Cleanliness detection method, apparatus, system, and storage medium

Abstract

The application provides a cleanliness detection method, device, system and storage medium, which are used for cleanliness detection of an optical glass assembly of a laser collimator mirror; the optical glass assembly comprises a plurality of end faces, and a pre-detected end face in the plurality of end faces is defined as a target end face; the cleanliness detection method comprises the following steps: acquiring an image of the target end face according to position data of the target end face; dividing the target end face image according to cleanliness division data, performing gray scale processing on the divided target end face image to obtain corresponding gray image division areas; and comparing a gray value change amplitude of each gray image division area with a gray value change amplitude threshold of each corresponding gray image division area to determine whether the cleanliness of the target end face is qualified. In the application, the cleanliness detection method is used for cleanliness detection of the optical glass assembly of the laser collimator mirror, and the detection efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of laser, in particular to a cleanliness detection method, device, system and storage medium. BACKGROUND

[0002] In the prior art, the dust and damage of the mirror surface, i.e. the end surface, of a laser collimator are observed by a high-power microscope to detect whether the laser collimator is qualified.

[0003] Since the entire detection process is completed by manual observation through a high-power microscope, the dust and damage of the end surface cannot be archived. Meanwhile, since the laser collimator has multiple end surfaces to be detected, the distance between the camera and the end surface needs to be adjusted to focus the camera and the end surface when different end surfaces are detected, but the focusing process is completed manually, which results in low detection efficiency. SUMMARY

[0004] Embodiments of the present application provide a cleanliness detection method, device, system and storage medium.

[0005] In a first aspect, embodiments of the present application provide a cleanliness detection method for detecting the cleanliness of an optical glass assembly of a laser collimator; the optical glass assembly includes multiple end surfaces, and a pre-detected end surface in the multiple end surfaces is defined as a target end surface; the cleanliness detection method includes:

[0006] obtaining an image of the target end surface according to position data of the target end surface;

[0007] partitioning the target end surface image according to cleanliness partition data, and performing grayscale processing on the partitioned target end surface image to obtain corresponding grayscale image partition regions; and

[0008] comparing the grayscale value change amplitudes of each grayscale image partition region with corresponding grayscale value change amplitude thresholds of each grayscale image partition region to determine whether the cleanliness of the target end surface is qualified.

[0009] In a second aspect, embodiments of the present application provide a cleanliness detection device for detecting the cleanliness of an optical glass assembly; the optical glass assembly includes multiple end surfaces, and a pre-detected end surface in the multiple end surfaces is defined as a target end surface; the cleanliness detection device includes:

[0010] an obtaining module configured to obtain an image of the target end surface according to position data of the target end surface;

[0011] a processing module configured to partition the target end surface image according to cleanliness partition data, and perform grayscale processing on the partitioned target end surface image to obtain corresponding grayscale image partition regions; and

[0012] The determining module is configured to compare the gray value variation range of each of the gray image sub-regions with a corresponding gray value variation range threshold of each of the gray image sub-regions, and determine whether the cleanliness of the target end surface is qualified.

[0013] In a third aspect, the embodiments of the present application provide a cleanliness detection system, which comprises:

[0014] a memory storing computer readable instructions;

[0015] a processor reading the computer readable instructions stored in the memory to execute the cleanliness detection method;

[0016] a visual camera connected to the processor, and the processor is configured to control the visual camera to take pictures and acquire target end surface images from the visual camera; and

[0017] a driving motor connected to the processor, and the processor is configured to control the driving motor to move to drive the visual camera to move, so that the visual camera is in focus with the target end surface.

[0018] In a fourth aspect, the embodiments of the present application further provide a storage medium having computer readable instructions stored thereon, and when the computer readable instructions are executed by a processor of a computer, the computer executes the cleanliness detection method.

[0019] In the embodiments of the present application, the laser collimator mirror can be fixed to the seat body, the camera assembly is driven to move by the running assembly, so that the visual camera is in focus with the target end surface; the image information of the target end surface can be acquired and saved by the visual camera; the visual camera can be quickly focused with the target end surface by the running assembly, and the detection efficiency is improved.

[0020] In the embodiments of the present application, the cleanliness detection method is used to detect the cleanliness of the optical glass assembly of the laser collimator mirror, and the detection efficiency is improved. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0022] Figure 1 The flowchart of the cleanliness detection method provided by the embodiments of the present application.

[0023] Figure 2 The application scenario schematic diagram of the cleanliness detection method provided by the embodiments of the present application.

[0024] Figure 3 FIG. 1 is a schematic diagram of a laser collimator according to an embodiment of the present application. Figure 2 FIG. 2 is a schematic diagram of a structure of a laser collimator according to an embodiment of the present application.

[0025] Figure 4 FIG. 3 is an enlarged view of a part A in FIG. 2. Figure 3

[0026] Figure 5 FIG. 4 is a schematic diagram of a cleaning degree zoning of a target end face of the laser collimator shown in FIG. 2. Figure 3

[0027] Figure 6 FIG. 5 is a schematic diagram of a cleaning degree zoning of an image of the target end face acquired by a visual camera shown in FIG. 2. Figure 2

[0028] FIG. 6 is a first schematic diagram of a cleaning degree detection device according to an embodiment of the present application. Figure 7

[0029] FIG. 7 is a second schematic diagram of a cleaning degree detection device according to an embodiment of the present application. Figure 8

[0030] FIG. 8 is a working flow chart of a cleaning degree detection system according to an embodiment of the present application. Figure 9 DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the present application.

[0032] In the description of the present application, it should be understood that the terms "first", "second", "third", "fourth" are only used for description purpose, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features.

[0033] Referring to FIG. 1, a laser collimator 40 according to an embodiment of the present application includes a laser 10, a laser collimator 20, and a target 30. Figures 1-4 The embodiments of the present application provide a cleaning degree detection method, which is used for cleaning degree detection of an optical glass assembly 420 of a laser collimator 40. The optical glass assembly 420 includes a plurality of end faces, and a pre-detected end face in the plurality of end faces is defined as a target end face 421.

[0034] It can be understood that, referring to FIG. 1, the laser collimator 40 according to an embodiment of the present application includes a laser 10, a laser collimator 20, and a target 30. Figures 3-4 ​​​The laser collimator 40 comprises a housing 440, which encloses a first accommodating space. The optical glass assembly 420 comprises a protective mirror 422 and an end cap 424. The protective mirror 422 and the end cap 424 are arranged in the first accommodating space in a spaced manner. The plurality of end faces comprises a first end face 4222, a second end face 4224, a third end face 4242 and a fourth end face 4244. The first end face 4222 is a light energy input end of the protective mirror 422, and the second end face 4224 is a light energy output end of the protective mirror 422. The third end face 4242 is a light energy input end of the end cap 424, and the fourth end face 4244 is a light energy output end of the end cap 424. The central axis of the first end face 4222, the central axis of the second end face 4224, the central axis of the third end face 4242 and the central axis of the fourth end face 4244 are collinear.

[0035] The target end face 421 can be any one of the first end face 4222, the second end face 4224, the third end face 4242 and the fourth end face 4244.

[0036] Referring to Figure 2 The cleanliness detection method provided by the embodiment of the present application is used to detect a plurality of laser collimators 40 arranged along the Z-axis direction. The camera assembly 620 is driven to move by the running assembly 320 capable of running along the X-axis direction and the Z-axis direction, so as to focus the vision camera 626 in the camera assembly 620 and the target end face 421.

[0037] The camera assembly 620 comprises, in sequence, an aperture 622, a high-power lens 624 and a vision camera 626; and is used to acquire image information of the target end face 421 after the vision camera 626 is focused on the target end face 421, so as to acquire cleanliness information and smoothness information of the target end face 421.

[0038] The running assembly 320 is used to drive the camera assembly 620 to move, so that the vision camera 626 is focused on the target end face 421. The running assembly 320 comprises a first running mechanism 322 and a second running mechanism 324. The first running mechanism 322 is connected with the camera assembly 620, and is used to drive the camera assembly 620 to make reciprocating motion along the X-axis direction. The second running mechanism 324 is connected with the first running mechanism 322, and is used to drive the first running mechanism 322 and the camera assembly 620 to make reciprocating motion along the Y-axis direction. The first running mechanism 322 is driven to move by a driving motor. The second running mechanism 324 is driven to move by a driving motor.

[0039] Referring to Figure 1 , Figure 2 , Figure 5 and Figure 6 The cleanliness detection method provided by the embodiment of the present application comprises:

[0040] 201、acquire the image of the target end face 421 according to the position data of the target end face 421;

[0041] 202、partition the target end face image 500 according to the cleanliness partition data, and perform grayscale processing on the partitioned target end face image 500 to obtain the corresponding gray image partition area; and

[0042] 203、compare the gray value change range of each gray image partition area with the gray value change range threshold of the corresponding gray image partition area, and determine whether the cleanliness of the target end face 421 is qualified.

[0043] For example, the cleanliness detection method further comprises, before acquiring the image of the target end face 421 according to the position data of the target end face 421, reading the position information of the target end face, the size information of the target end face, the cleanliness partition data of the target end face, and the cleanliness evaluation data of the target end face.

[0044] The cleanliness partition data includes boundary data of each cleanliness partition of the target end face 421 and cleanliness level data of each cleanliness partition; the target end face 421 is divided into multiple cleanliness partitions through the cleanliness partition data, and each cleanliness partition corresponds to a cleanliness level.

[0045] The cleanliness evaluation data includes the gray value change range threshold of the gray image partition area corresponding to each cleanliness partition and the defect pixel point number threshold.

[0046] For example, the step 201 of acquiring the image of the target end face 421 according to the position data of the target end face 421 comprises:

[0047] According to the position data, the driving motor is controlled to move to drive the visual camera to move, so that the visual camera is focused on the target end face 421;

[0048] After the visual camera is focused on the target end face 421, the visual camera is controlled to take a photo of the target end face 421 to acquire the target end face image 500; and

[0049] The target end face image 500 is stored.

[0050] For example, the step 202 of partitioning the target end face image 500 according to the cleanliness partition data comprises:

[0051] The target end face image area 540 in the target end face image 500 is recognized, and the size information of the target end face image area 540 is acquired; a scale is calculated according to the acquired size information of the target end face image area 540 and the read size information of the target end face 421; and

[0052] Based on the scale and the cleanliness zoning data, the target end face image region 540 is divided into at least one first-class region and at least one second-class region, wherein the cleanliness level of at least one first-class region is higher than that of at least one second-class region.

[0053] For example, step 202, which involves performing grayscale processing on the partitioned target end face image 500, includes:

[0054] By performing grayscale processing on the target image, at least one first-class region and at least one second-class region are transformed into corresponding grayscale image partition regions.

[0055] For example, please refer to Figure 5 The target end face 421 includes multiple test areas. The target end face 421 can be divided into multiple test areas based on cleanliness. These multiple test areas include a first test area 4212, a second test area 4214, a third test area 4216, and a fourth test area 4218. The cleanliness levels of the third test area 4216, the fourth test area 4218, the second test area 4214, and the first test area 4212 decrease sequentially.

[0056] The cleanliness classification data includes the boundary data of the first test area 4212, the second test area 4214, the third test area 4216, and the fourth test area 4218 of the target end face 421, as well as the cleanliness level data of the first test area 4212, the second test area 4214, the third test area 4216, and the fourth test area 4218.

[0057] Understandably, please refer to Figure 6 The target end face image 500 includes a background image region 520 and a target end face image region 540.

[0058] For example, please refer to Figures 5-6 Based on the scale and the cleanliness zoning data, the target end face image area 540 is divided into three regions: the third region 546, the fourth region 548, the second region 544, and the first region 542, arranged from high to low cleanliness levels.

[0059] Figure 6 The third region 546 and Figure 5 The third test region, 4216, corresponds to this. Figure 6 The fourth region 548 and Figure 5 The fourth test region, 4218, corresponds to this. Figure 6 The second region 544 and Figure 5 The second test region 4214 corresponds to this. Figure 6 The first region 542 and Figure 5The first to-be-tested region 4212 in the gray image corresponds to.

[0060] For example, referring to Figure 6 The fourth region 548 is a circular region, the third region 546 is an annular region surrounding the fourth region 548, the second region 544 is an annular region surrounding the third region 546, and the first region 542 is an annular region surrounding the second region 544. The third region 546 is a first type of region, the fourth region 548 and the second region 544 are both second type of regions, and the first region 542 is a third type of region. The cleanliness levels are arranged in descending order as the first type of region, the second type of region, and the third type of region.

[0061] For example, the comparison of the gray value variation range of each gray image partition region with the corresponding gray value variation range threshold in step 203 to determine whether the cleanliness of the target end face 421 is qualified includes:

[0062] According to the first determination step, it is determined whether the cleanliness of the target end face region corresponding to at least one first type of region is qualified.

[0063] When the cleanliness of the target end face region corresponding to at least one first type of region is unqualified, the cleanliness of the target end face 421 is determined to be unqualified.

[0064] When the cleanliness of the target end face region corresponding to at least one first type of region is qualified, according to the second determination step, it is determined whether the cleanliness of the target end face region corresponding to at least one second type of region is qualified. When the cleanliness of the target end face region corresponding to at least one second type of region is qualified, the cleanliness of the target end face 421 is determined to be qualified.

[0065] The first determination step is different from the second determination step.

[0066] For example, when the target end face image region 540 is partitioned, the gray image partition region is divided into the third region 546, the fourth region 548, the second region 544, and the first region 542 arranged in descending order of cleanliness level, and the third region 546 is a first type of region, the fourth region 548 and the second region 544 are second type of regions, and the first region 542 is a third type of region, the first type of region and the second type of region correspond to the middle region of the target end face 421, and the cleanliness needs to be determined to be qualified. The third type of region corresponds to the edge region of the target end face 421, and the cleanliness does not need to be determined to be qualified. According to the first determination step, it is determined whether the cleanliness of the target end face region corresponding to the third region 546 is qualified, and according to the second determination step, it is determined whether the cleanliness of the target end face region corresponding to the fourth region 548 and the second region 544 is qualified.

[0067] According to the first judging step, whether the cleanliness of the target end face region corresponding to the at least one first type region is qualified or not includes:

[0068] The absolute value of the gray scale difference between any two adjacent pixel points in the first type region is calculated, and the calculation result is compared with the gray scale difference threshold of the first type region:

[0069] If the absolute value of the gray scale difference between any two adjacent pixel points in the first type region is greater than the gray scale difference threshold of the first type region, the cleanliness of the target end face region corresponding to the at least one first type region is unqualified.

[0070] If the absolute value of the gray scale difference between any two adjacent pixel points in the first type region is less than or equal to the gray scale difference threshold of the first type region, the cleanliness of the target end face region corresponding to the at least one first type region is qualified.

[0071] According to the second judging step, whether the cleanliness of the target end face region corresponding to the at least one second type region is qualified or not includes:

[0072] The absolute value of the gray scale difference between any two adjacent pixel points in the second type region is calculated, and the adjacent two pixel points with the absolute value of the gray scale difference greater than the gray scale difference threshold of the second type region are marked as defect pixel points; the sum of the number of defect pixel points in the second type region is compared with the defect pixel point number threshold of the second type region:

[0073] If the sum of the number of defect pixel points in the second type region is greater than the defect pixel point number threshold of the second type region, the cleanliness of the target end face region corresponding to the at least one second type region is unqualified.

[0074] If the sum of the number of defect pixel points in the second type region is less than or equal to the defect pixel point number threshold of the second type region, the cleanliness of the target end face region corresponding to the at least one second type region is qualified.

[0075] Referring to Figure 7 The embodiment of the present application provides a cleanliness detection device 800 for detecting the cleanliness of an optical glass assembly; the optical glass assembly includes a plurality of end faces, and a pre-detected end face in the plurality of end faces is defined as a target end face 421. The cleanliness detection device 800 includes an acquisition module 820, a processing module 840 and a determination module 860.

[0076] The acquisition module 820 is used for acquiring an image of the target end face 421 according to position data.

[0077] The processing module 840 is used to partition the target end face image 500 according to the cleanliness zoning data, and to perform grayscale processing on the partitioned target end face image 500 to obtain the corresponding gray image partition areas.

[0078] The determination module 860 is used to compare the gray value change range of each gray image partition with the corresponding gray value change range threshold of each gray image partition to determine whether the cleanliness of the target end face 421 is qualified.

[0079] For example, please refer to Figure 8 The cleanliness detection device 800 also includes a reading module 810, which is used to read the position information of the target end face 421, the size information of the target end face 421, the cleanliness zoning data of the target end face 421, and the cleanliness evaluation data of the target end face 421.

[0080] For example, the acquisition module 820 includes a driving unit 822, a photographing unit 824, and a storage unit 826.

[0081] The drive unit 822 is used to control the movement of the drive motor according to the position data to drive the movement of the vision camera, so that the vision camera can focus on the target end face 421.

[0082] The imaging unit 824 is used to control the vision camera to take a picture of the target end face 421 and obtain the target end face image 500 after the vision camera focuses on the target end face 421.

[0083] Storage unit 826 is used to store target end face image 500.

[0084] For example, the processing module 840 includes a scale acquisition unit 842, a partition processing unit 844, and a preprocessing unit 846.

[0085] The scale acquisition unit 842 is used to identify the target end face image region 540 in the target end face image 500 and acquire the size information of the target end face image region 540; it is used to calculate the scale based on the acquired size information of the target end face image region 540 and the read size information of the target end face 421.

[0086] The partitioning processing unit 844 is used to partition the target end face image region 540 according to the scale and the cleanliness zoning data to obtain at least one first type region and at least one second type region, wherein the cleanliness level of at least one first type region is higher than that of at least one second type region.

[0087] The preprocessing unit 846 is used to perform grayscale processing on the target image, so that at least one first-class region and at least one second-class region are transformed into corresponding gray regions.

[0088] The partition processing unit 844 is configured to partition the target end face image region 540 according to the scale and the cleanliness zoning data, to obtain a third region 546, a fourth region 548, a second region 544 and a first region 542 arranged from high to low in cleanliness level. The fourth region 548 is a circular region, the third region 546 is an annular region surrounding the fourth region 548, the second region 544 is an annular region surrounding the third region 546, and the first region 542 is an annular region surrounding the second region 544.

[0089] The preprocessing unit 846 is configured to perform grayscale processing on the target image, so that the first region 542, the second region 544, the third region 546 and the fourth region 548 are converted into corresponding gray image partition regions.

[0090] The determination module 860 includes a first judgment unit 862 and a second judgment unit 864.

[0091] The first judgment unit 862 is configured to calculate the absolute value of the gray difference between any two adjacent pixel points in the first type region, and compare the calculation result with the gray difference threshold of the first type region.

[0092] If the absolute value of the gray difference between any two adjacent pixel points in the first type region is greater than the gray difference threshold of the first type region, the gray value variation amplitude of the first type region exceeds the gray value variation amplitude threshold of the first type region.

[0093] If the absolute value of the gray difference between any two adjacent pixel points in the first type region is less than or equal to the gray difference threshold of the first type region, the gray value variation amplitude of the first type region does not exceed the gray value variation amplitude threshold of the first type region.

[0094] The second judgment unit 864 is configured to calculate the absolute value of the gray difference between any two adjacent pixel points in the second type region, mark the adjacent two pixel points with the absolute value of the gray difference greater than the gray difference threshold of the second type region as defect pixel points, and compare the sum of the number of defect pixel points in the second type region with the defect pixel point number threshold of the second type region.

[0095] If the sum of the number of defect pixel points in the second type region is greater than the defect pixel point number threshold of the second type region, the cleanliness of the target end face region corresponding to at least one second type region is unqualified.

[0096] If the sum of the number of defect pixel points in the second type region is less than or equal to the defect pixel point number threshold of the second type region, the cleanliness of the target end face region corresponding to at least one second type region is qualified.

[0097] It can be understood that the embodiment of the application provides a cleanliness detection system, comprising a memory, a processor and a driving motor.

[0098] The memory stores computer readable instructions.

[0099] The processor reads the computer readable instructions stored in the memory to execute the cleanliness detection method; the visual camera is connected with the processor; the processor is used to control the visual camera to take pictures and is used to obtain the target end face image 500 from the visual camera.

[0100] The driving motor is connected with the processor; the processor is used to control the driving motor to move to drive the visual camera to move, so that the visual camera is in focus with the target end face 421.

[0101] It can be understood that the embodiment of the application provides a storage medium, which stores computer readable instructions, when the computer readable instructions are executed by the processor of the computer, the computer executes the cleanliness detection method.

[0102] It can be understood that please refer to Figure 9 The embodiment of the application also provides a laser collimator cleanliness detection method, comprising:

[0103] According to step 201, the image of the first end face of the laser collimator is obtained; according to steps 202 and 203, the image algorithm processing is performed on the image of the first end face to determine whether the cleanliness of the first end face is qualified;

[0104] When it is determined that the cleanliness of the first end face is unqualified, manual wiping is performed and then rechecked; if the rechecking is still unqualified after more than three times, it is determined that the laser collimator is an unqualified product;

[0105] When it is determined that the cleanliness of the first end face is qualified, according to step 201, the images of the second end face, the third end face and the fourth end face of the laser collimator are obtained; according to steps 202 and 203, the image algorithm processing is performed on the images of the second end face, the third end face and the fourth end face respectively to determine whether the cleanliness of the second end face, the third end face and the fourth end face is qualified;

[0106] When it is determined that at least one of the cleanliness of the second end face, the third end face and the fourth end face is unqualified, manual wiping is performed on the unqualified end face and then rechecked; if the rechecking is still unqualified after more than three times, it is determined that the laser collimator is an unqualified product;

[0107] When it is determined that the cleanliness of the second end face, the third end face and the fourth end face is qualified, the images of the first end face, the second end face, the third end face and the fourth end face are saved, and the laser collimator is qualified and put into storage.

[0108] The cleaning degree detection method, device, system and storage medium provided by the embodiments of the present application are described in detail above, and the principles and implementation manners of the present application are described by applying specific examples. The above description of the embodiments is only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges will be changed, and the above description of the present application should not be understood as a limitation.

Claims

1. A cleanliness detection method characterized by, The application discloses a cleanliness detection method for an optical glass assembly of a laser collimator, and relates to the technical field of optical glass assembly cleanliness detection. According to the position data of the target end face, an image of the target end face is acquired. A target end face image region in the target end face image is identified, and size information of the target end face image region is acquired; a scale is calculated according to the acquired size information of the target end face image region and read size information of the target end face; and According to the scale and cleanliness zoning data, the target end face image region is zoned to obtain at least one first type region and at least one second type region, wherein a cleanliness level of the at least one first type region is higher than that of the at least one second type region; The at least one first type region and the at least one second type region are converted into corresponding gray image zoning regions through gray processing of the target end face image, a gray value change range of each gray image zoning region is compared with a gray value change range threshold of the corresponding gray image zoning region, and whether the cleanliness of the target end face is qualified is judged; and According to the first judgment step, whether the cleanliness of a target end face region corresponding to the at least one first type region is qualified is judged. When the cleanliness of the target end face region corresponding to the at least one first type region is unqualified, the cleanliness of the target end face is judged as unqualified. When the cleanliness of the target end face region corresponding to the at least one first type region is qualified, whether the cleanliness of a target end face region corresponding to the at least one second type region is qualified is judged according to the second judgment step; when the cleanliness of the target end face region corresponding to the at least one second type region is qualified, the cleanliness of the target end face is judged as qualified; wherein the first judgment step is different from the second judgment step.

2. The cleanliness detection method according to claim 1, characterized by, According to the position data of the target end face, an image of the target end face is acquired. According to the position data, a driving motor is controlled to move to drive a visual camera to move, so that the visual camera is in focus with the target end face; After the visual camera is in focus with the target end face, the visual camera is controlled to take a photograph of the target end face to acquire a target end face image; and The target end face image is stored.

3. The cleanliness detection method according to claim 1, characterized by, According to the first judgment step, whether the cleanliness of a target end face region corresponding to the at least one first type region is qualified is judged, including: The absolute value of a gray difference of any two adjacent pixel points in the first type region is calculated, and the calculation result is compared with a gray difference threshold of the first type region: If the absolute value of the gray difference of the two adjacent pixel points in the first type region is greater than the gray difference threshold of the first type region, the cleanliness of the target end face region corresponding to the at least one first type region is unqualified. If the absolute value of the gray scale difference of any two adjacent pixels in the first type region is less than or equal to the gray scale difference threshold of the first type region, the cleanliness of the target end face region corresponding to the at least one first type region is qualified.

4. The cleanliness detection method according to claim 3, characterized in that, the judging whether the cleanliness of the target end face region corresponding to the at least one second type region is qualified according to the second judging step further comprises: calculating the absolute value of the gray scale difference of any two adjacent pixels in the second type region, marking the two adjacent pixels in the second type region whose absolute value of the gray scale difference is greater than the gray scale difference threshold of the second type region as defect pixels respectively; comparing the sum of the number of the defect pixels in the second type region with the defect pixel number threshold of the second type region: if the sum of the number of the defect pixels in the second type region is greater than the defect pixel number threshold of the second type region, the cleanliness of the target end face region corresponding to the at least one second type region is unqualified; if the sum of the number of the defect pixels in the second type region is less than or equal to the defect pixel number threshold of the second type region, the cleanliness of the target end face region corresponding to the at least one second type region is qualified.

5. A cleanliness detection device characterized by comprising: The application relates to a cleanliness detection device for detecting the cleanliness of an optical glass assembly; the optical glass assembly comprises a plurality of end faces, and a pre-detected end face in the plurality of end faces is defined as a target end face; the cleanliness detection device comprises: an acquisition module configured to acquire an image of the target end face according to position data of the target end face; a processing module configured to divide the target end face image according to cleanliness zoning data, and perform gray scale processing on the divided target end face image to obtain corresponding gray image zoning regions; and a determination module configured to compare the gray value variation range of each gray image zoning region with the gray value variation range threshold of each corresponding gray image zoning region, and judge whether the cleanliness of the target end face is qualified; wherein the processing module comprises: a scale acquisition unit configured to identify a target end face image region in the target end face image, and acquire size information of the target end face image region; and calculate a scale according to the acquired size information of the target end face image region and the read size information of the target end face; a zoning processing unit configured to divide the target end face image region according to the scale and the cleanliness zoning data, to obtain at least one first type region and at least one second type region, wherein the cleanliness level of the at least one first type region is higher than that of the at least one second type region; and a preprocessing unit configured to perform gray scale processing on the target end face image, so that the at least one first type region and the at least one second type region are converted into corresponding gray regions, the gray value variation range of each gray image zoning region is compared with the gray value variation range threshold of each corresponding gray image zoning region, and whether the cleanliness of the target end face is qualified is judged. The determining module is configured to determine whether the cleanliness of the target end face region corresponding to the at least one first type region is qualified according to a first judging step; when the cleanliness of the target end face region corresponding to the at least one first type region is unqualified, determine that the cleanliness of the target end face is unqualified; when the cleanliness of the target end face region corresponding to the at least one first type region is qualified, determine whether the cleanliness of the target end face region corresponding to the at least one second type region is qualified according to a second judging step; when the cleanliness of the target end face region corresponding to the at least one second type region is qualified, determine that the cleanliness of the target end face is qualified; and the first judging step is different from the second judging step.

6. The cleanliness detection apparatus according to claim 5, characterized by The obtaining module comprises: a driving unit configured to control a driving motor to move to drive a visual camera to move, so that the visual camera is in focus with the target end face; a photographing unit configured to control the visual camera to take a photograph of the target end face to obtain a target end face image after the visual camera is in focus with the target end face; and a storage unit configured to store the target end face image.

7. The cleanliness detection apparatus according to claim 5, characterized by The determining module comprises: a first judging unit configured to calculate the absolute value of the gray difference between any two adjacent pixel points in the first type region, and compare the calculation result with the gray difference threshold of the first type region; if the absolute value of the gray difference between any two adjacent pixel points in the first type region is greater than the gray difference threshold of the first type region, the gray value variation amplitude of the first type region exceeds the gray value variation amplitude threshold of the first type region, and the cleanliness of the target end face region corresponding to the at least one first type region is unqualified; if the absolute value of the gray difference between any two adjacent pixel points in the first type region is less than or equal to the gray difference threshold of the first type region, the gray value variation amplitude of the first type region does not exceed the gray value variation amplitude threshold of the first type region, and the cleanliness of the target end face region corresponding to the at least one first type region is qualified; and a second judging unit configured to calculate the absolute value of the gray difference between any two adjacent pixel points in the second type region, mark the adjacent two pixel points with the absolute value of the gray difference greater than the gray difference threshold of the second type region as defect pixel points in the second type region, and compare the sum of the number of the defect pixel points in the second type region with the defect pixel point number threshold of the second type region: if the sum of the number of the defect pixel points in the second type region is greater than the defect pixel point number threshold of the second type region, the cleanliness of the target end face region corresponding to the at least one second type region is unqualified; if the sum of the number of the defect pixel points in the second type region is less than or equal to the defect pixel point number threshold of the second type region, the cleanliness of the target end face region corresponding to the at least one second type region is qualified.

8. A cleanliness detection system characterized by, comprise: a memory storing computer readable instructions; a processor reading computer readable instructions stored in a memory to perform the method of any one of claims 1 to 4; a vision camera connected to the processor; the processor is configured to control the vision camera to take a picture and to obtain a target end face image from the vision camera; and a driving motor connected to the processor; the processor is configured to control the driving motor to move to drive the vision camera to move so that the vision camera is in focus with the target end face. a computer readable medium having stored thereon computer readable instructions that, when executed by a processor of a computer, cause the computer to perform the method of any one of claims 1 to 4.

9. A storage medium, characterized by ​

Images