CN109949294A - A kind of fracture apperance figure crack defect extracting method based on OpenCV - Google Patents
A kind of fracture apperance figure crack defect extracting method based on OpenCV Download PDFInfo
- Publication number
- CN109949294A CN109949294A CN201910215988.6A CN201910215988A CN109949294A CN 109949294 A CN109949294 A CN 109949294A CN 201910215988 A CN201910215988 A CN 201910215988A CN 109949294 A CN109949294 A CN 109949294A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel
- value
- fracture
- point
- 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.)
- Pending
Links
Abstract
A kind of fracture apperance figure crack defect extracting method based on OpenCV, belongs to field of image processing.It is existing to metal fracture behavior judgment method empirically depending on, there is a problem of judgement inaccuracy.Stretching fracture feature image is pre-processed using thresholding method, by image crack pixel and background pixel graphical demarcation and highlights: processing of making an uproar of dispelling is carried out using mean filter or median filter method；The extraction of fracture apperance figure crack defect；Edge extracting is carried out to pretreated image, obtains the image of intuitive crack appearance；Image thinning process；Search the Contour moment of simultaneously drawing image.
Description
Technical field
The fracture apperance figure crack defect extracting method based on OpenCV that the present invention relates to a kind of.
Background technique
To the research of metal fracture behavior, there are two types of different methods at present.One is the method for fracture mechanics, it according to
According to being fracture and elastic plastic theory, material inherent shortcoming is considered；Another kind is the method for metal physics, it is from the micro- of material
Tissue, microdefect, even molecule and atom scale on study fracture mechanism.Fracture failure is primarily due in mechanical component
Or cracked in engineering goods or fracture, so, to study fracture failure cause or influence factor just must emphatically or first
Type, mechanism and the feature of analysis and research crackle or fracture, that is, the rule and influence factor of the crackle generation that analyze and research.
Fracture undergoes the formation extension of crackle until the different phases such as final fracture, each stage all with internal, external, mechanics
, the several factors such as chemical and physics it is related；Each stage of fracture process can leave accordingly on fracture again simultaneously
Trace, pattern and feature.Fracture analysis is exactly to be disclosed by observation, identification and the analysis to these traces, pattern and feature
The correlative factor of fracture process out, to distinguish the property and mechanism of fracture failure.Presently mainly according to the existing warp of expert
It tests and is qualitatively classified to a certain fracture apperance, then judge the specific failure mode of this failure product on this basis.Cause
This objectively carries out the basis that classification is failure analysis work to fracture apperance, significant.Metal fracture is Yi Menguan
The science for examining research metal fracture surface characteristics form plays a significant role in the engineerings such as evaluation of material and failure.Observation and
Analysis metal fracture can provide important information for the performance of research metal material and behavior etc., pass through the spy to metal fracture
The research of the problems such as sign, shape, pattern, can preferably disclose the mechanism of fracture process, and research influences fracture process and fracture
The various factors of form, thus preferably Instructing manufacture solid line.
Summary of the invention
The purpose of the present invention is to solve it is existing to metal fracture behavior judgment method empirically depending on, exist and sentence
The problem for inaccuracy of breaking, and propose a kind of fracture apperance figure crack defect extracting method based on OpenCV.
A kind of fracture apperance figure crack defect extracting method based on OpenCV, the method are realized by following steps:
Step 1: pre-processed to stretching fracture feature image using thresholding method, by image crack pixel with
Background pixel graphical demarcation simultaneously highlights:
Firstly, finding the pixel to be divided of image, corresponding threshold value then is set to these pixels；Wherein, it sets
The gray value of threshold pixels point to have stronger contrast to facilitate observation and subsequent image to handle.
Step 2: carrying out processing of making an uproar of dispelling using mean filter or median filter method；
Step 3: the extraction of fracture apperance figure crack defect:
Edge extracting is carried out to pretreated image, obtains the image of intuitive crack appearance；
Step 4: image thinning process；
Step 5: searching the Contour moment of simultaneously drawing image.
The invention has the benefit that
The present invention is that the OpenCV being related to is the cross-platform computer vision library based on BSD license (open source) distribution,
It may operate in Linux, Windows, Android and Mac OS operating system.It increase income and efficiently -- by a series of C functions
It is constituted with a small amount of C++ class, while providing the interface of the language such as Python, Ruby, MATLAB, realize image procossing and calculating
Many general-purpose algorithms of machine visual aspects.Using in OpenCV function library, AdaptiveThreshold () instruction be may be implemented
Threshold automatically selects, i.e., adaptive threshold method.Variable's attribute based on threshold, is automatically selected by histogram, to reach most
Good effect.Median filtering belongs to nonlinear filtering, and the gray scale of the pixel is replaced using the intermediate value of neighborhood of pixel points gray value
Value is effective when to fracture apperance figure crack defect image procossing to inhibit noise.There is the gradient value much put larger in neighborhood,
Edge detection is carried out to image using the sobel operator in OpenCV edge detection, can effectively be given up non-edge point
It abandons.
Detailed description of the invention
Fig. 1 is flow chart of the method for the present invention；
Fig. 2 is block of pixels of the present invention diagram；
Fig. 3 is that each pixel is labeled in Fig. 2, and indicates that foreground pixel, light color indicate background pixel with dark color；
Specific embodiment
Specific embodiment 1:
A kind of fracture apperance figure crack defect extracting method based on OpenCV of present embodiment, the method by with
Lower step is realized:
Step 1: pre-processed to stretching fracture feature image using thresholding method, by image crack pixel with
Background pixel graphical demarcation simultaneously highlights:
Firstly, finding the pixel to be divided of image, corresponding threshold value then is set to these pixels；Wherein, it sets
The gray value of threshold pixels point to have stronger contrast to facilitate observation and subsequent image to handle.By to stretching fracture figure
As analysis, crackle characteristic can be confirmed, but image, in collection process, the interference of the factors such as ambient enviroment, light causes crackle
It is inadequate with the discrimination of background.The pattern dividing method of threshold can extract necessary pixel from figure.
Step 2: carrying out processing of making an uproar of dispelling using mean filter or median filter method；
Step 3: the extraction of fracture apperance figure crack defect:
After stretching fracture image preprocessing, available complete crack image.Edge is carried out to pretreated image
It extracts, obtains the image of intuitive crack appearance；
Step 4: image thinning process；
Step 5: searching the Contour moment of simultaneously drawing image.
Specific embodiment 2:
Unlike specific embodiment one, a kind of fracture apperance figure crackle based on OpenCV of present embodiment is lacked
Extracting method is fallen into, in the step two, dispel using mean filter method the process of processing of making an uproar are as follows:
The filter that median filtering considers be it is linear, i.e., the sum of the response of two signals and each response are equal
's.Linear filtering is easy to construct, and is analyzed from frequency response angle.However neighborhood picture is used in many cases,
The nonlinear filtering of element can obtain better effect.Such as when noise is shot noise, with linear filterings pair such as mean filters
When image is handled, noise will not be removed, they can only be converted into more soft but still visible shot.This is just
It needs to use median filtering.
In OpenCV, function Threshold () can complete above-mentioned five kinds of thresholdizations operation.But function
Threshold () has an apparent defect, as reaches preferable image processing effect, in operation will be to parameter
Threshold is repeatedly modified.However in OpenCV, oneself of threshold is may be implemented in function AdaptiveThreshold ()
Dynamic selection, i.e., adaptive threshold method.This is improved threshold technology, and wherein threshold itself is a variable, certainly based on histogram
Dynamic selection threshold, to reach optimum efficiency.
Adaptive threshold function prototype:
C++:void adaptiveThreshold(InputArray src,OutputArray dst,double
maxValue,int adaptiveMethod,int thresholdType,int blockSize,double C)
Discovery binaryzation threshold obviously distinguishes crackle with background with truncation threshold after being handled by thresholdization.But
It is integrally partially dark using truncation threshold treated image crack color and background color.So final choice binaryzation threshold is handled
Image, because image is only left black-and-white two color, and such picture contrast is high, after the processing of binaryzation threshold convenient for observation.
Mean filter is typical linear filtering algorithm, and main method is neighborhood averaging, i.e., with a piece of image-region
The average value of each pixel replaces each pixel values of source images, provides a template, the mould to object pixel on the image
Plate includes surrounding adjacent pixels, and original pixel is being replaced with the average value of the entire pixels in template, that is, is being treated
The current pixel point (x, y) of processing selects a template, which is made of several pixels closed on, seek template in all pictures
The mean value of element, then current pixel point (x, y) is assigned the mean value, as the gray scale point g (x, y) of image after processing at that point,
That is:
Wherein, m is in the template comprising the total number of pixels including current pixel；
In the step two, dispel using median filter method the process of processing of making an uproar are as follows: the median filtering is
A kind of typical nonlinear filtering technique, basic thought are the ashes that the pixel is replaced with the intermediate value of neighborhood of pixel points gray value
Angle value；Median filtering is the nonlinear signal processing technology for effectively inhibiting noise based on the theoretical one kind of sequencing statistical, basic
Principle is that the intermediate value that the value of any in digital picture or Serial No. is used in each point value in a neighborhood of the point replaces, and allows week
The pixel value enclosed is close to true value, to eliminate isolated noise spot；Median filtering is and line in processing consecutive image window function
Property filtering working method it is similar, but filtering is no longer ranking operation.
Median filtering has apparent advantage compared to mean filter.In mean filter, since noise contribution is placed into
In average value, so output receives the influence of noise.But in median filtering, since noise contribution is difficult to be chosen, institute
Hardly to influence output.Therefore in same region, the ability that median filtering eliminates the method for noise is even better.
Median filtering is all a good method either in terms of eliminating noise and still saving edge.
Specific embodiment 3:
Unlike specific embodiment one, a kind of fracture apperance figure crackle based on OpenCV of present embodiment is lacked
Extracting method is fallen into, in the step three, edge extracting is carried out to pretreated image, obtains intuitive crack appearance
Image process specifically:
After stretching fracture image preprocessing, available complete crack image carries out edge to pretreated image
Detection and Extraction crackle obtains the image of intuitive crack appearance: thus by pixel has in some neighborhood of pixels in image
There is the pixel (i.e. marginal point) of acute variation pixel value to extract.
Specific embodiment 4:
Unlike specific embodiment three, a kind of fracture apperance figure crackle based on OpenCV of present embodiment is lacked
Extracting method is fallen into, the edge detection process is realized by following steps:
1) filter edge detection algorithm:
Single order and second dervative based on image intensity, derivative, need to be using filtering come improving image quality to noise-sensitive；
Pretreated image is carried out using adaptive median filter, can reach the image request using edge detection；
2) enhance edge algorithms:
After the changing value for determining each vertex neighborhood intensity of image, gradient magnitude is calculated by enhancing algorithm, by image grayscale point
The point that neighborhood intensity value has significant change highlights.
3) detection algorithm:
Image after enhancing has the gradient value much put larger in neighborhood, is detected by threshold method, by non-edge
Point is given up；
Edge detection is carried out to image using the sobel operator in OpenCV edge detection；Sobel operator is one main
Discrete differential operator for edge detection；It combines Gaussian smoothing and differential derivation, for calculating image grayscale function
Approximate gradient；This operator is used at any point of image, can all generate corresponding gradient vector or its law vector.
Wherein, the calculating process of Sobel operator:
1) respectively in x and y both direction derivation；
1. level variation: I (being applied image is I) and the kernel Gx of an odd sized are carried out convolution.For example, when interior
When core size is 3, the calculated result of Gx are as follows:
2. vertical change: the kernel of I and odd sized is carried out convolution；For example, being calculated when kernel size is 3
As a result are as follows:
2) in the every bit of image, approximate solution is found out in conjunction with two above result:
Or it is replaced using following formula:
G=| Gx|+|Gy|。
Specific embodiment 5:
Unlike specific embodiment four, a kind of fracture apperance figure crackle based on OpenCV of present embodiment is lacked
Extracting method is fallen into, in the step four, image thinning process is to generate the process of the skeleton of an object, and skeleton is usually
One two value object；Skeleton indicates that the shape of object, all these pixels are all structural pixels with less pixel, thus
All it is necessary pixel.Image thinning process can simply be defined as finding out the necessary picture of those description object shapes in object
The process of element, the necessary pixel of these description object shapes are Skeleton pixel, binary map of the set of composition as micronization processes
Picture, background are black, and being worth is 0, and the foreground object pixel value to be refined is 255；Whether judge object pixel as follows
The pixel fallen for the refinement:
Such as the block of pixels that Fig. 2 is 3 × 3, each pixel is labeled.Indicate that foreground pixel, light color indicate back with dark color
Scene element, as shown in figure 3, P9 is foreground pixel, P4 is background pixel, and P1 is object pixel, that is, need to judge should
The pixel being refined.
Algorithm is divided into two steps:
Step 1: the foreground pixel point that circulation is all, is labeled as deleting to the pixel for meeting following condition:
1,2≤N (P1)≤6, wherein N (P1) is indicated in 8 pixels adjacent with P1, is the number of foreground pixel point；
2, S (P1)=1, wherein S (P1) is indicated sequentially the number that paired value is 0,1 is distinguished in front and back between P2-P9-P2；
3, P2*P4*P6=0；
4, P4*P6*P8=0；
Step 2: the foreground pixel point that circulation is all, is labeled as deleting to the pixel for meeting following condition；
1,2≤N (P1)≤6, wherein be the number of foreground pixel point in N (P1) expression 8 pixels adjacent with P1；
2, S (P1)=1, wherein S (P1) is indicated sequentially the number that paired value is 0,1 is distinguished in front and back between P2-P9-P2；
3, P2*P4*P8=0；
4, P2*P6*P8=0.
Claims (5)
1. a kind of fracture apperance figure crack defect extracting method based on OpenCV, it is characterised in that: the method passes through following
Step is realized:
Step 1: being pre-processed to stretching fracture feature image using thresholding method, by image crack pixel and background
Pixel graphics are distinguished and are highlighted:
Firstly, finding the pixel to be divided of image, corresponding threshold value then is set to these pixels；Wherein, the threshold of setting
The gray value of value pixel will have stronger contrast to facilitate observation and subsequent image to handle；
Step 2: carrying out processing of making an uproar of dispelling using mean filter or median filter method；
Step 3: the extraction of fracture apperance figure crack defect:
Edge extracting is carried out to pretreated image, obtains the image of intuitive crack appearance；
Step 4: image thinning process；
Step 5: searching the Contour moment of simultaneously drawing image.
2. a kind of fracture apperance figure crack defect extracting method based on OpenCV according to claim 1, it is characterised in that:
In the step two, dispel using mean filter method the process of processing of making an uproar are as follows:
Each pixel value that source images are replaced with the average value of each pixel of a piece of image-region, on the image to target picture
Element provides a template, which includes surrounding adjacent pixels, with the average value of the entire pixels in template come generation
For original pixel, i.e., to current pixel point (x, y) to be processed, select a template, the template is by several pixels for closing on
Composition, the mean value for the middle all pixels that seek template, then current pixel point (x, y) is assigned the mean value, as image after processing at this
Gray scale point g (x, y) on point, it may be assumed that
Wherein, m is in the template comprising the total number of pixels including current pixel；
In the step two, dispel using median filter method the process of processing of making an uproar are as follows: with neighborhood of pixel points gray value
Intermediate value replaces the gray value of the pixel；The value of any in digital picture or Serial No. is used in a neighborhood of the point
The intermediate value of each point value replaces, and allows the pixel value of surrounding close to true value, to eliminate isolated noise spot.
3. a kind of fracture apperance figure crack defect extracting method based on OpenCV according to claim 2, it is characterised in that:
In the step three, edge extracting is carried out to pretreated image, obtains the mistake of the image of intuitive crack appearance
Journey specifically:
Edge detection is carried out to pretreated image and extracts crackle, obtains the image of intuitive crack appearance: thus will
There is pixel the pixel of acute variation pixel value to extract in some neighborhood of pixels in image.
4. a kind of fracture apperance figure crack defect extracting method based on OpenCV according to claim 3, it is characterised in that:
The edge detection process is realized by following steps:
1) filter edge detection algorithm:
Single order and second dervative based on image intensity, derivative, need to be using filtering come improving image quality to noise-sensitive；It uses
Adaptive median filter carries out pretreated image；
2) enhance edge algorithms:
After the changing value for determining each vertex neighborhood intensity of image, gradient magnitude is calculated by enhancing algorithm, by image grayscale vertex neighborhood
The point that intensity value has significant change highlights；
3) detection algorithm:
Image after enhancing has the gradient value much put larger in neighborhood, is detected by threshold method, by non-edge point into
Row is given up；
Edge detection is carried out to image using the sobel operator in OpenCV edge detection；Calculate the approximation of image grayscale function
Gradient:
Wherein, the calculating process of Sobel operator:
1) respectively in x and y both direction derivation；
1. level variation: I (being applied image is I) and the kernel Gx of an odd sized are carried out convolution；For example, when kernel is big
It is small when being 3, the calculated result of Gx are as follows:
2. vertical change: the kernel of I and odd sized is carried out convolution；When kernel size is 3, calculated result are as follows:
2) in the every bit of image, approximate solution is found out in conjunction with two above result:
Or it is replaced using following formula:
G=| Gx|+|Gy|。
5. a kind of fracture apperance figure crack defect extracting method based on OpenCV according to claim 4, it is characterised in that:
In the step four, image thinning process is to find out the process of the necessary pixel of description object shape in object, these descriptions
The necessary pixel of object shapes is Skeleton pixel, and bianry image of the set as micronization processes of composition, background is black, is worth and is
0, the foreground object pixel value to be refined is 255；Judge whether object pixel is pixel that the refinement is fallen as follows:
Step 1: the foreground pixel point that circulation is all, is labeled as deleting to the pixel for meeting following condition:
1,2≤N (P1)≤6, wherein N (P1) is indicated in 8 pixels adjacent with P1, is the number of foreground pixel point；
2, S (P1)=1, wherein S (P1) is indicated sequentially the number that paired value is 0,1 is distinguished in front and back between P2-P9-P2；
3, P2*P4*P6=0；
4, P4*P6*P8=0；
Step 2: the foreground pixel point that circulation is all, is labeled as deleting to the pixel for meeting following condition；
1,2≤N (P1)≤6, wherein be the number of foreground pixel point in N (P1) expression 8 pixels adjacent with P1；
2, S (P1)=1, wherein S (P1) is indicated sequentially the number that paired value is 0,1 is distinguished in front and back between P2-P9-P2；
3, P2*P4*P8=0；
4, P2*P6*P8=0.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910215988.6A CN109949294A (en) | 2019-03-21 | 2019-03-21 | A kind of fracture apperance figure crack defect extracting method based on OpenCV |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910215988.6A CN109949294A (en) | 2019-03-21 | 2019-03-21 | A kind of fracture apperance figure crack defect extracting method based on OpenCV |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109949294A true CN109949294A (en) | 2019-06-28 |
Family
ID=67011127
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910215988.6A Pending CN109949294A (en) | 2019-03-21 | 2019-03-21 | A kind of fracture apperance figure crack defect extracting method based on OpenCV |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109949294A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110766743A (en) * | 2019-10-23 | 2020-02-07 | 中冶赛迪重庆信息技术有限公司 | Material flow detection method, device, equipment and medium based on image recognition |
| CN112396635A (en) * | 2020-11-30 | 2021-02-23 | 深圳职业技术学院 | Multi-target detection method based on multiple devices in complex environment |
| CN112487642A (en) * | 2020-11-27 | 2021-03-12 | 成都大学 | Fatigue fracture morphology feature extraction method based on flooding filling algorithm |
| CN114441546A (en) * | 2022-04-08 | 2022-05-06 | 湖南万航科技有限公司 | Crack measurement endoscope |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103839268A (en) * | 2014-03-18 | 2014-06-04 | 北京交通大学 | Method for detecting fissure on surface of subway tunnel |
| CN106780526A (en) * | 2016-11-21 | 2017-05-31 | 浙江工业大学 | A kind of ferrite wafer alligatoring recognition methods |
| CN107169967A (en) * | 2017-07-07 | 2017-09-15 | 哈尔滨理工大学 | A kind of diamond saw blade crack detecting method based on image procossing |
-
2019
- 2019-03-21 CN CN201910215988.6A patent/CN109949294A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103839268A (en) * | 2014-03-18 | 2014-06-04 | 北京交通大学 | Method for detecting fissure on surface of subway tunnel |
| CN106780526A (en) * | 2016-11-21 | 2017-05-31 | 浙江工业大学 | A kind of ferrite wafer alligatoring recognition methods |
| CN107169967A (en) * | 2017-07-07 | 2017-09-15 | 哈尔滨理工大学 | A kind of diamond saw blade crack detecting method based on image procossing |
Non-Patent Citations (6)
| Title |
|---|
| 何东健 等: "《数字图像处理》", 28 February 2018 * |
| 孟庆娇: "中碳钢铸坯角部横裂纹预测方法的研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》 * |
| 毛星云 等: "《OpenCV3编程入门》", 28 February 2015 * |
| 牛硕: "金属膜片表面缺陷检测", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 王健荣: "基于图像特征的钢轨表面瑕疵识别方法", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 董晴晴 等: "基于图像处理技术的管道裂缝检测方法研究", 《应用科技》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110766743A (en) * | 2019-10-23 | 2020-02-07 | 中冶赛迪重庆信息技术有限公司 | Material flow detection method, device, equipment and medium based on image recognition |
| CN112487642A (en) * | 2020-11-27 | 2021-03-12 | 成都大学 | Fatigue fracture morphology feature extraction method based on flooding filling algorithm |
| CN112487642B (en) * | 2020-11-27 | 2024-02-13 | 成都大学 | Fatigue fracture morphology feature extraction method based on water-flooding filling algorithm |
| CN112396635A (en) * | 2020-11-30 | 2021-02-23 | 深圳职业技术学院 | Multi-target detection method based on multiple devices in complex environment |
| CN112396635B (en) * | 2020-11-30 | 2021-07-06 | 深圳职业技术学院 | Multi-target detection method based on multiple devices in complex environment |
| CN114441546A (en) * | 2022-04-08 | 2022-05-06 | 湖南万航科技有限公司 | Crack measurement endoscope |
| CN114441546B (en) * | 2022-04-08 | 2022-06-24 | 湖南万航科技有限公司 | Crack measurement endoscope |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109949294A (en) | A kind of fracture apperance figure crack defect extracting method based on OpenCV | |
| Liu et al. | Single image dehazing via large sky region segmentation and multiscale opening dark channel model | |
| CN108416766B (en) | Double-side light-entering type light guide plate defect visual detection method | |
| CN109472788B (en) | Method for detecting flaw on surface of airplane rivet | |
| CN109978848B (en) | Method for detecting hard exudation in fundus image based on multi-light-source color constancy model | |
| CN109544571A (en) | A kind of metallic phase image edge detection method based on mathematical morphology | |
| CN108830857B (en) | Self-adaptive Chinese character copy label image binarization segmentation method | |
| CN114399522A (en) | High-low threshold-based Canny operator edge detection method | |
| CN115908415B (en) | Edge-based defect detection method, device, equipment and storage medium | |
| Mustafa | Feature selection using sequential backward method in melanoma recognition | |
| CN113780110A (en) | Method and device for detecting weak and small targets in image sequence in real time | |
| EP3293672A1 (en) | Particle boundary identification | |
| CN112750089B (en) | Optical remote sensing image defogging method based on local block maximum and minimum pixel prior | |
| Valliammal et al. | A hybrid method for enhancement of plant leaf recognition | |
| Chen et al. | An improved edge detection in noisy image using fuzzy enhancement | |
| CN104102911A (en) | Image processing for AOI (automated optical inspection)-based bullet appearance defect detection system | |
| WO2019181072A1 (en) | Image processing method, computer program, and recording medium | |
| CN110930358A (en) | Solar panel image processing method based on self-adaptive algorithm | |
| Zhang et al. | Motion detection based on improved Sobel and ViBe algorithm | |
| CN106600662B (en) | Method and device for drawing main lines in image | |
| CN110458042B (en) | Method for detecting number of probes in fluorescent CTC | |
| Shao et al. | An adaptive image contrast enhancement algorithm based on retinex | |
| Khan et al. | Segmentation of single and overlapping leaves by extracting appropriate contours | |
| CN109949245B (en) | Cross laser detection positioning method and device, storage medium and computer equipment | |
| CN112614062B (en) | Colony counting method, colony counting device and computer storage medium |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190628 |
|
| WD01 | Invention patent application deemed withdrawn after publication |