CN107145909B - Method for classifying surface to which optical element damage belongs - Google Patents

Method for classifying surface to which optical element damage belongs Download PDF

Info

Publication number
CN107145909B
CN107145909B CN201710317105.3A CN201710317105A CN107145909B CN 107145909 B CN107145909 B CN 107145909B CN 201710317105 A CN201710317105 A CN 201710317105A CN 107145909 B CN107145909 B CN 107145909B
Authority
CN
China
Prior art keywords
damage
optical element
light
classification model
classifying
Prior art date
Legal status (The legal status 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 status listed.)
Active
Application number
CN201710317105.3A
Other languages
Chinese (zh)
Other versions
CN107145909A (en
Inventor
陈凤东
刘国栋
魏富鹏
刘炳国
彭志涛
唐军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Harbin Institute of Technology
Laser Fusion Research Center China Academy of Engineering Physics
Original Assignee
Harbin Institute of Technology
Laser Fusion Research Center China Academy of Engineering Physics
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 Harbin Institute of Technology, Laser Fusion Research Center China Academy of Engineering Physics filed Critical Harbin Institute of Technology
Priority to CN201710317105.3A priority Critical patent/CN107145909B/en
Publication of CN107145909A publication Critical patent/CN107145909A/en
Application granted granted Critical
Publication of CN107145909B publication Critical patent/CN107145909B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
    • G06F18/24133—Distances to prototypes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/217—Validation; Performance evaluation; Active pattern learning techniques

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

The invention provides a method for classifying optical element damage belonging to a surface by automatically distinguishing light incident surface damage or light emergent surface damage, belonging to the field of optical damage classification. The invention specifically comprises the following steps: selecting an optical element; acquiring damage online images of the optical element vacuum isolation sheet by using an FODI system, and marking all damage points in the light transmission caliber range of the optical element vacuum isolation sheet; scanning the light incident surface and the light emergent surface of the vacuum isolation sheet of the optical element within the light transmission caliber range one by one, and recording the position and the form of a damage point as offline data; matching the off-line data to the acquired on-line image by using geometric transformation to obtain a training sample set with damages to the light incident surface and the light emergent surface; establishing a classification model, and training the classification model by adopting a training sample set to obtain the optimal parameters of the classification model; and classifying the damage of the optical element by using a classification model with optimal parameters to determine the damage of the light incident surface or the light emergent surface. The invention is used for identifying and classifying the damage in the FODI system.

Description

Method for classifying surface to which optical element damage belongs
Technical Field
The present invention relates to a classification method, and more particularly, to a classification method for a surface to which an optical element is damaged.
Background
The large solid laser device has large scale, numerous optical elements and high output energy and power, and is a main force device for Inertial Confinement Fusion (ICF) research. Under high power conditions, optical element damage becomes a troublesome problem that people must solve. In order to ensure that the growth process of Damage is found and tracked in time, an on-line Damage detection system (FODI) of the terminal optical element collects images of the terminal optical element after each targeting experiment, a data processing module is used for identifying and classifying the Damage in the images, all possible damages are marked, but the surface to which the Damage belongs cannot be classified, and the technical requirement of the FODI system is not met.
Disclosure of Invention
In view of the above disadvantages, the present invention provides a method for classifying damaged surfaces of optical elements, which can automatically distinguish the damage of the light incident surface or the light emitting surface.
The method for classifying the surface to which the optical element damage belongs comprises the following steps:
the method comprises the following steps: selecting an optical element;
step two: acquiring damage online images of the optical element vacuum isolation sheet by using an FODI system, marking all damage points in the light transmission aperture range of the optical element vacuum isolation sheet, and representing each damage point by using a characteristic vector;
step three: scanning the light incident surface and the light emergent surface of the vacuum isolation sheet of the optical element within the light transmission caliber range one by one, and recording the position and the form of a damage point as offline data;
step four: matching the off-line data in the step three to the on-line image acquired in the step two by using geometric transformation to obtain a training sample set with damages to the light incident surface and the light emergent surface;
step five: establishing a classification model, and training the classification model by adopting a training sample set to obtain the optimal parameters of the classification model;
step six: and classifying the damage of the optical element by using a classification model with optimal parameters to determine the damage of the light incident surface or the light emergent surface.
Preferably, the fourth step further comprises:
classifying the test sample set by using a classification model with optimal parameters, judging the classification accuracy, and if the classification accuracy meets the requirements, turning to the sixth step; otherwise, increasing the set sample collection number, and turning to the step one;
the first step is as follows: and selecting the optical elements according to the set sample collection quantity.
Preferably, the feature vector includes a pixel area of the damage point, a signal gray sum, a noise gray sum, a signal mean, a signal variance, a noise mean, a signal maximum gray value, a noise maximum gray value, a sum of local signal-to-noise ratios, a signal-to-noise energy ratio, a saturation area ratio, a saturation gray ratio, a long axis length of a matching ellipse and a short axis length of a matching ellipse, an abscissa of an image where the damage point is located, and an ordinate of an image where the damage point is located.
Preferably, the classification model is implemented using an overrun learning machine in the form of a kernel in machine learning.
Preferably, the fourth step is:
and (3) matching the off-line data in the step three to the on-line image acquired in the step two by using geometric transformation to obtain a sample set with the damage of the light inlet surface and the light outlet surface, and dividing the sample set with the damage of the light inlet surface and the light outlet surface into two parts, wherein one part is used as a training sample set, and the other part is used as a testing sample set.
The features mentioned above can be combined in various suitable ways or replaced by equivalent features as long as the object of the invention is achieved.
The method has the advantages that the method analyzes the light emitting characteristics of the damages of the light inlet surface and the light outlet surface based on the finite time domain difference method and the Fourier optical angular spectrum theory, then uses the characteristic vector to represent the damages of the light inlet surface and the light outlet surface in the FODI online image, finally uses the ultralimit learning machine to realize the classification of the surfaces to which the damages belong, and automatically distinguishes whether the damages are on the light inlet surface or the light outlet surface.
Drawings
FIG. 1 is a schematic diagram illustrating the principle of imaging damage to a light incident surface;
FIG. 2 is a schematic diagram illustrating a principle of light-emitting surface damage imaging;
FIG. 3 is a schematic two-dimensional topography of smooth arc-shaped pits and burred arc-shaped pits;
FIG. 4 is a diagram showing the light intensity distribution of a smooth circular-arc-shaped pit, in which FIG. 41 shows the light intensity distribution at the lens when only the illumination light is emitted, FIG. 42 shows the light intensity distribution at the CCD when only the illumination light is emitted, FIG. 43 shows the light intensity distribution at the lens when the illumination light plus stray light is emitted, and FIG. 44 shows the light intensity distribution at the CCD when the illumination light plus stray light is emitted;
FIG. 5 is a diagram showing the light intensity distribution of the burred arc-shaped pits, in which FIG. 51 shows the light intensity distribution at the lens when only the illumination light is emitted, FIG. 52 shows the light intensity distribution at the CCD when only the illumination light is emitted, FIG. 53 shows the light intensity distribution at the lens when the illumination light plus stray light is emitted, and FIG. 54 shows the light intensity distribution at the CCD when the illumination light plus stray light is emitted;
FIG. 6 is an FODI online image of the damage to the light incident surface;
FIG. 7 is an FODI online image of a damaged light emitting surface;
FIG. 8 is a schematic diagram illustrating a method for classifying a surface to which an optical element is damaged according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The invention is further described with reference to the following drawings and specific examples, which are not intended to be limiting.
As shown in fig. 1 and 2, it is assumed that two damage points have the same shape and size and are respectively located on the light incident surface and the light emitting surface of the vacuum isolation sheet of the optical element, and after the illumination on the optical element side is turned on, the total internal reflection light inside the optical element is broken at the damage points. Diffuse reflection and diffuse transmission occur at the point of injury, assuming that the light energy radiated to the point of injury is the same. For the damage of the light incident surface, part of the diffusely reflected light is radiated onto an imaging system to form a damage image of the light incident surface on the CCD; for the damage of the light-emitting surface, part of the light which is diffused and transmitted is radiated onto the imaging system, and a damage image of the light-emitting surface is formed on the CCD. Since the scattering and transmission properties of the damaged surface are different in most cases, the scattering and transmission energies of the surface will be different, and the resulting properties such as gray scale morphology will be different. For most blast-pit type damages in a superluminescent-III high power laser device, it was experimentally found that the depth h was typically about one fifth of the lateral diameter d. Taking a damage with a depth of 20 μm and a lateral dimension of 100 μm as an example, it is assumed that the damaged surface is divided into a smooth shape and a circular arc shape with burrs as shown in FIG. 3.
Assuming that the amplitude of the internally reflected illumination light irradiated to the damaged point is 1, the amplitude of the stray light is 0.5. And solving far-field light intensity distribution at the position of the imaging lens by using a finite time domain difference method, and then calculating the light intensity distribution at the surface of the CCD by using an angular spectrum method.
As can be seen from fig. 4 and 5, the light intensity distributions of the damage points of the light incident surface and the light emergent surface at the imaging lens and the CCD image surface are different, especially the far-field light intensity distribution at the CCD image surface, and the waveform difference between the two is shown in the 12 characteristic attributes listed on the left side of table 1.
A damage point with the dimension of about 300 mu m on the light inlet surface of the vacuum isolation sheet of the optical element is selected, and an online image is shot by using a FODI camera. Rotating optical element 180°And enabling the damage point to be positioned on the light-emitting surface and shooting an online image by using a FODI camera. The images of the same damage on the light incident surface and the light emergent surface of the optical element are shown in fig. 6 and 7.
As can be seen from fig. 6 and 7, when the same damage is located at the light incident surface and the light emergent surface, the bright spots formed in the FODI image are different. Comparing the waveform differences of the theoretical simulation with the 12 features on the left side in table 1, the present embodiment correspondingly proposes the 1 st to 12 th FODI online image features on the right side of table 1. In addition, in order to characterize the difference in spot shape, the major axis length and the minor axis length of the matching ellipse are added, so the damage difference between the light incident surface and the light emitting surface is shown in table 1.
TABLE 1 far-field luminescence characteristics of light incident and emergent surfaces and FODI image characteristics
Figure BDA0001288788550000041
The FODI system employs a side illumination technique, a significant feature of which is illumination non-uniformity. The luminous power of the miniature semiconductor laser LD (laser diode) is set to be 1.58W, the light wavelength is 808nm, the divergence angle is 70 degrees, and the light polarization degree is 0.9. The refractive index of the fused silica under illumination at a wavelength of 808nm corresponds to 1.45319. The optical element vacuum insulation sheet has dimensions of 430mm × 430mm × 10 mm. The light scattering problem caused by the surface roughness of the element is processed by a Bidirectional Scattering Distribution Function (BSDF) model. When using ray tracing, the light propagation cut-off threshold is set to 1% (i.e. the ray starts from the light source and splits during tracing until the sub-ray terminates propagation when its carrying flux accounts for less than 1% of the starting flux). The sizes of the metal frame and the LD light emitting surface are set according to actual conditions. And obtaining the distribution simulation result of the internal reflection illumination light field of the vacuum isolation sheet.
According to simulation results, the optical field distribution inside the optical element changes along with the change of the position, and in order to represent the influence of the illumination nonuniformity on imaging, two characteristics of coordinates (X, Y) are added when representing the damage in the FODI image, and the position of the damage is represented. From this, the feature vector x ═ x can be obtained(1),...,x(16)]Wherein x is(1)Is the area of a pixel, x(2)For signal gray sum, x(3)Is the sum of the noise gray levels, x(4)Is the mean value of the signal, x(5)Is the signal variance, x(6)Is the mean value of the noise, x(7)For maximum gray value of the signal, x(8)Is the noise maximum gray value, x(9)Is the sum of local signal-to-noise ratios, x(10)Is the signal to noise energy ratio, x(11)Is the saturation area ratio, x(12)To saturate the gray scale ratio, x(13)To match the major axis length of the ellipse, x(14)To match the minor axis length, x, of the ellipse(15)Is the abscissa X, X of the image of the lesion(16)Is the ordinate Y of the image on which the lesion is located. I.e. a feature vector of a total of 16 features
Figure BDA0001288788550000051
To characterize the ith lesion.
In the present embodiment, a Kernel-based Extreme Learning Machine (Kernel-based Extreme Learning Machine) in Machine Learning is used, and the classification model of the Kernel-based Extreme Learning Machine is:
Figure BDA0001288788550000052
wherein K (x, x)i) Is a kernel function, x ═ x(1),...,x(16)]For the input of the sample to be classified,
Figure BDA0001288788550000053
Figure BDA0001288788550000054
for training samples, M is the number of training samples, I is the identity matrix, C is a constant, ΩtrainFor training the kernel matrix formed by the samples, omegatrain i,j＝K(xi,xj)，(i,j＝1,…,M),T＝[y1,…,yM]TA class label matrix for the training samples.
The vector output form corresponding to the N test samples is as follows:
Figure BDA0001288788550000055
wherein F (x) is [ ((x))1),…,f(xN)]T，ΩtestFor the kernel matrix formed by the test samples and the training samples, omegatest i,j＝K(xi,xj) (i ═ 1, …, N; j ═ 1, …, M). Kernel function selection K (x, x)i)＝exp(-γ||x-xi||2)。
According to the above discussion, as shown in fig. 8, the method for classifying a surface to which an optical element damage belongs according to the present embodiment includes the steps of:
the method comprises the following steps: selecting optical elements according to the set sample collection quantity;
step two: acquiring damage online images of the optical element vacuum isolation sheet by using an FODI system, marking all damage points in the light transmission aperture range of the optical element vacuum isolation sheet, and representing each damage point by using a characteristic vector;
step three: scanning the light incident surface and the light emergent surface of the vacuum isolation sheet of the optical element in the light transmission caliber range one by using a microscope, and recording the position and the shape of a damaged point as offline data;
step four: matching the off-line data in the step three to the on-line image acquired in the step two by using geometric transformation to obtain a training sample set with damages to the light incident surface and the light emergent surface;
step five: establishing a classification model, and training the classification model by adopting a training sample set to obtain the optimal parameters of the classification model; classifying the test sample set by using a classification model with optimal parameters, judging the classification accuracy, and if the classification accuracy meets the requirements, turning to the sixth step; otherwise, increasing the set sample collection number, and turning to the step one;
step six: and classifying the damage of the optical element by using a classification model with optimal parameters to determine the damage of the light incident surface or the light emergent surface.
Although the invention herein has been described with reference to particular embodiments, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present invention. It is therefore to be understood that numerous modifications may be made to the illustrative embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that features described in different dependent claims and herein may be combined in ways different from those described in the original claims. It is also to be understood that features described in connection with individual embodiments may be used in other described embodiments.

Claims (2)

1. A method of classifying a surface to which an optical element is damaged, the method comprising the steps of:
the method comprises the following steps: selecting optical elements according to the set sample collection quantity;
step two: acquiring damage online images of the optical element vacuum isolation sheet by using an FODI system, marking all damage points in the light transmission aperture range of the optical element vacuum isolation sheet, and representing each damage point by using a characteristic vector;
step three: scanning the light incident surface and the light emergent surface of the vacuum isolation sheet of the optical element within the light transmission caliber range one by one, and recording the position and the form of a damage point as offline data;
step four: matching the off-line data in the step three to the on-line image acquired in the step two by using geometric transformation to obtain a sample set with damages to the light inlet surface and the light outlet surface, and dividing the sample set into two parts, wherein one part is used as a training sample set, and the other part is used as a test sample set;
step five: establishing a classification model, and training the classification model by adopting a training sample set to obtain the optimal parameters of the classification model;
step six: classifying damage of the optical element by using a classification model with optimal parameters, and determining the damage of the optical element as light incident surface damage or light emergent surface damage;
the classification model is as follows:
Figure FDA0002802081890000011
wherein K (x, x)i) Is a kernel function, x ═ x(1),...,x(16)]For the input feature vector, x, of the sample to be classified(1)Is the area of a pixel, x(2)For signal gray sum, x(3)Is the sum of the noise gray levels, x(4)Is the mean value of the signal, x(5)Is the signal variance, x(6)Is the mean value of the noise, x(7)For maximum gray value of the signal, x(8)Is the noise maximum gray value, x(9)Is the sum of local signal-to-noise ratios, x(10)Is the signal to noise energy ratio, x(11)Is the saturation area ratio, x(12)To saturate the gray scale ratio, x(13)To match the major axis length of the ellipse, x(14)To match the minor axis length, x, of the ellipse(15)Is the abscissa, x, of the image on which the lesion is located(16)As the ordinate of the image on which the lesion is located,
Figure FDA0002802081890000012
for training samples, I is 1, …, M is the number of training samples, I is the identity matrix, C is a constant, ΩtrainA kernel matrix constructed for training samples, T ═ y1,...,yM]TClass label matrix for training sample;
the vector output form corresponding to the N test samples is as follows:
Figure FDA0002802081890000013
wherein F (x) is [ ((x))1),...,f(xN)]T，ΩtestSelecting K (x, x) for kernel function of kernel matrix formed by test sample and training samplei)＝exp(-γ||x-xi||2)。
2. The method of classifying a surface to which an optical element damage belongs according to claim 1, wherein the step five further comprises:
classifying the test sample set by using a classification model with optimal parameters, judging the classification accuracy, and if the classification accuracy meets the requirements, turning to the sixth step; otherwise, increasing the set sample collection number and turning to the step one.
CN201710317105.3A 2017-05-08 2017-05-08 Method for classifying surface to which optical element damage belongs Active CN107145909B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710317105.3A CN107145909B (en) 2017-05-08 2017-05-08 Method for classifying surface to which optical element damage belongs

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710317105.3A CN107145909B (en) 2017-05-08 2017-05-08 Method for classifying surface to which optical element damage belongs

Publications (2)

Publication Number Publication Date
CN107145909A CN107145909A (en) 2017-09-08
CN107145909B true CN107145909B (en) 2021-02-02

Family

ID=59776753

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710317105.3A Active CN107145909B (en) 2017-05-08 2017-05-08 Method for classifying surface to which optical element damage belongs

Country Status (1)

Country Link
CN (1) CN107145909B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108668080B (en) * 2018-06-22 2020-09-08 北京小米移动软件有限公司 Method and device for prompting degree of dirt of lens and electronic equipment
CN112580264B (en) * 2020-12-25 2023-07-07 中国人民解放军国防科技大学 BP neural network algorithm-based damage point size distribution prediction method and system
CN113237888B (en) * 2021-05-21 2022-06-14 哈尔滨工业大学 Large-caliber optical element damage point online and offline dark field image matching method

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010271177A (en) * 2009-05-21 2010-12-02 Honda Motor Co Ltd Surface examination device
CN105512609A (en) * 2015-11-25 2016-04-20 北京工业大学 Multi-mode fusion video emotion identification method based on kernel-based over-limit learning machine
WO2016097520A1 (en) * 2014-12-15 2016-06-23 Compagnie Generale Des Etablissements Michelin Method for detecting a defect on a surface of a tyre

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103389310B (en) * 2013-07-31 2015-04-15 哈尔滨工业大学 Online sub-pixel optical component damage detection method based on radiation calibration
CN103743750B (en) * 2014-01-14 2016-08-17 中国科学院自动化研究所 A kind of generation method of distribution diagram of surface damage of heavy calibre optical element
CN105092608B (en) * 2015-09-24 2017-11-03 哈尔滨工业大学 The elimination method of twin image in final-optics element damage on-line checking

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010271177A (en) * 2009-05-21 2010-12-02 Honda Motor Co Ltd Surface examination device
WO2016097520A1 (en) * 2014-12-15 2016-06-23 Compagnie Generale Des Etablissements Michelin Method for detecting a defect on a surface of a tyre
CN105512609A (en) * 2015-11-25 2016-04-20 北京工业大学 Multi-mode fusion video emotion identification method based on kernel-based over-limit learning machine

Also Published As

Publication number Publication date
CN107145909A (en) 2017-09-08

Similar Documents

Publication Publication Date Title
CN108332689B (en) Optical measurement system and method for detecting surface roughness and surface damage
US10281408B2 (en) Inspection object imaging apparatus, inspection object imaging method, surface inspection apparatus, and surface inspection method
CN107735674B (en) Surface defect detection device, surface defect detection method, and steel product manufacturing method
CN107145909B (en) Method for classifying surface to which optical element damage belongs
CN109099859B (en) Device and method for measuring surface defect three-dimensional morphology of large-caliber optical element
CN103635794A (en) Method and system for optically inspecting parts
CN109829883B (en) Product quality detection method and device
CN110390661A (en) It checks management system, check managing device and checks management method
WO2016208626A1 (en) Surface flaw detection method, surface flaw detection device, and manufacturing method for steel material
CN108662993A (en) A kind of Surface roughness measurement system based on optical scattering principle
JP2023512280A (en) Detector for object recognition
Jiang et al. Measuring particle size distribution using LED-illumination
Inoshita et al. Shape from single scattering for translucent objects
CN106895793A (en) The method and apparatus of double mode depth survey
Wei et al. Automatic classification of true and false laser-induced damage in large aperture optics
CN110044931A (en) A kind of detection device on bend glass surface and internal flaw
Shi et al. Research on laser stripe characteristics and center extraction algorithm for desktop laser scanner
JP2018072260A (en) Inspection condition determination device, inspection condition determination method and inspection condition determination program
JP6782449B2 (en) Surface inspection method and its equipment
McCarthy et al. LED-based collimating line-light combining freeform and fresnel optics
Torng et al. Development of an automated optical inspection system for mobile phone panels
CN112858341B (en) Detection method, shooting system and detection system
TW202227894A (en) Bullet casing illumination module and forensic analysis system using the same
CN113483692A (en) Hole detection optical system
Ekwongmunkong et al. Automated machine vision system for inspecting cutting quality of cubic zirconia

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
CB03 Change of inventor or designer information

Inventor after: Chen Fengdong

Inventor after: Liu Guodong

Inventor after: Wei Fupeng

Inventor after: Liu Bingguo

Inventor after: Peng Zhitao

Inventor after: Tang Jun

Inventor before: Liu Guodong

Inventor before: Chen Fengdong

Inventor before: Wei Fupeng

Inventor before: Liu Bingguo

Inventor before: Peng Zhitao

Inventor before: Tang Jun

CB03 Change of inventor or designer information
GR01 Patent grant
GR01 Patent grant