WO2022104900A1 - 脏污图像检测方法、脏污图像检测装置及脏污图像检测机构 - Google Patents

脏污图像检测方法、脏污图像检测装置及脏污图像检测机构 Download PDF

Info

Publication number
WO2022104900A1
WO2022104900A1 PCT/CN2020/132652 CN2020132652W WO2022104900A1 WO 2022104900 A1 WO2022104900 A1 WO 2022104900A1 CN 2020132652 W CN2020132652 W CN 2020132652W WO 2022104900 A1 WO2022104900 A1 WO 2022104900A1
Authority
WO
WIPO (PCT)
Prior art keywords
dirt
area
unit
image
detection area
Prior art date
Application number
PCT/CN2020/132652
Other languages
English (en)
French (fr)
Inventor
罗涛
Original Assignee
诚瑞光学(深圳)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 诚瑞光学(深圳)有限公司 filed Critical 诚瑞光学(深圳)有限公司
Publication of WO2022104900A1 publication Critical patent/WO2022104900A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/90—Dynamic range modification of images or parts thereof
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image

Definitions

  • the present application relates to the technical field of contamination detection, and in particular, to a contamination image detection method, a contamination image detection device, and a contamination image detection mechanism.
  • Mobile phone lens refers to an optical element made of lens and used for shooting.
  • the surface of the mobile phone lens is always contaminated due to direct or indirect contact, which affects the shooting effect of the mobile phone. Therefore, the contamination detection of the lens is an essential process.
  • the detection of contamination mainly relies on manual identification under a high-power microscope, which requires a large workload and has a certain degree of subjective awareness, thereby affecting the stability of lens detection.
  • a dirty image detection method comprising
  • the presence or absence of dirt in the detection area is detected.
  • a dirty image detection device comprising:
  • the image acquisition unit is used to acquire an image of the lens to be detected
  • the first processing unit is used to determine the range of the detection area
  • the second processing unit is configured to determine the dirt particles within the detection area
  • a detection unit configured to detect whether there is dirt in the detection area.
  • the determining the dirty particles within the detection area includes: dividing the detection area into multiple local areas; capturing multiple pixel points in the local area; selecting the pixel points to be determined ; Compare the pixel point with other surrounding pixel points to obtain the difference in gray level between the pixel point and other surrounding pixel points; based on the difference value, determine whether the pixel point is the dirty particle ;
  • the detecting whether there is dirt in the detection area based on the dirt particles includes one, two or three of the following three methods: determining whether there is individual-type dirt in the detection area; Determining whether there is aggregated contamination in the detection area; determining whether there is regional contamination in the detection area;
  • the determining whether there is an individual type of dirt in the detection area includes: selecting the dirt particles to be determined; calculating the area of the dirt particles; and determining the dirt particles based on the area of the dirt particles. Whether the said dirt particles belong to the individual type of dirt;
  • the determining whether there is aggregation-type dirt in the detection area includes: selecting the dirt particles to be determined; defining an aggregation-type distance; connecting the dirt particles with other dirt particles within the aggregation-type distance. contamination particles to form a contamination particle set; calculating the area of the contamination particle set; determining whether the contamination particle set belongs to aggregated contamination based on the area of the contamination particle set;
  • the judging whether there is regional-type dirt in the detection area includes: defining a judging area; calculating the number and area of the dirt particles in the judging area; based on the number and area of the dirt particles , and determine whether the determination area belongs to the area-type contamination.
  • a dirty image detection device comprising:
  • the image acquisition unit is used to acquire an image of the lens to be detected
  • the first processing unit is used to determine the range of the detection area
  • the second processing unit is configured to determine the dirt particles within the detection area
  • a detection unit which is used to detect whether there is dirt in the detection area
  • the second processing unit includes: a dividing unit, which is used for dividing the detection area into a plurality of local areas; a capturing unit, which is used for capturing a plurality of pixels in the local area point; image selection unit, the image selection unit is used to select the pixel point to be determined; comparison unit, the comparison unit is used to compare the pixel point with other surrounding pixel points to obtain the pixel point and The difference between the gray levels of other surrounding pixels; a dirty particle acquisition unit, the dirty particle acquisition unit is configured to determine whether the pixel point is the dirty particle based on the difference value;
  • the detection unit includes one, two or three of the following three types of units: a first determination unit, which is used to determine whether there is contamination of an individual type in the detection area; a second determination unit unit, the second determination unit is used to determine whether there is aggregated contamination in the detection area; the third determination unit is used to determine whether there is area-type contamination in the detection area ;
  • the first determination unit includes: a first selection unit for selecting the dirt particles to be determined; a first calculation unit for calculating the dirt particles area; a first determination subunit, the first determination subunit is used to determine whether the dirt particles belong to individual-type dirt;
  • the second determination unit includes: a second selection unit for selecting the dirty particles to be determined; a first definition unit for defining an aggregation-type distance; a dirt particle connecting unit, the dirt particle connecting unit is used for connecting the dirt particles and other dirt particles within the aggregation-type distance to form a dirt particle set; a second computing unit, the second The calculation unit is used for calculating the area of the dirt particle set; the second judgment subunit is used for judging whether the dirt particle set belongs to aggregated dirt;
  • the third determination unit includes: a second definition unit, which is used to define a determination area; and a third calculation unit, which is used to calculate the amount of the dirt particles in the determination area. The number and area; a third determination subunit, the third determination subunit is used to determine whether the determination area belongs to area-type dirt.
  • the present application also provides a dirty image detection mechanism, which is used to realize the above dirty image detection method, and specifically includes a camera for collecting the image of the lens to be collected, and for driving the camera to move in a vertical direction.
  • the driving component and the light-emitting component for illuminating the lens to be collected.
  • the image of the lens is obtained first, and then the range of the detection area to be detected is determined according to the obtained image, and then the dirty particles in the detection area are obtained. Based on the dirt particles, the dirt in the detection area is judged.
  • FIG. 1 is a schematic flowchart of a dirty image detection method provided by a first embodiment of the present application.
  • FIG. 2 is a schematic partial flowchart of a dirty image detection method provided by a second embodiment of the present application.
  • FIG. 3 is a schematic flowchart of a dirty image detection method provided by a third embodiment of the present application.
  • FIG. 4 is a schematic partial flowchart of a dirty image detection method provided by a fourth embodiment of the present application.
  • FIG. 5 is a schematic partial flowchart of a dirty image detection method provided by a fifth embodiment of the present application.
  • FIG. 6 is a schematic flowchart of a dirty image detection method provided by a sixth embodiment of the present application.
  • FIG. 7 is a schematic partial flowchart of a dirty image detection method provided by a seventh embodiment of the present application.
  • FIG. 8 is a schematic partial flowchart of a dirty image detection method provided by an eighth embodiment of the present application.
  • FIG. 9 is a schematic partial flowchart of a dirty image detection method provided by an eighth embodiment of the present application.
  • FIG. 10 is a schematic structural diagram of a dirty image detection device provided by the present application.
  • FIG. 11 is a schematic structural diagram of a detection unit of the dirty image detection device as provided in FIG. 10 .
  • FIG. 12 is a schematic structural diagram of the first determination unit of the detection unit provided in FIG. 11 .
  • FIG. 13 is a schematic structural diagram of the second determination unit of the detection unit provided in FIG. 11 .
  • FIG. 14 is a schematic structural diagram of the third determination unit of the detection unit provided in FIG. 11 .
  • FIG. 15 is a schematic structural diagram of the second processing unit of the dirty image detection apparatus as provided in FIG. 10 .
  • FIG. 16 is a schematic structural diagram of a dirty image detection mechanism provided by the present application.
  • FIG. 1 shows a schematic flowchart of a dirty image detection method provided by a first embodiment of the present application.
  • the first embodiment of the present application will be described in detail below with reference to FIG. 1 .
  • the image contains the image of the lens to be tested, and also includes the image of the periphery of the lens to be tested.
  • S200 Determine the range of the detection area based on the image.
  • the main function of this step is to remove the image around the lens to be detected, so as to avoid the interference of the peripheral image to the contamination detection.
  • S300 Determine the dirt particles within the detection area.
  • Dirty particles refer to particles that may be judged as dirty.
  • S400 Determine the dirt in the detection area based on the dirt particles.
  • the above-mentioned dirt particles that may be dirt are judged by comparing the characteristics of dirt.
  • each different lens has different criteria for judging dirt.
  • the dirt particles on the lens can be detected stably and objectively without manual operation, which greatly improves the dirt detection efficiency.
  • FIG. 2 shows a partial schematic flowchart of a dirty image detection method provided by the second embodiment of the present application.
  • the second embodiment of the present application will be described in detail below with reference to FIG. 2 .
  • the center coordinates of the detection area are calculated by a visual algorithm.
  • the center coordinates of the detection area are calculated by a caliper algorithm.
  • the caliper algorithm first divides the detection area to form a uniform block, then takes points in each divided block, and finally obtains a circle or a straight line through circle fitting or straight line fitting, and finally obtains the detection area. Center coordinates.
  • S220 Define the radius of the detection area.
  • the radius of the detection area is defined according to the model of the lens.
  • FIG. 3 shows a schematic flowchart of a dirty image detection method provided by a third embodiment of the present application.
  • the third embodiment of the present application will be described in detail below with reference to FIG. 3 .
  • step S250 is also included.
  • Bright spots are spots formed by light hitting the lens.
  • the bright spot exists in the image of the lens, and is easily judged to be dirty, thereby affecting the judgment result of soiling. Therefore, removing bright spots that affect the contamination determination result can ensure the accuracy of the contamination determination result.
  • FIG. 4 shows a schematic flowchart of part of the dirty image detection method provided by the fourth embodiment of the present application.
  • the fourth embodiment of the present application will be described in detail below with reference to FIG. 4 .
  • Step S250 specifically further includes the following steps.
  • S251 Define the maximum radius of the long side of the circumscribed ellipse for dirt.
  • S252 Remove the bright spots whose maximum radius of the short side of the circumscribed ellipse is greater than the radius of the long side of the maximum circumscribed ellipse of dirt.
  • the maximum radius of the long side of the circumscribed ellipse of the dirt has been determined, that is, the bright spot larger than the radius of the long side of the maximum circumscribed ellipse of the dirt cannot be dirty.
  • FIG. 5 shows a schematic flowchart of part of the dirty image detection method provided by the fifth embodiment of the present application.
  • the fifth embodiment of the present application will be described in detail below with reference to FIG. 5 .
  • Step S300 specifically further includes the following steps.
  • the detection area is composed of multiple local areas.
  • S320 Capture multiple pixel points in the local area.
  • a pixel is composed of small squares of an image. These small squares have a clear position and assigned color value. The color and position of the small squares determine the appearance of the image.
  • a digital image is an image with only one sampled color per pixel, and such images are usually displayed in grayscale ranging from the darkest black to the brightest white.
  • grayscale the relationship between white and black is divided into several levels according to the logarithmic relationship, which is called "gray level”. The range is generally from 0 to 255, white is 255, black is 0, so black and white images are also called grayscale images.
  • Steps S330 and S340 are repeated multiple times to compare and determine all the pixel points.
  • a dynamic threshold method may be used to assist in completing step S300.
  • the dynamic threshold method refers to the method of dividing the image into blocks according to the left side and selecting a threshold for each block.
  • FIG. 6 shows a schematic flowchart of the dirty image detection method provided by the sixth embodiment of the present application.
  • the sixth embodiment of the present application will be described in detail below with reference to FIG. 6 .
  • Step S400 includes any one or more of the following three methods.
  • S410 Determine whether there is contamination of the individual type in the detection area.
  • this method is mainly used to determine dirty particles with a large area.
  • S420 Determine whether or not there is aggregated contamination in the detection area.
  • this method is mainly used to determine a plurality of dirt particles that are relatively close together or relatively concentrated.
  • S430 Determine whether there is area-type contamination in the detection area.
  • this method is mainly used to determine a plurality of relatively dispersed dirt particles.
  • FIG. 7 shows a schematic flowchart of part of a dirty image detection method provided by the seventh embodiment of the present application.
  • the seventh embodiment of the present application will be described in detail below with reference to FIG. 7 .
  • Step S410 specifically includes the following steps.
  • S411 Select the dirty particles to be determined.
  • step S230 the range of the detection area and the coordinates of each point in the detection area are obtained.
  • the area of the dirty particles can be obtained according to the coordinates of the dirty particles.
  • S413 Based on the area of the dirt particles, determine whether the dirt particles belong to individual types of dirt.
  • the data of the contamination is collected.
  • FIG. 8 shows a schematic flowchart of part of the dirty image detection method provided by the eighth embodiment of the present application.
  • the eighth embodiment of the present application will be described in detail below with reference to FIG. 8 .
  • Step S420 specifically includes the following steps.
  • the size of the converging distance is determined by the model of the lens.
  • step S230 the range of the detection area and the coordinates of each point in the detection area are obtained.
  • the area of the set of dirty particles can be obtained according to the coordinates of the set of dirty particles.
  • S425 Based on the area of the contamination particle set, it is determined whether the contamination particle set belongs to aggregated contamination.
  • a closed operation may be used to assist in completing step S420.
  • the closing operation is defined as dilation followed by erosion.
  • FIG. 9 shows a schematic flowchart of part of the dirty image detection method provided by the ninth embodiment of the present application.
  • the ninth embodiment of the present application will be described in detail below with reference to FIG. 9 .
  • Step S430 specifically includes the following steps.
  • the detection area is divided into a plurality of judgment areas.
  • S432 Calculate the number and area of dirt particles in the determination area.
  • step S230 the range of the detection area and the coordinates of each point in the detection area are obtained. In this way, the total area of the dirty particles can be obtained according to the coordinates of each dirty particle.
  • S433 Based on the number and area of the dirt particles, it is determined whether the determination area belongs to the area-type dirt.
  • the data of the contamination is collected.
  • a dirty image detection device 1000 including:
  • the image acquisition unit 1100 is used to acquire an image of the lens to be detected
  • the first processing unit 1200 is used to determine the range of the detection area
  • the second processing unit 1300 the second processing unit 1300 is used to determine the dirt particles within the detection area
  • the detection unit 1400 is used to detect whether there is dirt in the detection area.
  • the second processing unit 1300 specifically includes the following units.
  • the dividing unit 1310 is used to divide the detection area into a plurality of partial areas
  • the capturing unit 1320 is used to capture a plurality of pixel points in the local area;
  • Image selection unit 1330 the image selection unit 1330 is used to select the pixel to be determined
  • a comparison unit 1340 the comparison unit 1340 is used to compare the pixel point with other surrounding pixel points, so as to obtain the difference in gray level between the pixel point and other surrounding pixel points;
  • the dirty particle acquiring unit 1350 is configured to determine whether the pixel point is a dirty particle based on the difference value.
  • the detection unit 1400 includes one, two or three of the following three units:
  • the first determination unit 1410 the first determination unit 1410 is used to determine whether there is contamination of the individual type in the detection area;
  • the second determination unit 1420 is configured to determine whether there is aggregated contamination in the detection area
  • the third determination unit 1430 is used to determine whether there is regional-type contamination in the detection area.
  • the first determination unit 1410 includes:
  • the first selection unit 1411 the first selection unit 1411 is used to select the dirty particles to be determined
  • the first calculation unit 1412 is used to calculate the area of the dirt particles
  • the first determination subunit 1413 is used to determine whether the dirt particles belong to individual-type dirt.
  • the second determination unit 1420 includes:
  • the second selection unit 1421 the second selection unit 1421 is used to select the dirty particles to be determined
  • the first defining unit 1422 is used to define an aggregated distance
  • Dirty particle connecting unit 1423 which is used to connect the dirty particle and other dirty particles within the aggregation-type distance to form a dirty particle set
  • the second calculation unit 1424 is used to calculate the area of the dirty particle set
  • the second determination subunit 1425 is used to determine whether the set of dirt particles belongs to aggregated dirt.
  • the third determination unit 1430 includes:
  • the second definition unit 1431, the second definition unit 1431 is used to define the determination area
  • the third calculation unit 1432 the third calculation unit 1432 is used to calculate the number and area of the dirt particles in the determination area;
  • the third determination sub-unit 1433 is used to determine whether the determination area belongs to the area-type contamination.
  • the present application also provides a dirty image detection mechanism, which specifically includes a camera 10 for capturing an image of a lens to be captured, a driving assembly 20 for driving the camera 10 to move in a vertical direction, and a The light-emitting component 30 of the lens to be collected is illuminated.
  • the drive assembly 20 includes a power element 21 , a lead screw (not shown in the figure), a moving nut (not shown in the figure), a guide rail 22 and a slider 23 .
  • the output shaft of the power element 21 is connected with the screw rod, the moving nut is sleeved and engaged with the screw rod, and the moving nut is fixedly connected with the slider 23, the slider 23 is slidably arranged on the guide rail 22, and the camera 10 It is mounted on the slider 23 by a mounting member 40 . In this way, the power element 21 can drive the camera 10 to move longitudinally.
  • the power element 21 rotates the servo motor to increase the precision and stability of the lens point movement.
  • the camera 10 selects a bi-telecentric lens with low depth of field and high resolution to ensure that the dirt can be clearly imaged. Further, the camera 10 adopts a CMOS global exposure camera 10 to ensure fast and stable image acquisition.
  • the light-emitting assembly 30 adopts a combination of a green light and a condenser lens, and the installation angle is at a 45-degree angle to the lens axis, and illuminates from bottom to top, so that the surface of the finished lens to be detected has multiple light spots; these light spots start from the center, and the brightness is higher. Slowly darkens, so that the surface of the finished lens to be inspected presents uneven light distribution; in this way, slight dirt can be better presented.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Quality & Reliability (AREA)
  • Studio Devices (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

脏污图像检测方法、脏污图像检测装置及脏污图像检测机构，其中采集镜片脏污方法包括获取待检测镜片的影像（S100），基于影像确定检测区域的范围（S200），确定检测区域范围内的脏污颗粒（S300），基于脏污颗粒判定检测区域内的脏污（S400）；脏污图像检测装置（1000）包括影像获取单元（1100）、第一处理单元（1200）、第二处理单元（1300）以及检测单元（1400）。通过上述提供的脏污图像检测方法及脏污图像检测装置，可稳定且客观地检测出镜片上的脏污颗粒，无需人工操作，极大程度提高了脏污的检测效率。

Description

脏污图像检测方法、脏污图像检测装置及脏污图像检测机构 技术领域
本申请涉及脏污检测技术领域，尤其涉及脏污图像检测方法、脏污图像检测装置及脏污图像检测机构。
 背景技术
手机镜片，指的是一种由透镜制成且用于拍摄的光学元件。在镜片生产过程中，总会因直接接触或间接接触导致手机镜片表面存在脏污，以至于影响到手机的拍摄效果。因此对于镜片的脏污检测是一道必不可少的工序。
 技术问题
现有技术中，对于脏污的检测主要依赖于人工在高倍显微镜下识别，其工作量较大，且存在一定的主观意识，从而影响镜片检测的稳定性。
 技术解决方案
基于此，有必要提供一种可解决镜片检测稳定性差的问题的脏污图像检测方法。
一种脏污图像检测方法，包括
获取待检测镜片的影像；
基于所述影像，确定检测区域的范围；
确定所述检测区域范围内的脏污颗粒；
基于所述脏污颗粒，检测所述检测区域内是否存在的脏污。
为了解决上述问题，本申请还提供一种脏污图像检测装置，包括：
影像获取单元，所述影像获取单元用于获取待检测镜片的影像；
第一处理单元，所述第一处理单元用于确定检测区域的范围；
第二处理单元，所述第二处理单元用于确定所述检测区域范围内的脏污颗粒；
检测单元，所述检测单元用于检测所述检测区域内是否存在脏污。
其中，所述确定所述检测区域范围内的脏污颗粒，包括：将所述检测区域划分为多个局部区域；捕捉所述局部区域内的多个像素点；选择待判定的所述像素点；对比所述像素点与周边其它像素点，以得出所述像素点与周边其它像素点之间灰度的差值；基于所述差值，判定所述像素点是否为所述脏污颗粒；
所述基于所述脏污颗粒，检测所述检测区域内是否存在脏污，包括以下三种方法中的一种、两种或三种：判定所述检测区域内是否存在个体型的脏污；判定所述检测区域内是否存在聚集型的脏污；判定所述检测区域内是否存在区域型的脏污；
其中，所述判定所述检测区域内是否存在个体型的脏污，包括：选择待判定的所述脏污颗粒；计算所述脏污颗粒的面积；基于所述脏污颗粒的面积，判定所述脏污颗粒是否属于个体型的脏污；
所述判定所述检测区域内是否存在聚集型的脏污，包括：选择待判定的所述脏污颗粒；定义聚集型距离；连接所述脏污颗粒与其在所述聚集型距离内的其它脏污颗粒，以形成脏污颗粒集；计算所述脏污颗粒集的面积；基于所述脏污颗粒集的面积，判定所述脏污颗粒集是否属于聚集型的脏污；
所述判定所述检测区域内是否存在区域型的脏污，包括：定义判定区域；计算所述判定区域内所述脏污颗粒的个数以及面积；基于所述脏污颗粒的个数以及面积，判定所述判定区域是否属于区域型的脏污。
为了解决上述问题，本申请还提供一种脏污图像检测装置，包括：
影像获取单元，所述影像获取单元用于获取待检测镜片的影像；
第一处理单元，所述第一处理单元用于确定检测区域的范围；
第二处理单元，所述第二处理单元用于确定所述检测区域范围内的脏污颗粒；
检测单元，所述检测单元用于检测所述检测区域内是否存在脏污；
其中，所述第二处理单元包括：划分单元，所述划分单元用于将所述检测区域划分为多个局部区域；捕捉单元，所述捕捉单元用于捕捉所述局部区域内的多个像素点；选像单元，所述选像单元用于选择待判定的所述像素点；对比单元，所述对比单元用于对比所述像素点与周边其它像素点，以得出所述像素点与周边其它像素点之间灰度的差值；脏污颗粒获取单元，所述脏污颗粒获取单元用于基于所述差值，判定所述像素点是否为所述脏污颗粒；
所述检测单元包括以下三种单元中的一种、两种或三种：第一判定单元，所述第一判定单元用于判定所述检测区域内是否存在个体型的脏污；第二判定单元，所述第二判定单元用于判定所述检测区域内是否存在聚集型的脏污；第三判定单元，所述第三判定单元用于判定所述检测区域内是否存在区域型的脏污；
所述第一判定单元包括：第一选择单元，所述第一选择单元用于选择待判定的所述脏污颗粒；第一计算单元，所述第一计算单元用于计算所述脏污颗粒的面积；第一判定子单元，所述第一判定子单元用于判定所述脏污颗粒是否属于个体型脏污；
所述第二判定单元包括：第二选择单元，所述第二选择单元用于选择待判定的所述脏污颗粒；第一定义单元，所述第一定义单元用于定义聚集型距离；脏污颗粒连接单元，所述脏污颗粒连接单元用于连接所述脏污颗粒与其在所述聚集型距离内的其它脏污颗粒，以形成脏污颗粒集；第二计算单元，所述第二计算单元用于计算所述脏污颗粒集的面积；第二判定子单元，所述第二判定子单元用于判定所述脏污颗粒集是否属于聚集型的脏污；
所述第三判定单元包括：第二定义单元，所述第二定义单元用于定义判定区域；第三计算单元，所述第三计算单元用于计算所述判定区域内所述脏污颗粒的个数以及面积；第三判定子单元，所述第三判定子单元用于判定所述判定区域是否属于区域型的脏污。
为了解决上述问题，本申请还提供一种脏污图像检测机构，用于实现上述脏污图像检测方法，具体包括用于采集待采集镜片影像的相机、用于驱动所述相机沿竖直方向移动的驱动组件以及用于照射待采集镜片的发光组件。
 有益效果
采用上述提供的脏污图像检测方法及脏污图像检测装置，先获取镜片的影像，再根据所获得的影像确定需要检测的检测区域的范围，然后得出检测区域范围内的脏污颗粒，最后根据脏污颗粒，判断检测区域内的脏污。通过这种方法，可稳定且客观地检测出镜片上的脏污颗粒，无需人工操作，极大程度提高了脏污的检测效率。
 附图说明
图1为本申请第一实施例提供的脏污图像检测方法的流程示意图。
图2为本申请第二实施例提供的脏污图像检测方法的部分流程示意图。
图3为本申请第三实施例提供的脏污图像检测方法的流程示意图。
图4为本申请第四实施例提供的脏污图像检测方法的部分流程示意图。
图5为本申请第五实施例提供的脏污图像检测方法的部分流程示意图。
图6为本申请第六实施例提供的脏污图像检测方法的流程示意图。
图7为本申请第七实施例提供的脏污图像检测方法的部分流程示意图。
图8为本申请第八实施例提供的脏污图像检测方法的部分流程示意图。
图9为本申请第八实施例提供的脏污图像检测方法的部分流程示意图。
图10为本申请提供的脏污图像检测装置的结构示意图。
图11为如图10提供的脏污图像检测装置的检测单元的结构示意图。
图12为如图11提供的检测单元的第一判定单元结构示意图。
图13为如图11提供的检测单元的第二判定单元结构示意图。
图14为如图11提供的检测单元的第三判定单元结构示意图。
图15为如图10提供的脏污图像检测装置的第二处理单元的结构示意图。
图16为本申请提供的脏污图像检测机构的结构示意图。
 本发明的实施方式
下面结合附图和实施方式对本申请作进一步说明。
图1示出了本申请第一实施例提供的脏污图像检测方法的流程示意图，以下参考图1对本申请第一实施例进行详细的阐述。
S100：获取待检测镜片的影像。
可以理解的是，该影像内含有待检测镜片的影像，还包括待检测镜片周边的影像。
S200：基于影像，确定检测区域的范围。
该步骤的主要作用是用于去除待检测镜片周边的影像，以避免周边影像对于脏污检测的干扰。
S300：确定检测区域范围内的脏污颗粒。
脏污颗粒，指的是可能判断为脏污的颗粒。
S400：基于脏污颗粒，判定检测区域内的脏污。
对比脏污的特性，对上述可能为脏污的脏污颗粒进行判定。
需要补充的是，每款不同的镜片，其脏污的判定标准不同。
采用上述提供的脏污图像检测方法，先获取镜片的影像，再根据所获得的影像确定需要检测的检测区域的范围，然后得出检测区域范围内的脏污颗粒，最后根据脏污颗粒，判断检测区域内的脏污。通过这种方法，可稳定且客观地检测出镜片上的脏污颗粒，无需人工操作，极大程度提高了脏污的检测效率。
基于本申请第一实施例，本申请提出第二实施例。图2示出了本申请第二实施例提供的脏污图像检测方法的部分流程示意图，以下参考图2对本申请第二实施例进行详细的阐述。
S210：计算检测区域的中心坐标。
需要说明的是，在该步骤中，通过视觉算法来计算检测区域的中心坐标。可选地，通过卡尺算法来计算检测区域的中心坐标。
卡尺算法，首先对检测区域进行分割，形成均匀的区块，然后在每个划分的区块内取点，最后通过圆拟合或者直线拟合的方式，得到圆或直线，最终得到检测区域的中心坐标。
S220：定义检测区域的半径。
可以理解的是，不同型号的产品，其镜片的面积均不相同。因此，在本步骤中，根据镜片的型号来定义检测区域的半径。
S230：基于中心坐标及半径，计算检测区域的范围。
可以理解的是，已知检测区域的中心坐标以及检测区域，即可推算出该检测区域的面积以及检测区域内每个点的坐标。
基于本申请第一实施例，本申请提出第三实施例。图3示出了本申请第三实施例提供的脏污图像检测方法的流程示意图，以下参考图3对本申请第三实施例进行详细的阐述。
在步骤S200之后，还包括步骤S250。
S250：去除影响脏污判定结果的亮斑。
亮斑，指的是因光线照射在镜片上形成的斑点。该亮斑存在镜片的影像中，且易被判定为脏污，从而对脏污的判定结果产生影响。因此，去除影响脏污判定结果的亮斑，可保证脏污判定结果的准确性。
基于本申请第三实施例，本申请提出第四实施例。图4示出了本申请第四实施例提供的脏污图像检测方法的部分流程示意图，以下参考图4对本申请第四实施例进行详细的阐述。
步骤S250具体还包括如下步骤。
S251：定义脏污的最大外接椭圆长边半径。
可以理解的是，不同型号的镜片，其面积均不相同。面积相对较大的镜片，判定为脏污的标准已不相同。
S252：去除最大外接椭圆短边半径大于脏污最大外接椭圆长边半径的亮斑。
可以理解的是，脏污的最大外接椭圆长边半径已经确定，即大于脏污最大外接椭圆长边半径的亮斑不可能为脏污。
基于本申请第一实施例，本申请提出第五实施例。图5示出了本申请第五实施例提供的脏污图像检测方法的部分流程示意图，以下参考图5对本申请第五实施例进行详细的阐述。
步骤S300具体还包括如下步骤。
S310：将检测区域划分为多个局部区域。
可以理解的是，检测区域是由多个局部区域组成。
S320：捕捉局部区域内的多个像素点。
需要说明的是，像素是指由图像的小方格组成的，这些小方块都有一个明确的位置和被分配的色彩数值，小方格颜色和位置就决定该图像所呈现出来的样子。
S330：选择待判定的像素点。
S340：对比像素点与周边其它像素点，以得出像素点与周边其它像素点之间灰度的差值。
需要说明的是，数字图像是每个像素只有一个采样颜色的图像，这类图像通常显示为从最暗黑色到最亮的白色的灰度。具体地，把白色与黑色之间按对数关系分成若干级，称为“灰度等级”。范围一般从0到255，白色为255，黑色为0，故黑白图片也称灰度图像。
S350：基于差值，判定像素点是否为脏污颗粒。
需要补充的是，在判定某一像素点是否为脏污颗粒时，都需要用对比该点与周边其它像素点灰度的差值。重复多次步骤S330以及S340，以对比确定所有的像素点。
另外，可使用动态阈值法辅助完成步骤S300。动态阈值法，指的是将图像按左边分块，并对每一块分别选一阈值进行分割的方法。
基于本申请第一实施例，本申请提出第六实施例。图6示出了本申请第六实施例提供的脏污图像检测方法的流程示意图，以下参考图6对本申请第六实施例进行详细的阐述。
步骤S400包括以下三种方法的中的任意一种或多种。
S410：判定检测区域内是否存在个体型的脏污。
需要说明的是，该方法主要用于判定面积较大的脏污颗粒。
S420：判定检测区域内是否存在聚集型的脏污。
需要说明的是，该方法主要用于判定相聚较近或较为集中的多个脏污颗粒。
S430：判定检测区域内是否存在区域型的脏污。
需要说明的是，该方法主要用于判定较为分散的多个脏污颗粒。
基于本申请第六实施例，本申请提出第七实施例。图7示出了本申请第七实施例提供的脏污图像检测方法的部分流程示意图，以下参考图7对本申请第七实施例进行详细的阐述。
步骤S410具体包括如下步骤。
S411：选择待判定的脏污颗粒。
S412：计算脏污颗粒的面积。
可以理解的是，在步骤S230中以得知检测区域的范围以及检测区域内各点的坐标。如此，可根据脏污颗粒的坐标，得出脏污颗粒的面积。
S413：基于脏污颗粒的面积，判定脏污颗粒是否属于个体型的脏污。
可以理解的是，不同型号的手机镜片，其判定脏污的标准不相同。
若判断结果为个体型的脏污，便收集该脏污的数据。
基于本申请第六实施例，本申请提出第八实施例。图8示出了本申请第八实施例提供的脏污图像检测方法的部分流程示意图，以下参考图8对本申请第八实施例进行详细的阐述。
步骤S420具体包括如下步骤。
S421：选择待判定的脏污颗粒。
S422：定义聚集型距离。
可以理解的是，该聚集型距离的大小由镜片的型号决定。
S423：连接脏污颗粒与其聚集型距离内的其它脏污颗粒，以形成脏污颗粒集。
可以理解的是，多个脏污颗粒聚集在某区域内，其效果相当于一个较大的脏污颗粒。该步骤用于连接距离较近的脏污颗粒，从而形成一个较大的脏污颗粒。
S424：计算脏污颗粒集的面积。
可以理解的是，在步骤S230中以得知检测区域的范围以及检测区域内各点的坐标。如此，可根据脏污颗粒集的坐标，得出脏污颗粒集的面积。
S425：基于脏污颗粒集的面积，判定脏污颗粒集是否属于聚集型的脏污。
若判断结果为聚集型的脏污，便收集该脏污的数据。
另外，可采用闭运算来辅助完成步骤S420。在数学形态学中，闭运算被定义为先膨胀后腐蚀。
基于本申请第六实施例，本申请提出第九实施例。图9示出了本申请第九实施例提供的脏污图像检测方法的部分流程示意图，以下参考图9对本申请第九实施例进行详细的阐述。
步骤S430具体包括如下步骤。
S431：定义判定区域。
将检测区域分割为多个判定区域。
S432：计算判定区域内脏污颗粒的个数以及面积。
可以理解的是，在步骤S230中以得知检测区域的范围以及检测区域内各点的坐标。如此，可根据每一个脏污颗粒的坐标，得出脏污颗粒的总面积。
S433：基于脏污颗粒的个数以及面积，判定判定区域是否属于区域型的脏污。
若判断结果为区域型的脏污，便收集该脏污的数据。
另外，请参阅图10-图15，本申请还提供一种脏污图像检测装置1000，包括：
影像获取单元1100，影像获取单元1100用于获取待检测镜片的影像；
第一处理单元1200，第一处理单元1200用于确定检测区域的范围；
第二处理单元1300，第二处理单元1300用于确定检测区域范围内的脏污颗粒；
检测单元1400，检测单元1400用于检测检测区域内是否存在脏污。
具体地，第二处理单元1300具体包括如下单元。
划分单元1310，划分单元1310用于将检测区域划分为多个局部区域；
捕捉单元1320，捕捉单元1320用于捕捉局部区域内的多个像素点；
选像单元1330，选像单元1330用于选择待判定的像素点；
对比单元1340，对比单元1340用于对比像素点与周边其它像素点，以得出像素点与周边其它像素点之间灰度的差值；
脏污颗粒获取单元1350，脏污颗粒获取单元1350用于基于差值，判定像素点是否为脏污颗粒。
具体地，检测单元1400包括以下三种单元中的一种、两种或三种：
第一判定单元1410，第一判定单元1410用于判定检测区域内是否存在个体型的脏污；
第二判定单元1420，第二判定单元1420用于判定检测区域内是否存在聚集型的脏污；
第三判定单元1430，第三判定单元1430用于判定检测区域内是否存在区域型的脏污。
其中，第一判定单元1410包括：
第一选择单元1411，第一选择单元1411用于选择待判定的脏污颗粒；
第一计算单元1412，第一计算单元1412用于计算脏污颗粒的面积；
第一判定子单元1413，第一判定子单元1413用于判定脏污颗粒是否属于个体型脏污。
第二判定单元1420包括：
第二选择单元1421，第二选择单元1421用于选择待判定的脏污颗粒；
第一定义单元1422，第一定义单元1422用于定义聚集型距离；
脏污颗粒连接单元1423，脏污颗粒连接单元1423用于连接脏污颗粒与其在聚集型距离内的其它脏污颗粒，以形成脏污颗粒集；
第二计算单元1424，第二计算单元1424用于计算脏污颗粒集的面积；
第二判定子单元1425，第二判定子单元1425用于判定脏污颗粒集是否属于聚集型的脏污。
第三判定单元1430包括：
第二定义单元1431，第二定义单元1431用于定义判定区域；
第三计算单元1432，第三计算单元1432用于计算判定区域内脏污颗粒的个数以及面积；
第三判定子单元1433，第三判定子单元1433用于判定判定区域是否属于区域型的脏污。
需要说明的是，通过前述的脏污图像检测方法，本领域技术人员可清 楚了解到本实施例提供的脏污图像检测装置1000，为了方便和简洁，上述单元 以及装置的具体工作过程，可参考前述方法实施例中对应的过程，在此不再赘述。
另外，请参阅图16，本申请还提供一种脏污图像检测机构，具体包括包括用于采集待采集镜片影像的相机10、用于驱动相机10沿竖直方向移动的驱动组件20以及用于照射待采集镜片的发光组件30。
该驱动组件20包括动力元件21、丝杆（图中未示出）、移动螺母（图中未示出）、导轨22以及滑块23。该动力元件21的输出轴与丝杆连接，移动螺母套设且啮合在丝杆上，且该移动螺母与滑块23固定连接，该滑块23可滑动设置在导轨22上，再者相机10通过一安装件40安装在滑块23上。如此，可使动力元件21驱动相机10纵向运动。
优选地，动力元件21旋转伺服电机，以增加镜头点位移动的精度及稳定性。
可选地，该相机10选用低景深高分辨率的双远心镜头，以保证脏污能够成像清晰。进一步地，该相机10采用CMOS全局曝光相机10，保证既快速又稳定的获得影像。
进一步地，该发光组件30采用绿光灯加聚光镜组合，安装角度与镜头轴线成45度角，从下往上照射，满足待检测的镜头成品表面有多个光斑；这些光斑从中心开始，亮度慢慢变暗，使得待检测的镜头成品表面呈现不均匀的光照分布；这样，能够使得轻微脏污更好的呈现出来。
以上所述的仅是本申请的实施方式，在此应当指出，对于本领域的普通技术人员来说，在不脱离本申请创造构思的前提下，还可以做出改进，但这些均属于本申请的保护范围。

Claims (6)

  1. 一种脏污图像检测方法，包括：
    获取待检测镜片的影像；
    基于所述影像，确定检测区域的范围；
    确定所述检测区域范围内的脏污颗粒；
    基于所述脏污颗粒，检测所述检测区域内是否存在脏污；
    其中，所述确定所述检测区域范围内的脏污颗粒，包括：将所述检测区域划分为多个局部区域；捕捉所述局部区域内的多个像素点；选择待判定的所述像素点；对比所述像素点与周边其它像素点，以得出所述像素点与周边其它像素点之间灰度的差值；基于所述差值，判定所述像素点是否为所述脏污颗粒；
    所述基于所述脏污颗粒，检测所述检测区域内是否存在脏污，包括以下三种方法中的一种、两种或三种：判定所述检测区域内是否存在个体型的脏污；判定所述检测区域内是否存在聚集型的脏污；判定所述检测区域内是否存在区域型的脏污；
    其中，所述判定所述检测区域内是否存在个体型的脏污，包括：选择待判定的所述脏污颗粒；计算所述脏污颗粒的面积；基于所述脏污颗粒的面积，判定所述脏污颗粒是否属于个体型的脏污；
    所述判定所述检测区域内是否存在聚集型的脏污，包括：选择待判定的所述脏污颗粒；定义聚集型距离；连接所述脏污颗粒与其在所述聚集型距离内的其它脏污颗粒，以形成脏污颗粒集；计算所述脏污颗粒集的面积；基于所述脏污颗粒集的面积，判定所述脏污颗粒集是否属于聚集型的脏污；
    所述判定所述检测区域内是否存在区域型的脏污，包括：定义判定区域；计算所述判定区域内所述脏污颗粒的个数以及面积；基于所述脏污颗粒的个数以及面积，判定所述判定区域是否属于区域型的脏污。
  2. 根据权利要求1所述的脏污图像检测方法，其特征在于，所述基于所述影像，确定检测区域的范围，包括：
    计算所述检测区域的中心坐标；
    定义所述检测区域的半径；
    基于所述中心坐标及所述半径，计算所述检测区域的范围。
  3. 根据权利要求1所述的脏污图像检测方法，其特征在于，在所述基于所述影像，确定检测区域的范围之后，在所述确定所述检测区域范围内的脏污颗粒之前，还包括：
    去除影响脏污判定结果的亮斑。
  4. 根据权利要求3所述的脏污图像检测方法，其特征在于，所述去除影响脏污判定结果的亮斑，包括：
    定义脏污的最大外接椭圆长边半径；
    去除最大外接椭圆短边半径大于脏污最大外接椭圆长边半径的亮斑。
  5. 一种脏污图像检测装置，其特征在于，包括：
    影像获取单元，所述影像获取单元用于获取待检测镜片的影像；
    第一处理单元，所述第一处理单元用于确定检测区域的范围；
    第二处理单元，所述第二处理单元用于确定所述检测区域范围内的脏污颗粒；
    检测单元，所述检测单元用于检测所述检测区域内是否存在脏污；
    其中，所述第二处理单元包括：划分单元，所述划分单元用于将所述检测区域划分为多个局部区域；捕捉单元，所述捕捉单元用于捕捉所述局部区域内的多个像素点；选像单元，所述选像单元用于选择待判定的所述像素点；对比单元，所述对比单元用于对比所述像素点与周边其它像素点，以得出所述像素点与周边其它像素点之间灰度的差值；脏污颗粒获取单元，所述脏污颗粒获取单元用于基于所述差值，判定所述像素点是否为所述脏污颗粒；
    所述检测单元包括以下三种单元中的一种、两种或三种：第一判定单元，所述第一判定单元用于判定所述检测区域内是否存在个体型的脏污；第二判定单元，所述第二判定单元用于判定所述检测区域内是否存在聚集型的脏污；第三判定单元，所述第三判定单元用于判定所述检测区域内是否存在区域型的脏污；
    所述第一判定单元包括：第一选择单元，所述第一选择单元用于选择待判定的所述脏污颗粒；第一计算单元，所述第一计算单元用于计算所述脏污颗粒的面积；第一判定子单元，所述第一判定子单元用于判定所述脏污颗粒是否属于个体型脏污；
    所述第二判定单元包括：第二选择单元，所述第二选择单元用于选择待判定的所述脏污颗粒；第一定义单元，所述第一定义单元用于定义聚集型距离；脏污颗粒连接单元，所述脏污颗粒连接单元用于连接所述脏污颗粒与其在所述聚集型距离内的其它脏污颗粒，以形成脏污颗粒集；第二计算单元，所述第二计算单元用于计算所述脏污颗粒集的面积；第二判定子单元，所述第二判定子单元用于判定所述脏污颗粒集是否属于聚集型的脏污；
    所述第三判定单元包括：第二定义单元，所述第二定义单元用于定义判定区域；第三计算单元，所述第三计算单元用于计算所述判定区域内所述脏污颗粒的个数以及面积；第三判定子单元，所述第三判定子单元用于判定所述判定区域是否属于区域型的脏污。
  6. 一种脏污图像检测机构，其特征在于，用于实现如权利要求1-4任意一项所述的脏污图像检测方法，具体包括用于采集待采集镜片影像的相机、用于驱动所述相机沿竖直方向移动的驱动组件以及用于照射待采集镜片的发光组件。
PCT/CN2020/132652 2020-11-17 2020-11-30 脏污图像检测方法、脏污图像检测装置及脏污图像检测机构 WO2022104900A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011282276.5 2020-11-17
CN202011282276.5A CN112102319B (zh) 2020-11-17 2020-11-17 脏污图像检测方法、脏污图像检测装置及脏污图像检测机构

Publications (1)

Publication Number Publication Date
WO2022104900A1 true WO2022104900A1 (zh) 2022-05-27

Family

ID=73785682

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/132652 WO2022104900A1 (zh) 2020-11-17 2020-11-30 脏污图像检测方法、脏污图像检测装置及脏污图像检测机构

Country Status (2)

Country Link
CN (1) CN112102319B (zh)
WO (1) WO2022104900A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116183940A (zh) * 2023-02-07 2023-05-30 泰州奥尔斯顿生物科技有限公司 基于污点分布鉴别的生物检测分析装置

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113570597A (zh) * 2021-09-01 2021-10-29 南通中煌工具有限公司 基于人工智能的泥头车车厢脏污程度的判定方法及***

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160004144A1 (en) * 2014-07-04 2016-01-07 The Lightco Inc. Methods and apparatus relating to detection and/or indicating a dirty lens condition
CN106231297A (zh) * 2016-08-29 2016-12-14 深圳天珑无线科技有限公司 摄像头的检测方法及装置
CN111246204A (zh) * 2020-03-24 2020-06-05 昆山丘钛微电子科技有限公司 一种基于相对亮度偏差的脏污检测方法和装置
CN111726612A (zh) * 2020-07-07 2020-09-29 歌尔科技有限公司 镜头模组脏污检测方法、***、设备及计算机存储介质

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6317223B1 (en) * 1998-12-14 2001-11-13 Eastman Kodak Company Image processing system for reducing vertically disposed patterns on images produced by scanning
CN102413354B (zh) * 2011-10-05 2014-04-30 深圳市联德合微电子有限公司 一种手机摄像模组自动光学检测方法、装置及***
CN102410974A (zh) * 2011-12-14 2012-04-11 华北电力大学 气流输送管道中颗粒料粒度分布及形状分布在线测量方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160004144A1 (en) * 2014-07-04 2016-01-07 The Lightco Inc. Methods and apparatus relating to detection and/or indicating a dirty lens condition
CN106231297A (zh) * 2016-08-29 2016-12-14 深圳天珑无线科技有限公司 摄像头的检测方法及装置
CN111246204A (zh) * 2020-03-24 2020-06-05 昆山丘钛微电子科技有限公司 一种基于相对亮度偏差的脏污检测方法和装置
CN111726612A (zh) * 2020-07-07 2020-09-29 歌尔科技有限公司 镜头模组脏污检测方法、***、设备及计算机存储介质

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116183940A (zh) * 2023-02-07 2023-05-30 泰州奥尔斯顿生物科技有限公司 基于污点分布鉴别的生物检测分析装置
CN116183940B (zh) * 2023-02-07 2024-05-14 广东蓝莺高科有限公司 基于污点分布鉴别的生物检测分析装置

Also Published As

Publication number Publication date
CN112102319B (zh) 2021-02-12
CN112102319A (zh) 2020-12-18

Similar Documents

Publication Publication Date Title
CN110766684B (zh) 一种基于机器视觉的定子表面缺陷检测***及检测方法
WO2022082904A1 (zh) 镜头脏污检测方法、装置和设备
WO2022104900A1 (zh) 脏污图像检测方法、脏污图像检测装置及脏污图像检测机构
CN103743761B (zh) 一种镜片水印疵病图像检测装置
CN110108711A (zh) 圆环侧壁缺陷的视觉检测***
CN107610085A (zh) 一种基于计算机视觉的焊点缺陷检测***
CN110751635B (zh) 一种基于帧间差分和hsv颜色空间的口腔检测方法
CN107274403B (zh) 一种浮选表面质量的评价方法
CN111551350A (zh) 一种基于U_Net网络的光学镜片表面划痕检测方法
CN109461156B (zh) 基于视觉的螺纹密封塞装配检测方法
JP7393313B2 (ja) 欠陥分類装置、欠陥分類方法及びプログラム
TWI512284B (zh) 玻璃氣泡瑕疵檢測系統
CN114577805A (zh) 一种MiniLED背光面板缺陷检测方法及装置
JP2017227474A (ja) 照明装置、及び、画像検査装置
CN110570412B (zh) 一种零件误差视觉判断***
US11493453B2 (en) Belt inspection system, belt inspection method, and recording medium for belt inspection program
CN105606623A (zh) 一种基于线阵相机的钢板表面缺陷检测***
CN111008960B (zh) 基于机器视觉的铝电解电容底部外观检测方法及装置
CN109990742A (zh) 基于图像处理技术的板栗检测方法
CN111833350A (zh) 机器视觉检测方法与***
CN111724375A (zh) 一种屏幕检测方法及***
CN104881652A (zh) 一种基于玉米穗凸性特征的行数自动检测算法
CN111815705B (zh) 激光跟踪仪滤光保护镜片污染识别方法、装置及电子设备
CN115272173A (zh) 锡球缺陷检测方法及其装置、计算机设备、存储介质
JP2007285868A (ja) 輝度勾配検出方法、欠陥検出方法、輝度勾配検出装置および欠陥検出装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20962191

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20962191

Country of ref document: EP

Kind code of ref document: A1