is the inherent attribute of the image, and then it is transferred to the real number domain to find R(x, y), which is expressed by formulas (2) - (6) as:

[0020] log S(x，y) = log R(x，y)·L(x，y) Formula (2)

[0021] log S(x，y) = log R(x，y) + log L(x，y) Formula (3)

[0022] s(x, y) = r(x, y) + I(x, y) Formula (4)

[0023] r(x, y) = s(x, y) - I(x, y) Formula (5)

[0024] R(x, y) = log r(x，y) Formula (6)

[0025] After calculating R(x, y), perform R, G, B three-color channel decomposition.

[0026] Furthermore, the specific method for target feature extraction in step (5) is as follows: for the R, G, B three-color image channels after color space decomposition, first use the histogram equalization algorithm to enhance the R, G, B three-color channels respectively. The histogram equalization algorithm is as follows:[[]]

[0027] Assume an image has L levels of gray scale, n k is the number of pixels with gray scale k in the image f(x, y), and n is the total number of pixels in the image. Then the probability P S (S k ) is shown in Formulas (7) and (8) as follows:[[]]

[0028] where K = 0, 1, 2, 3,... L - 1 Formula (7)

[0029] S k = K / (L - 1) Formula (8)

[0030] Taking S k as the independent variable and P S (S k ) as the function, the obtained curve is the histogram of the image. Its information entropy is shown in Formula (9) as follows:[[]]

[0031]

[0032] The information entropy H of the entire image is shown in Formulas (10) and (11) as follows:[[]]

[0033]

[0034]

[0035] An image with a uniform distribution of histograms has the maximum information entropy value H. That is, when the formula (11) is satisfied, the formula (10) has the maximum information entropy value H max . Transforming the histogram of the image into an approximately uniform distribution, this process is the histogram equalization of the image, as shown in formula (12):

[0036]

[0037] where t k is the gray level of each pixel after histogram equalization. In actual calculations, t k needs to be rounded.

[0038] After performing histogram equalization on the R, G, and B color channels, then select one color channel with the largest contrast between the target and the background in the image. For example, in the R color channel, if the contrast between target 1 and the background is the largest, then select the R color channel; in the G channel, if the contrast between target 2 and the background is the largest, then select the G color channel; in the B color channel, if the contrast between target 3 and the background is the largest, then select the B color channel;

[0039] Next, use the sobel operator to extract the target contour in the selected color channel. The sobel edge detection operator performs gray-level weighted operations on the upper, lower, left, and right neighborhoods of the pixel points in the image using a 3×3 template. The sobel operator is shown in formulas (13) and (14):

[0040] G x [i, j] = f[i - 1, j + 1] + 2×f[i, j + 1] + f[i + 1, j + 1] Formula (13)

[0041] -f[i - 1, j - 1] - 2×f[i, j - 1] - f[i + 1, j - 1]

[0042] G y [i, j] = f[i + 1, j - 1] + 2×f[i + 1, j] + f[i + 1, j + 1] Formula (14)

[0043] -f[i - 1, j - 1] - 2×f[i - 1, j] - f[i - 1, j + 1]

[0044] The convolution templates are represented by formulas (15) and (16):

[0045]

[0046]

[0047] Finally, the mathematical morphology method is used to process the target contour in the image extracted by the Sobel operator, removing the isolation outside the target, filling the small holes inside the contour, connecting adjacent objects, and smoothing the boundary;

[0048] The formula for opening operation is: represents the erosion operation, that is, the original image A is eroded by the structuring element S, represents that the original image A is eroded by the structuring element S and then dilated by the structuring element S;

[0049] The formula for closing operation is: represents the dilation operation, that is, the original image A is dilated by the structuring element S, represents that the original image A is dilated by the structuring element S and then eroded by the structuring element S.

[0050] The beneficial effects of the present invention are as follows: Compared with the prior art, the present invention has the following advantages:

[0051] (1) The present invention adopts the method of energizing with two probes, completely energizing from the bottom of the laser module, so that the laser module with an insulated outer cylinder can be normally lit in the improved two-probe JAS-15 chuck;

[0052] (2) Usually, the laser module cone assembly symmetry adjustment tooling uses 4 optical power acquisition chips placed at the four vertices of a square to receive optical power data in real time. However, generally, the light intensity emitted by the laser module is not uniform in a 360-degree plane, and there are usually strong light sides and weak light sides. Obviously, using only 4 optical power acquisition chips placed at the four vertices of a square to collect optical signals cannot effectively adjust the symmetry of the cone assembly on the laser module. With the laser module cone assembly optical power adjustment tooling proposed by the present invention, two optical power acquisition chips are installed on a PCB board behind the head of an optical power acquisition component at a certain distance. In this way, the strong light emitted by the laser module can be sandwiched between the two optical power acquisition chips. Similarly, the weak light emitted by the laser module can be sandwiched between the two optical power acquisition chips. In this way, the four groups of optical power acquisition components clamp the directions of two strong lights and two weak lights, which can effectively improve the adjustment accuracy and shorten the adjustment time;

[0053] (3) The image enhancement scheme using the Retinex theory proposed by the present invention can effectively solve the problem of uneven illumination of the image, remove the ambient light, and improve the quality of the target image;

[0054] (4) The present invention decomposes the enhanced image into three color channels of R, G, and B, which can distinguish different targets from the background, reduce the difficulty of extracting target features, and then uses the histogram equalization algorithm to perform gray-scale enhancement processing on each color channel to further improve the contrast between the target and the background, and can further improve the integrity and efficiency of extracting the required target features;

[0055] (5) The present invention uses the sobel operator to extract the target contour and processes the target contour using the mathematical morphology method, that is, uses the opening operation to remove the isolated points outside the target, and then uses the closing operation to fill the small holes inside the contour, connect adjacent objects, and smooth the boundary. This method can not only effectively remove the isolated points in the target contour extracted by the sobel operator and fill the contour holes, but also further improve the integrity of the image edge contour and make the shape feature extraction more accurate. Compared with the prior art, the edge integrity of the extracted graphic target features is better. Description of the Drawings

[0056] Figure 1 is the flowchart of the implementation of the present invention;

[0057] Figure 2 is the schematic diagram of the laser module cone symmetry adjustment tooling for the embodiment of the present invention;

[0058] Figure 3 is the target contour extracted by the sobel operator in the embodiment;

[0059] Figure 4 is the schematic diagram of the target image after the opening operation in the embodiment;

[0060] Figure 5 is the schematic diagram of the target image after the closing operation in the embodiment. Detailed Embodiment

[0061] The following further describes the specific implementation manners of the present invention in conjunction with the drawings:

[0062] As Figures 1-5 shown, a method for adjusting the cone component optical power symmetry of a double-probe lighting laser module includes the following steps:

[0063] (1) Design a laser module cone component symmetry adjustment device, and use the laser module cone component symmetry adjustment tooling to adjust the accurate placement position of the cone component on the laser module;

[0064] (2) Collect the bottom feature image of the laser module, find effective and reliable target features, calculate the position information of the current target features, and thus perform position and angle compensation on the laser module;

[0065] (3) Design a lighting scheme to clearly highlight the image information at the bottom of the laser module;

[0066] (4) Image processing: perform image enhancement processing on the bottom image of the laser module collected by the industrial camera to remove the influence of ambient light, decompose the enhanced target image into color channels, and obtain a grayscale image with the largest contrast between the target features and the background;

[0067] (5) Target feature extraction: After performing histogram equalization processing on a single color channel, use the sobel operator to extract the target contour, and then use mathematical morphology methods for processing to obtain a good target contour;

[0068] (6) Image angle compensation: Compensate the position angle of the current laser module obtained in step (5) to rotate the laser module to the standard position, so that the laser module can be normally lit by powering on from the bottom of the module;

[0069] (7) Optical power symmetry adjustment

[0070] Automatically place the laser module on the adjustment station. After completing steps (4), (5), and (6), automatically place the cone assembly on the light-emitting body of the module, and adjust the cone assembly forward and backward, left and right. When the optical power values of the laser emitted after being reflected by the cone assembly in the symmetric direction are within the accuracy error range, keep the cone assembly stationary, apply ultraviolet glue for curing, and complete the position installation of the cone assembly of the laser module, that is, complete the optical power symmetry adjustment of the cone assembly of the laser module.

[0071] The bottom feature acquisition system of the laser module includes an AOI light source, an industrial camera MV-USB131GC from Medvision, an industrial control computer, an independent 24V DC power supply, and automated industrial software; the 24V DC voltage provides power for the AOI light source, the industrial control computer, and the industrial camera from Medvision, and the automated industrial software performs bottom image acquisition, target feature extraction, angle compensation, and adjustment of the installation position of the cone assembly of the laser module.

[0072] Further, the specific method for designing the symmetry adjustment device of the laser module cone assembly in step (1) is as follows: Place 4 optical power acquisition components at the vertices of a 200mm×200mm square platform respectively. Each optical power acquisition component is standardized and customized with metal materials, and the surface and interior of each optical power acquisition component are blackened. A PCB board is installed behind the head of each optical power acquisition component. The PCB board is evenly divided into 6 vertical strips, each strip with a width of 4mm, and is marked as positions 1, 2, 3, 4, 5, and 6 in sequence. Then, two optical power acquisition chips are installed at positions 1 and 6, and the remaining positions are left vacant. The 8 optical power acquisition chips inside the optical power acquisition components at the vertices of the 200mm×200mm square platform are numbered clockwise as: 1, 2, 3, 4, 5, 6, 7, 8. Connect the 8 optical power acquisition chips to the single-chip microcomputer in sequence according to the numbers. Install a JAS-15 chuck at the center of the 200mm×200mm square platform, connect the chuck to the single-chip microcomputer, and power on the laser module. Transmit the optoelectronic data collected by the 8 optical power acquisition chips to the single-chip microcomputer for real-time data processing to obtain the real-time optical power data collected by each optical power acquisition chip.

[0073] Further, the detailed process of the image processing method for the bottom of the laser module collected in step (4) is as follows: Use the Retinex theory to enhance the image of the bottom of the collected laser module. The Retinex theory assumes that the original image S(x, y) is the product of the illumination image L(x, y) and the reflectance image R(x, y), which is expressed by formula (1):

[0074] S(x, y) = R(x, y)·L(x, y) Formula (1)

[0075] The purpose of image enhancement based on Retinex is to estimate the illumination L(x, y) from the original image S(x, y), so as to decompose R(x, y), eliminate the influence of uneven illumination, and improve the visual effect of the image. Usually, the image is transferred to the logarithmic domain, that is, s(x, y) = log S(x，y) , l(x, y) = log L(x，y) , r(x, y) = log R(x，y) , and then the product relationship is converted into a sum relationship. The core of the Retinex method is to estimate the illuminance L(x, y), estimate the L(x, y) component from the image S(x, y), solve the reflection component r(x, y) from S(x, y) and I(x, y). The reflection component r(x, y) is the intrinsic property of the image, and then it is transferred to the real number domain to find R(x, y), which is expressed by formulas (2) - (6) as:

[0076] log S(x，y) = log R(x，y)·L(x，y) Formula (2)

[0077] log S(x，y) = log R(x，y) + log L(x，y) Formula (3)

[0078] s(x, y) = r(x, y) + I(x, y) Formula (4)

[0079] r(x, y) = s(x, y) - I(x, y) Formula (5)

[0080] R(x, y) = log r(x，y) Formula (6)

[0081] After calculating R(x, y), perform R, G, B three-color channel decomposition.

[0082] Furthermore, the specific method of target feature extraction in step (5) is as follows: for the R, G, B three-color image channels after color space decomposition, first use the histogram equalization algorithm to enhance the R, G, B three-color channels respectively. The histogram equalization algorithm is as follows:[[]]

[0083] Assume an image has L gray levels, n k is the number of pixels with gray level k in the image f(x, y), and n is the total number of pixels in the image. Then the probability P S (S k ) is shown in Formulas (7) and (8):

[0084] where K = 0, 1, 2, 3,... L - 1 Formula (7)

[0085] S k = K / (L - 1) Formula (8)

[0086] Taking S k as the independent variable and P S (S k ) as the function, the obtained curve is the histogram of the image. Its information entropy is shown in Formula (9):

[0087]

[0088] The information entropy H of the entire image is shown in Formulas (10) and (11):

[0089]

[0090]

[0091] An image with a uniform distribution of histograms has the maximum information entropy value H. That is, when the formula (11) is satisfied, the formula (10) has the maximum information entropy value H max Making the histogram of the image approximately uniformly distributed, this process is the histogram equalization of the image, as shown in formula (12):

[0092]

[0093] where t k is the gray level of each pixel after histogram equalization. In actual calculation, t k needs to be rounded.

[0094] After performing histogram equalization on the R, G, and B color channels, then select the color channel with the largest contrast between the target and the background in the image. For example, in the R color channel, if the contrast between target 1 and the background is the largest, then select the R color channel; in the G channel, if the contrast between target 2 and the background is the largest, then select the G color channel; in the B color channel, if the contrast between target 3 and the background is the largest, then select the B color channel.

[0095] Next, use the sobel operator to extract the target contour in the selected color channel. The sobel edge detection operator performs a gray-level weighted operation on the upper, lower, left, and right neighborhoods of the pixel points in the image using a 3×3 template. The sobel operator is shown in formulas (13) and (14):

[0096] G x [i, j] = f[i - 1, j + 1] + 2×f[i, j + 1] + f[i + 1, j + 1] Formula (13)

[0097] -f[i - 1, j - 1] - 2×f[i, j - 1] - f[i + 1, j - 1]

[0098] G y [i, j] = f[i + 1, j - 1] + 2×f[i + 1, j] + f[i + 1, j + 1] Formula (14)

[0099] -f[i - 1, j - 1] - 2×f[i - 1, j] - f[i - 1, i + 1]

[0100] The convolution templates are represented by formulas (15) and (16):

[0101]

[0102]

[0103] Finally, the mathematical morphology method is used to process the target contour in the image extracted by the Sobel operator, remove the isolation outside the target, fill the small holes inside the contour, connect adjacent objects, and smooth the boundary.

[0104] The formula for opening operation is: represents the erosion operation, that is, the original image A is eroded by the structuring element S, represents that the original image A is eroded by the structuring element S and then dilated by the structuring element S;

[0105] The formula for closing operation is: represents the dilation operation, that is, the original image A is dilated by the structuring element S, represents that the original image A is dilated by the structuring element S and then eroded by the structuring element S.

[0106] The above embodiments and descriptions in the specification only illustrate the principle and the best embodiments of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed.

Claims

1. Method for adjusting the optical power symmetry of the cone component of a dual-probe lighting laser module, characterized in that: It includes the following steps: (1) Design a device for adjusting the symmetry of the cone component of the laser module, and use the tooling for adjusting the symmetry of the cone component of the laser module to adjust the accurate placement position of the cone component on the laser module; (2) Collect the characteristic image at the bottom of the laser module, find effective and reliable target features, calculate the position information of the current target features, so as to perform position and angle compensation on the laser module; (3) Design a lighting scheme to clearly highlight the image information at the bottom of the laser module; (4) Image processing, perform image enhancement processing on the image of the bottom of the laser module collected by the industrial camera to remove the influence of ambient light, decompose the enhanced target image into color channels, and obtain a grayscale image with the largest contrast between the target features and the background; (5) Target feature extraction: After performing histogram equalization processing on a single color channel, use the sobel operator to extract the target contour, and then use the mathematical morphology method for processing to obtain a good target contour; (6) Image angle compensation, compensate the position and angle of the current laser module obtained in step (5) to rotate the laser module to the standard position, so that the laser module can be normally lit by being powered on from the bottom of the module; (7) Optical power symmetry adjustment Automatically place the laser module on the adjustment station. After completing steps (4), (5), and (6), automatically place the cone component on the light-emitting body of the module, and adjust the cone component forward, backward, left, and right. When the optical power values of the laser emitted after being reflected by the cone component in the symmetric direction are within the accuracy error range, keep the cone component stationary, apply ultraviolet glue for curing, complete the position installation of the cone component of the laser module, that is, complete the adjustment of the optical power symmetry of the cone component of the laser module; The specific method of target feature extraction in step (5) is: for the R, G, and B color image channels after color space decomposition, first use the histogram equalization algorithm to enhance the R, G, and B color channels respectively. The histogram equalization algorithm is as follows: Assume that an image has L gray levels, and n k is the number of pixels with gray level k in the image f(x, y), and n is the total number of pixels in the image. Then the probability P of the k-th gray level occurring s (S k ) is as shown in Formulas (7) and (8): where K = 0, 1, 2, 3, … L-1 Formula (7) S k = K / (L - 1) Equation (8) With S k as the independent variable and P s (S k ) as the function, the resulting curve is the histogram of the image, and its information entropy is as shown in formula (9): The information entropy H of the entire image is shown in formulas (10) and (11): An image with a uniform distribution of histograms has the maximum information entropy value H. That is, when the formula (11) is satisfied, the formula (10) has the maximum information entropy value H max , making the histogram of the image approximately uniformly distributed. This process is the histogram equalization of the image, as shown in formula (12): where t k is the gray level of each pixel after histogram equalization. In actual calculation, t k needs to be rounded. After performing histogram equalization on the R, G, and B color channels, then select a color channel with the largest contrast between the target and the background in the image. For example, in the R color channel, the contrast between target 1 and the background is the largest, so select the R color channel; In the G channel, the contrast between target 2 and the background is the largest, so select the G color channel; in the B color channel, the contrast between target 3 and the background is the largest, so select the B color channel; Then, use the sobel operator in the selected color channel to extract the target contour in the image. The sobel edge detection operator performs a gray-scale weighted operation on the upper, lower, left, and right neighborhoods of the pixel points in the image with a 3×3 template. The sobel operator is shown in formulas (13) and (14): G x [i, j] = r[i - 1, j + 1] + 2 × f[i, j + 1] + f[i + 1, j + 1] Formula (13) -f[i - 1, j - 1] - 2×f[i, j - 1] - r[i + 1, j - 1] G y [i, j] = f[i + 1, j - 1] + 2 × f[i + 1, j] + f[i + 1, j + 1] Equation (14) -f[i - 1, j - 1] - 2×f[i - 1, j] - f[i - 1, j + 1] The convolution template is represented by formulas (15) and (16): Finally, the mathematical morphology method is used to process the target contour in the image extracted by the Sobel operator, removing the isolation outside the target, filling the small holes inside the contour, connecting adjacent objects, and smoothing the boundary; The formula for opening operation is as follows: represents the erosion operation, that is, the original image A is eroded by the structuring element S, represents that the original image A is eroded by the structuring element S and then dilated by the structuring element S; The formula for closing operation is: represents the dilation operation, that is, the original image A is dilated by the structuring element S. represents that the original image A is dilated by the structuring element S and then eroded by the structuring element S.

2. The method for adjusting the optical power symmetry of the cone component of the double-probe lit laser module according to claim 1, characterized in that: The laser module bottom feature acquisition system includes an AOI light source, a Medvision industrial camera MV-USB131GC, an industrial control computer, an independent 24V DC power supply, and automation industrial software; the 24V DC voltage provides power for the AOI light source, the industrial control computer, and the Medvision industrial camera, and the automation industrial software performs laser module bottom image acquisition, target feature extraction, angle compensation, and cone component installation position adjustment.

3. The method for adjusting the optical power symmetry of the cone component of the double-probe lit laser module according to claim 2, characterized in that: The specific method for designing the laser module cone component symmetry adjustment device in step (1) is as follows: Place 4 optical power acquisition components at the vertices of a 200mm×200mm square platform respectively. Each optical power acquisition component is standardized and customized with metal materials, and the surface and interior of each optical power acquisition component are blackened; A PCB board is installed behind the head of each optical power acquisition component. The PCB board is evenly divided into 6 vertical strips, each strip with a width of 4mm, and is labeled as positions 1, 2, 3, 4, 5, and 6 in sequence. Then, two optical power acquisition chips are installed at positions 1 and 6, and the remaining positions are left vacant; The 8 optical power acquisition chips inside the optical power acquisition components at the vertices of the 200mm×200mm square platform are labeled clockwise in sequence as: 1, 2, 3, 4, 5, 6, 7, 8; Connect the 8 optical power acquisition chips to the single-chip microcomputer in sequence according to the labels; Install a JAS-15 chuck at the center of the 200mm×200mm square platform, connect the chuck to the single-chip microcomputer, and power on the laser module; Transmit the optoelectronic data collected by the 8 optical power acquisition chips to the single-chip microcomputer for real-time data processing to obtain the real-time optical power data collected by each optical power acquisition chip.

4. The method for adjusting the optical power symmetry of the cone component of the double-probe lit laser module according to claim 3, characterized in that: The detailed process of the image processing method for the bottom of the laser module collected in step (4) is as follows: Use the Retinex theory to enhance the image of the bottom of the laser module collected. The Retinex theory assumes that the original image S(x, y) is the product of the illumination image L(x, y) and the reflectance image R(x, y), which is represented by formula (1): S(x, y) = R(x, y)·L(x, y) Formula (1) The purpose of Retinex-based image enhancement is to estimate the illumination L(x, y) from the original image S(x, y), so as to decompose R(x, y), eliminate the influence of uneven illumination, and improve the visual effect of the image. Usually, the image is transformed into the logarithmic domain, that is, s(x, y) = log S(x，y) , l(x, y) = log L(x，y) , r(x, y) = log R(x，y) , and then the product relationship is converted into a sum relationship. The core of the Retinex method is to estimate the illumination L(x, y), estimate the L(x, y) component from the image S(x, y), solve the reflection component r(x, y) from S(x, y) and / (x, y). The reflection component r(x, y) is an inherent property of the image. Then, it is transformed back to the real domain to obtain R(x, y), which is expressed by formulas (2)-(6) as follows: log S(x，y) = log R(x，y)·L(x，y) Formula (2) log S(x，y) = log R(x，y) + log L(x，y) Formula (3) s(x, y) = r(x, y) + l(x, y) Formula (4) r(x, y) = s(x, y) - l(x, y) Formula (5) R(x, y) = log r(x，y) Formula (6) After calculating R(x, y), perform R, G, and B three-color channel decomposition.

Images