CN109584253A - Oil liquid abrasive grain image partition method - Google Patents

Oil liquid abrasive grain image partition method Download PDF

Info

Publication number
CN109584253A
CN109584253A CN201811559431.6A CN201811559431A CN109584253A CN 109584253 A CN109584253 A CN 109584253A CN 201811559431 A CN201811559431 A CN 201811559431A CN 109584253 A CN109584253 A CN 109584253A
Authority
CN
China
Prior art keywords
region
image
watershed
color
segmentation
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.)
Granted
Application number
CN201811559431.6A
Other languages
Chinese (zh)
Other versions
CN109584253B (en
Inventor
任松
徐雪茹
王小书
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.)
Chongqing Checkway Technology Co Ltd
Original Assignee
Chongqing Checkway Technology Co Ltd
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 Chongqing Checkway Technology Co Ltd filed Critical Chongqing Checkway Technology Co Ltd
Priority to CN201811559431.6A priority Critical patent/CN109584253B/en
Publication of CN109584253A publication Critical patent/CN109584253A/en
Application granted granted Critical
Publication of CN109584253B publication Critical patent/CN109584253B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • 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
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/155—Segmentation; Edge detection involving morphological operators
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics
    • 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
    • G06T2207/10024—Color image
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a kind of oil liquid abrasive grain image partition methods, comprising steps of 1) completing the watershed segmentation of Debris Image based on adaptive H-minima technology；2) Color characteristics parameters of Debris Image are extracted based on Lab color space；3) the textural characteristics parameter of Debris Image is extracted based on LBP map；4) similarity of watershed homogeneous region is determined based on Pasteur's distance；5) seed region is marked based on similarity matrix automatically；6) merging of watershed homogeneous region is completed based on region merging technique criterion；7) region merging technique image is modified based on Morphological scale-space, completes image segmentation.The present invention is greatly improved the pervasive of algorithm and is answered by the adaptive adjustment of the realization key parameter in the links of segmentation process.Compared with the conventional method, the present invention relates to the information for taking full advantage of Color Debris Image carrying, and cutting procedure complete display, segmentation result are stablized, and greatly improves particle partition work order of accuarcy.

Description

Oil liquid abrasive grain image partition method
Technical field
The present invention relates to oil liquid abrasive grain analysis technical fields, more particularly to a kind of pervasive iron to compose Debris Image segmentation side Method.
Background technique
Iron spectrum image segmentation is to carry out the key link of oil liquid abrasive grain analytical technology, and segmented image is abrasive grain automatic identification work Make the premise carried out, segmentation precision and segmentation efficiency directly affect the order of accuarcy of abrasive grain automatic identification work.It is transported by equipment The influence of row operating condition complexity, abrasive grain present in oil liquid show various shapes and texture diversity, and outstanding behaviours is in grain Diameter, thickness, edge, surface, background etc. cause Debris Image otherness and class inherited in significant class occur.? Existing iron spectrum Debris Image cutting techniques are applied to different abrasive grain figures there are under the influence of the significance difference opposite sex by Debris Image When as segmentation work, over-segmentation or less divided phenomenon are in uncontrollable state always, need to draw on the basis of partitioning algorithm Enter the processes such as morphology post-processing significantly to be corrected to segmentation result, therefore the quality of segmentation result is too dependent on The interactive process of segmentation personnel, although ensure that the Accurate Segmentation of every Debris Image under this method, with abrasive grain The increase of amount of images, the task amount for dividing work can obviously increase, so that the meaning of abrasive grain automatic identification work is difficult to obtain It highlights.Therefore, how to improve the adaptive degree of partitioning algorithm, reduce personnel to the maximum extent to the interactive shadow of cutting procedure It rings, is one of oil liquid abrasive grain image segmentation field assistant officer critical issue to be solved, and realize that wear Particles Recognition workflow is automatic One of critical issue of change.
At present in abrasive grain automatic identification link, the segmentation of iron spectrum Debris Image mainly uses Otsu threshold, color clustering etc. Method, these methods are low to the utilization rate of Debris Image carrying information, segmentation result is very unstable, lack a set of information utilization Height, process are reasonable, the adaptive stronger oil liquid abrasive grain image partition method of degree is next improves the automatic of wear Particles Recognition process comprehensively Change degree.
Summary of the invention
In view of this, the object of the present invention is to provide a kind of oil liquid abrasive grain image partition method, to solve in the prior art Iron spectrum Wear Particle Image Segmentation Method existing for the problems such as segmentation result is unstable, excessive dependence interactive process, utilize the set Process can complete the batch splitting work of Debris Image, improve the degree of automation of wear Particles Recognition work.
Oil liquid abrasive grain image partition method of the present invention, comprising the following steps:
1) it treats segmented image and carries out gray scale morphology reconstruct, extract gradient image, based on H-minima technology to gradient Image is modified, and the setting of H and its correction value H ' are as follows:
H '=β H (0 β≤1 <) (2)
In formula: M0、M1、M2Respectively indicate the mean value of modifying gradient image, the mean value of local minimum and local maximum Mean value, β are definite value modifying factor；
Watershed variation is carried out to modifying gradient image, completes the primary segmentation of Debris Image, obtains watershed homogeneity area Domain；
2) homogeneous region color feature extracted in watershed is completed in Lab color space: Debris Image is turned from rgb space Change to Lab space, the channel L, the channel a, b channel image are compressed to N number of color grade respectively, obtained in watershed segmentation image The color value of each homogeneous region, successively extracts the distribution of color histogram of each region, and it is normalized；
3) extraction of watershed homogeneous region textural characteristics is completed based on LBP map: determining P in the region that radius is R A sampled point obtains LBP map, and the grain distribution for calculating each homogeneous region in watershed segmentation image based on LBP map is straight Fang Tu, and it is normalized；
4) similarity measurement of color, texture normalization distribution histogram is respectively completed based on Bhattacharyya coefficient, Homogeneous region RmWith RnBetween color similarity ρcolorWith texture similarity ρLBPCalculating it is as follows:
In formula:Respectively region RmColor normalization histogram and texture normalization it is straight Fang Tu；Respectively region RnColor normalization histogram and texture normalization histogram；
During Fusion Features, definition region RmWith RnComprehensive similarity matrix w are as follows:
w(Rm,Rn)=wcolor(Rm,Rn)·ρcolor(Rm,Rn)+wLBP(Rm,Rn)·ρLBP(Rm,Rn) (7)
The comprehensive similarity matrix W of watershed homogeneous region are as follows:
5) seed region is marked based on similarity matrix automatically comprising following steps:
A, the maximum homogeneous region of product behind watershed segmentation the selection of background seed region and label: is defined as background Seed region Rb, it is 1 by the zone marker, i.e. L (Rb)=1；
B, the selection in foreground seeds region and label: tracking the edge contour of target area, to profile exterior pixel Mark value is judged that the corresponding mark value of external pixels point is adjacent with target area, thus obtains the label of neighboring region Value is subject to adjacent area if the homogeneous region mark value after watershed algorithm segmentation constitutes set s, foreground seeds region Specific deterministic process are as follows:
1. extracting background area RbNeighboring region mark value, be denoted as Rv,AndForeground seeds region is in RvIn It is selected；S indicates the mark value of homogeneous region after watershed algorithm segmentation, is set, and v is a subset of s, in v not Include b element；
2. successively obtaining RvNeighboring region, be denoted as Rvr,R is extracted from similarity matrix WvIt is adjacent region Similarity value w (Rv,Rr)；
3. according to similarity w successively zoning RvNode degree matrix D, corresponding region when matrix D is maximized Mark value v0It is defined as foreground area seed point, is 0 by the zone marker；
Calculation formula during foreground seeds zone marker is as follows:
v0=max (D (Rv)) (9)
6) merging of watershed homogeneous region is completed based on region merging technique criterion
Based on background seed region RbWith foreground seeds regionRegion Ri(i ∈ s and i ≠ b, i ≠ v0) merging rule Are as follows:
7) after the completion of merging work, unified morphology post-processing is carried out to the image after merging, completes the segmentation of image Work.
Beneficial effects of the present invention:
Oil liquid abrasive grain image partition method of the present invention has the advantages that compared with existing Debris Image cutting techniques
1) this method takes full advantage of the information of wear particle color image carrying, individual Debris Image is being divided It cuts and achieves best compromise between efficiency and segmentation precision, guarantee the stability of segmentation effect.
2) this method provides the Debris Images of complete set to divide process, has carried out in various degree to each segmentation link Optimization, take full advantage of the advantage of Morphology Algorithm but farthest reduce the dependence journey to morphology adjustment effect Degree, by introducing adaptive H value in H-minima technology, introducing adaptive weighting during Fusion Features, realization seed The methods of the automatic selection in region and label, realize the adaptive adjustment of partitioning algorithm, avoid the interaction in cutting procedure Formula processing, provides possibility for the batch splitting of Debris Image.
3) this method is suitable for all kinds of abrasive grains generated under typical wear mechanism, simultaneously as color characteristic and texture are special The extracting method of sign all has scale invariability, therefore this method is also applied for the Debris Image under various scales, has fine Universality.
Detailed description of the invention
Fig. 1 is the flow chart of oil liquid abrasive grain image segmentation；
Fig. 2 is Debris Image to be split；
Fig. 3 is the watershed tag image for extract after Morphological Reconstruction to gray level image to be split；
Fig. 4 is the watershed tag image extracted after being modified based on H-minima technology to gradient image；
Fig. 5 is that the color of the watershed homogeneous region 1 and 39 for the Debris Image to be split extracted based on Lab color space is returned One changes histogram；
Fig. 6 is the texture normalization of the watershed homogeneous region 1 and 39 for the Debris Image to be split extracted based on LBP map Histogram；
Fig. 7 is that the image after watershed homogeneous region merges is completed based on region merging technique criterion；
Fig. 8 is the image after being modified based on morphology post-processing to merging image；
Fig. 9 is the flow chart that unified morphology post-processing is carried out to the image after merging.
Specific embodiment
The invention will be further described with reference to the accompanying drawings and examples.
As shown, the present embodiment oil liquid abrasive grain image partition method, comprising the following steps:
1) it treats segmented image and carries out gray scale morphology reconstruct, extract gradient image, based on H-minima technology to gradient Image is modified, and the setting of H and its correction value H ' are as follows:
H '=β H (0 β≤1 <) (2)
In formula: M0、M1、M2Respectively indicate the mean value of modifying gradient image, the mean value of local minimum and local maximum Mean value, β are definite value modifying factor, take β=0.5 on the basis of for statistical analysis to test sample；
Watershed variation is carried out to modifying gradient image, completes the primary segmentation of Debris Image, obtains watershed homogeneity area Domain.
2) homogeneous region color feature extracted in watershed is completed in Lab color space: Debris Image is turned from rgb space Change to Lab space, the channel L, the channel a, b channel image are compressed to N number of color grade respectively, N=16 is taken in the present embodiment, obtains The color value of each homogeneous region in watershed segmentation image is taken, successively extracts the distribution of color histogram of each region, and right It is normalized.
3) extraction of watershed homogeneous region textural characteristics is completed based on LBP map: determining P in the region that radius is R A sampled point takes R=2, P=8 in the present embodiment；LBP map is obtained, is calculated based on LBP map each in watershed segmentation image The grain distribution histogram of a homogeneous region, and it is normalized.
4) similarity measurement of color, texture normalization distribution histogram is respectively completed based on Bhattacharyya coefficient, During Fusion Features, the weight of characteristic index is realized according to the similarity of feature to be divided in portion.Homogeneous region RmWith Rn Between color similarity ρcolorWith texture similarity ρLBPCalculating it is as follows:
In formula:Respectively region RmColor normalization histogram and texture normalization it is straight Fang Tu；Respectively region RnColor normalization histogram and texture normalization histogram.
During Fusion Features, definition region RmWith RnComprehensive similarity matrix w are as follows:
w(Rm,Rn)=wcolor(Rm,Rn)·ρcolor(Rm,Rn)+wLBP(Rm,Rn)·ρLBP(Rm,Rn) (7)
The comprehensive similarity matrix W of watershed homogeneous region are as follows:
5) seed region is marked based on similarity matrix automatically comprising following steps:
A, the maximum homogeneous region of product behind watershed segmentation the selection of background seed region and label: is defined as background Seed region Rb, it is 1 by the zone marker, i.e. L (Rb)=1；
B, the selection in foreground seeds region and label: tracking the edge contour of target area, to profile exterior pixel Mark value is judged that the corresponding mark value of external pixels point is adjacent with target area, thus obtains the label of neighboring region Value is subject to adjacent area if the homogeneous region mark value after watershed algorithm segmentation constitutes set s, foreground seeds region Specific deterministic process are as follows:
1. extracting background area RbNeighboring region mark value, be denoted as Rv,AndForeground seeds region is in RvIn It is selected；S indicates the mark value of homogeneous region after watershed algorithm segmentation, is set, and v is a subset of s, in v not Include b element；
2. successively obtaining RvNeighboring region, be denoted as Rvr,R is extracted from similarity matrix WvIt is adjacent region Similarity value w (Rv,Rr)；
3. according to similarity w successively zoning RvNode degree matrix D, corresponding region when matrix D is maximized Mark value v0It is defined as foreground area seed point, is 0 by the zone marker；
Calculation formula during foreground seeds zone marker is as follows:
v0=max (D (Rv)) (9)
6) merging of watershed homogeneous region is completed based on region merging technique criterion
Based on background seed region RbWith foreground seeds regionRegion Ri(i ∈ s and i ≠ b, i ≠ v0) merging rule Are as follows:
7) after the completion of merging work, unified morphology post-processing is carried out to the image after merging, completes the segmentation of image Work.The process for carrying out unified morphology post-processing to the image after merging is as follows:
The first step is handled the bianry image after merging based on morphological erosion operation, for eliminating in background Existing pseudo- cut-off rule selects circular configuration element, size se=1 in the step in the present embodiment；
Second step fills hole to through step 1 treated bianry image, for correcting abrasive particle surface because gray scale is violent The mistake segmenting pixels point of variation and appearance；
Third step, to through step 2 treated bianry image carries out deletion small area, guaranteeing only to include in segmentation visual field Target abrasive grain；
4th step carries out morphology opening operation to through step 3 treated bianry image, to eliminate segmentation noise spot simultaneously Smooth segmentation contour completes the segmentation work of Debris Image.Circular configuration element, size se are selected in the step in the present embodiment =3.
Finally, it is stated that the above examples are only used to illustrate the technical scheme of the present invention and are not limiting, although referring to compared with Good embodiment describes the invention in detail, those skilled in the art should understand that, it can be to skill of the invention Art scheme is modified or replaced equivalently, and without departing from the objective and range of technical solution of the present invention, should all be covered at this In the scope of the claims of invention.

Claims (1)

1. a kind of oil liquid abrasive grain image partition method, it is characterised in that: the following steps are included:
1) it treats segmented image and carries out gray scale morphology reconstruct, extract gradient image, based on H-minima technology to gradient image It is modified, the setting of H and its correction value H ' are as follows:
H '=β H (0 β≤1 <) (2)
In formula: M0、M1、M2Respectively indicate the equal of the mean value of modifying gradient image, the mean value of local minimum and local maximum Value, β are definite value modifying factor；
Watershed variation is carried out to modifying gradient image, completes the primary segmentation of Debris Image, obtains watershed homogeneous region；
2) extraction of watershed homogeneous region color characteristic is completed in Lab color space: Debris Image is converted from rgb space To Lab space, the channel L, the channel a, b channel image are compressed to N number of color grade respectively, obtained each in watershed segmentation image The color value of a homogeneous region, successively extracts the distribution of color histogram of each region, and it is normalized；
3) extraction of watershed homogeneous region textural characteristics is completed based on LBP map: determining that P are adopted in the region that radius is R Sampling point obtains LBP map, and the grain distribution histogram of each homogeneous region in watershed segmentation image is calculated based on LBP map, And it is normalized；
4) similarity measurement of color, texture normalization distribution histogram, homogeneity are respectively completed based on Bhattacharyya coefficient Region RmWith RnBetween color similarity ρcolorWith texture similarity ρLBPCalculating it is as follows:
In formula:Respectively region RmColor normalization histogram and texture normalize histogram Figure；Respectively region RnColor normalization histogram and texture normalization histogram；
During Fusion Features, definition region RmWith RnComprehensive similarity matrix w are as follows:
w(Rm,Rn)=wcolor(Rm,Rn)·ρcolor(Rm,Rn)+wLBP(Rm,Rn)·ρLBP(Rm,Rn) (7)
The comprehensive similarity matrix W of watershed homogeneous region are as follows:
5) seed region is marked based on similarity matrix automatically comprising following steps:
A, the maximum homogeneous region of product behind watershed segmentation the selection of background seed region and label: is defined as background seed Region Rb, it is 1 by the zone marker, i.e. L (Rb)=1；
B, the selection in foreground seeds region and label: tracking the edge contour of target area, to the label of profile exterior pixel Value is judged that the corresponding mark value of external pixels point is adjacent with target area, thus obtains the mark value of neighboring region, if Homogeneous region mark value after watershed algorithm segmentation constitutes set s, is subject to adjacent area, and the specific of foreground seeds region is sentenced Disconnected process are as follows:
1. extracting background area RbNeighboring region mark value, be denoted as Rv,AndForeground seeds region is in RvMiddle progress Selection；S indicates the mark value of homogeneous region after watershed algorithm segmentation, is a set, and v is a subset of s, does not include b in v Element；
2. successively obtaining RvNeighboring region, be denoted as Rvr,R is extracted from similarity matrix WvIt is adjacent the phase in region Like angle value w (Rv,Rr)；
3. according to similarity w successively zoning RvNode degree matrix D, corresponding zone marker value when matrix D is maximized v0It is defined as foreground area seed point, is 0 by the zone marker；
Calculation formula during foreground seeds zone marker is as follows:
v0=max (D (Rv)) (9)
6) merging of watershed homogeneous region is completed based on region merging technique criterion
Based on Background seed region RbWith foreground seeds regionRegion Ri(i ∈ s and i ≠ b, i ≠ v0) merging rule Are as follows:
7) after the completion of merging work, unified morphology post-processing is carried out to the image after merging, completes the segmentation work of image.
CN201811559431.6A 2018-12-20 2018-12-20 Oil abrasive particle image segmentation method Active CN109584253B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811559431.6A CN109584253B (en) 2018-12-20 2018-12-20 Oil abrasive particle image segmentation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811559431.6A CN109584253B (en) 2018-12-20 2018-12-20 Oil abrasive particle image segmentation method

Publications (2)

Publication Number Publication Date
CN109584253A true CN109584253A (en) 2019-04-05
CN109584253B CN109584253B (en) 2022-08-30

Family

ID=65930126

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811559431.6A Active CN109584253B (en) 2018-12-20 2018-12-20 Oil abrasive particle image segmentation method

Country Status (1)

Country Link
CN (1) CN109584253B (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110223299A (en) * 2019-06-12 2019-09-10 重庆邮电大学 A kind of particle partition method based on deposition process
CN111008949A (en) * 2019-08-16 2020-04-14 苏州喆安医疗科技有限公司 Soft and hard tissue detection method for tooth image
CN111445485A (en) * 2020-03-10 2020-07-24 西安工业大学 Online abrasive particle image data processing method
CN111681244A (en) * 2020-05-29 2020-09-18 山东大学 Blade image segmentation method, system, equipment and storage medium
CN112183556A (en) * 2020-09-27 2021-01-05 长光卫星技术有限公司 Port ore heap contour extraction method based on spatial clustering and watershed transformation

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050201618A1 (en) * 2004-03-12 2005-09-15 Huseyin Tek Local watershed operators for image segmentation
CN105809673A (en) * 2016-03-03 2016-07-27 上海大学 SURF (Speeded-Up Robust Features) algorithm and maximal similarity region merging based video foreground segmentation method
US20170098310A1 (en) * 2014-06-30 2017-04-06 Ventana Medical Systems, Inc. Edge-based local adaptive thresholding system and methods for foreground detection
CN106599793A (en) * 2016-11-21 2017-04-26 江苏大学 Marked watershed segmentation-based steel grain boundary automatic extraction method
CN107481241A (en) * 2017-08-24 2017-12-15 太仓安顺财务服务有限公司 A kind of color image segmentation method based on mixed method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050201618A1 (en) * 2004-03-12 2005-09-15 Huseyin Tek Local watershed operators for image segmentation
US20170098310A1 (en) * 2014-06-30 2017-04-06 Ventana Medical Systems, Inc. Edge-based local adaptive thresholding system and methods for foreground detection
CN105809673A (en) * 2016-03-03 2016-07-27 上海大学 SURF (Speeded-Up Robust Features) algorithm and maximal similarity region merging based video foreground segmentation method
CN106599793A (en) * 2016-11-21 2017-04-26 江苏大学 Marked watershed segmentation-based steel grain boundary automatic extraction method
CN107481241A (en) * 2017-08-24 2017-12-15 太仓安顺财务服务有限公司 A kind of color image segmentation method based on mixed method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
SUN HUI-JIE: ""Watershed Image Segmentation Algorithm Base on Particle Swarm and Region Growing"", 《2015 INTERNATIONAL CONFERENCE ON INTELLIGENT COMPUTING AND INTERNET OF THINGS (ICLT)》 *
李占波 等: ""基于改进分水岭和区域合并的彩色图像分割"", 《计算机工程与设计》 *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110223299A (en) * 2019-06-12 2019-09-10 重庆邮电大学 A kind of particle partition method based on deposition process
CN110223299B (en) * 2019-06-12 2021-06-18 重庆邮电大学 Abrasive particle segmentation method based on deposition process
CN111008949A (en) * 2019-08-16 2020-04-14 苏州喆安医疗科技有限公司 Soft and hard tissue detection method for tooth image
CN111008949B (en) * 2019-08-16 2021-09-14 苏州喆安医疗科技有限公司 Soft and hard tissue detection method for tooth image
CN111445485A (en) * 2020-03-10 2020-07-24 西安工业大学 Online abrasive particle image data processing method
CN111445485B (en) * 2020-03-10 2023-05-23 西安工业大学 Online abrasive particle image data processing method
CN111681244A (en) * 2020-05-29 2020-09-18 山东大学 Blade image segmentation method, system, equipment and storage medium
CN111681244B (en) * 2020-05-29 2022-06-21 山东大学 Blade image segmentation method, system, equipment and storage medium
CN112183556A (en) * 2020-09-27 2021-01-05 长光卫星技术有限公司 Port ore heap contour extraction method based on spatial clustering and watershed transformation

Also Published As

Publication number Publication date
CN109584253B (en) 2022-08-30

Similar Documents

Publication Publication Date Title
CN109584253A (en) Oil liquid abrasive grain image partition method
WO2023083059A1 (en) Road surface defect detection method and apparatus, and electronic device and readable storage medium
CN105718945B (en) Apple picking robot night image recognition method based on watershed and neural network
CN108537239B (en) Method for detecting image saliency target
Tosta et al. Segmentation methods of H&E-stained histological images of lymphoma: A review
Pang et al. Automatic segmentation of crop leaf spot disease images by integrating local threshold and seeded region growing
CN109636784A (en) Saliency object detection method based on maximum neighborhood and super-pixel segmentation
CN105894490A (en) Fuzzy integration multiple classifier integration-based uterine neck cell image identification method and device
Zhang et al. Segmentation of overlapping cells in cervical smears based on spatial relationship and overlapping translucency light transmission model
CN105931241B (en) A kind of automatic marking method of natural scene image
CN110796667A (en) Color image segmentation method based on improved wavelet clustering
CN109871900A (en) The recognition positioning method of apple under a kind of complex background based on image procossing
CN109087330A (en) It is a kind of based on by slightly to the moving target detecting method of smart image segmentation
Tareef et al. Automated three-stage nucleus and cytoplasm segmentation of overlapping cells
Wang et al. A hybrid method for the segmentation of a ferrograph image using marker-controlled watershed and grey clustering
CN108364300A (en) Vegetables leaf portion disease geo-radar image dividing method, system and computer readable storage medium
Zhao et al. Separate degree based Otsu and signed similarity driven level set for segmenting and counting anthrax spores
Zhang et al. Cytoplasm segmentation on cervical cell images using graph cut-based approach
He et al. Local and global Gaussian mixture models for hematoxylin and eosin stained histology image segmentation
CN111210447B (en) Hematoxylin-eosin staining pathological image hierarchical segmentation method and terminal
Wang et al. A fast image segmentation algorithm for detection of pseudo-foreign fibers in lint cotton
KR20240012738A (en) Cluster analysis system and method of artificial intelligence classification for cell nuclei of prostate cancer tissue
CN113850792A (en) Cell classification counting method and system based on computer vision
Schupp et al. Image segmentation via multiple active contour models and fuzzy clustering with biomedical applications
CN116188560A (en) Automatic grain identification and accurate quantization characterization method based on metallographic pictures

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
GR01 Patent grant
GR01 Patent grant