CN110647887A - Method for extracting internal marker in coal slime flotation foam image segmentation - Google Patents

Method for extracting internal marker in coal slime flotation foam image segmentation Download PDF

Info

Publication number
CN110647887A
CN110647887A CN201910667822.8A CN201910667822A CN110647887A CN 110647887 A CN110647887 A CN 110647887A CN 201910667822 A CN201910667822 A CN 201910667822A CN 110647887 A CN110647887 A CN 110647887A
Authority
CN
China
Prior art keywords
image
internal
extracting
coal slime
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.)
Pending
Application number
CN201910667822.8A
Other languages
Chinese (zh)
Inventor
田慕玲
王跃龙
孟海涛
武亚雄
武培雄
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.)
Taiyuan University of Technology
Original Assignee
Taiyuan University of Technology
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 Taiyuan University of Technology filed Critical Taiyuan University of Technology
Priority to CN201910667822.8A priority Critical patent/CN110647887A/en
Publication of CN110647887A publication Critical patent/CN110647887A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition
    • G06V30/14—Image acquisition
    • G06V30/148—Segmentation of character regions
    • G06V30/153—Segmentation of character regions using recognition of characters or words
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Probability & Statistics with Applications (AREA)
  • Multimedia (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

A method for extracting internal markers in coal slime flotation froth image segmentation is characterized in that aiming at the characteristics of large noise, low contrast ratio, difficult segmentation of bubbles and the like of a coal slime flotation froth image, a particle swarm optimization algorithm and a one-dimensional histogram weighted fuzzy C-means clustering algorithm are fused, the internal markers obtained by the two algorithms are superposed to be used as the internal markers of the image segmentation for extracting conventional marker points, capturing bright spots with darker bubbles and removing false bright spots, the internal marker obtained by the method overcomes the defects of the internal identification points obtained by one algorithm, supplements each other, prevents omission, ensures that the internal identification points are more accurate and real, the watershed image segmentation is more accurate, and the method is more suitable for extracting the internal marker in the watershed image segmentation of the coal slime flotation foam image of various coal preparation plants.

Description

Method for extracting internal marker in coal slime flotation foam image segmentation
Technical Field
The invention relates to an internal marker for coal slime flotation froth image segmentation, in particular to a method for extracting an internal marker symbol in the morphological watershed segmentation of a coal slime flotation froth image, and belongs to the field of digital image processing and coal slime flotation.
Background
The coal slime flotation froth image is an image acquired by using an industrial CCD camera and a matching device in real time in the actual flotation production process, and is different from nonferrous metal and mineral flotation froth, the coal slime flotation froth image has self specificity, the color and brightness information of the coal slime flotation froth image are not obvious, meanwhile, more dust exists in a flotation field, and the conditions are severe, so that the contrast of the acquired flotation froth image is poor, bubbles are mutually adhered and the edges are fuzzy, the bubbles are difficult to separate from the background, the edges of the bubbles are difficult to extract, an effective image internal marker and an effective image external dividing line are extracted in the coal slime flotation froth image segmentation, the gradient image is calibrated, and the calibrated image is segmented on the basis.
The outer segmentation line of the watershed image segmentation is obtained by extracting the watershed ridge line of the marker image after the marker image is transformed by the inner marker, so that the inner marker is the basis of the outer segmentation line; secondly, the internal marker and the external segmentation line are also the basis for performing morphological minimum value calibration on the gradient image, and the accuracy of the final watershed segmentation is determined by the internal marker and the external segmentation line together, so that the extraction of the internal marker is the key point of the watershed image segmentation, and if the internal identifier is not extracted sufficiently, the corresponding marker points can be lost, so that the segmentation area is reduced, and the under-segmentation of the watershed segmentation is caused; on the contrary, if too many false mark points are extracted, a plurality of unnecessary false mark points are added, so that the mark points are too many, over-segmentation in watershed segmentation is caused, and the image can be accurately segmented only by extracting correct internal mark symbols, so that the bubble size characteristics of the flotation froth image are obtained, the flotation working condition is grasped, and accurate prediction of product ash content is realized.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: how to better extract an internal marker used in the segmentation of a flotation froth image when the segmentation of the watershed image is carried out on the flotation froth image and a method for extracting the internal marker in the segmentation of the coal slime flotation froth image.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows: the method for extracting the internal marker in the coal slime flotation froth image segmentation provides an optimization algorithm for extracting the internal marker based on the fusion of a particle swarm optimization algorithm and a one-dimensional histogram weighted fuzzy C-means clustering algorithm in the coal slime flotation froth image segmentation of the coal slime flotation froth image obtained from a flotation field of a coal preparation plant.
The internal marker comprises identification points obtained by a particle swarm optimization algorithm and identification points obtained by a fuzzy C-means clustering algorithm weighted by a one-dimensional histogram, and the two identification points are subjected to morphological processing and then are superposed to serve as the internal marker for final image segmentation. By applying a particle swarm optimization algorithm, a two-dimensional threshold value is used as the position of a particle in the particle swarm optimization algorithm, the local entropy of the two-dimensional maximum entropy is used as a fitness function of the particle swarm optimization algorithm, the speed and the position of the particle are continuously adjusted according to a speed and position formula of the particle swarm, and the best position where each particle passes by the generation is found
Figure 718657DEST_PATH_IMAGE001
Global best position with current
Figure 508759DEST_PATH_IMAGE002
And after a plurality of generations of iterative operations, the positions of the particles are continuously updated, when a set termination condition is reached, a two-dimensional threshold value which enables the two-dimensional local entropy to obtain the maximum is finally found out and is used as a two-dimensional threshold value vector, the image is subjected to double-threshold value binary segmentation, a maximum value area of the binary image is extracted, and the result is used as an identification point through morphological corrosion operation.
Clustering the flotation froth images by adopting a one-dimensional histogram weighted fuzzy C-means clustering algorithm, and sampling
Figure 325405DEST_PATH_IMAGE003
，
Figure 327996DEST_PATH_IMAGE004
Is a gray scale, here
Figure 585802DEST_PATH_IMAGE005
256 samples in total, and the initial clustering center is taken as
Figure 77963DEST_PATH_IMAGE006
Wherein
Figure 319589DEST_PATH_IMAGE007
Is a gray-scale value of the image,
Figure 391450DEST_PATH_IMAGE008
is the minimum value of the gray scale of the image,is the median value of the gray levels of the image,
Figure 104508DEST_PATH_IMAGE010
is the maximum value of the gray scale of the image,
Figure 895747DEST_PATH_IMAGE011
，
Figure 974561DEST_PATH_IMAGE012
taking a random number of 0-1 and calculating a membership matrixAnd cluster center
Figure 509765DEST_PATH_IMAGE014
Simultaneously calculating the objective function according to the clustering objective function
Figure 725982DEST_PATH_IMAGE015
After continuously iterative updating, when the requirement is met
Figure 948102DEST_PATH_IMAGE016
Iteration is terminated, and fuzzy partition matrix is output
Figure 35007DEST_PATH_IMAGE017
And a cluster centerClustering the images of the coal slime flotation foam gray level images by a fuzzy C-means clustering algorithm weighted by a one-dimensional histogram, extracting local maximum value areas of the clustered images by adopting a local maximum value method for the clustered images, and extracting identification points by morphological corrosion operation; and performing superposition operation on the two identification points to serve as an internal marker of the final watershed image segmentation.
After the scheme is adopted, the method has the greatest advantages that the particle swarm optimization only needs to update through the internal speed in the threshold optimization process, complicated crossing and variation links are avoided, the two-dimensional threshold selection process is simpler and more efficient, the flotation image is clustered by fusing the one-dimensional histogram weighted fuzzy C-means clustering algorithm, some darker bright spots on bubbles are completely highlighted from the bubbles, some false bright spots generated due to emission are removed after clustering, internal markers obtained by the particle swarm optimization algorithm and the one-dimensional histogram weighted fuzzy C-means clustering algorithm are mutually supplemented, omission is prevented, the watershed image segmentation is more accurate and reasonable, the final whole watershed segmentation result is more accurate, a better image segmentation effect can be generated, and meanwhile, the phenomenon that the adjustment is performed by controlling the maximum value and minimum value depth parameter H in the maximum transformation algorithm based on expansion is abandoned The general algorithm of the obtained internal marker overcomes the defect of over-segmentation or under-segmentation generated in image segmentation due to distortion of the internal marker caused by blindness of H selection.
Drawings
FIG. 1 is a two-dimensional histogram of a coal slurry flotation froth image.
Detailed Description
The following provides a further detailed description of specific embodiments of the present invention.
As shown in fig. 1, by implementing the method for extracting an internal marker in coal slime flotation froth image segmentation provided by the present invention, a particle swarm optimization algorithm is used, and a local entropy of a two-dimensional maximum entropy is used as a fitness function to obtain a two-dimensional threshold value of image binarization segmentation, so that a maximum value region of an image after the image is subjected to dual-threshold binarization is extracted to obtain an identification point.
1. Random formation
Figure 477807DEST_PATH_IMAGE019
Individual particles are selected(ii) a At the respective positions ofPosition of the particles
Figure 546760DEST_PATH_IMAGE022
Dimension of
Figure 471991DEST_PATH_IMAGE023
Because the image binarization threshold is optimized, namely a two-dimensional threshold which enables the two-dimensional local entropy to obtain the maximum is searched
Figure 227457DEST_PATH_IMAGE024
(ii) a The position of the particle is the two-dimensional threshold value, so the selection is carried out
Figure 85692DEST_PATH_IMAGE025
(ii) a The interval is
Figure 370042DEST_PATH_IMAGE026
WhereinIs a gray scale of a gray-scale image,
Figure 76147DEST_PATH_IMAGE028
(ii) a The velocity of the particle, and correspondingly the particle velocity vector, is, the dimension,that is, the interval is, here, the maximum speed is taken, that is, the speed range is
Figure 788888DEST_PATH_IMAGE029
。
2. Selecting a foam image obtained by an industrial CCD camera in the coal slime flotation production site, wherein the pixel size of the foam image is 256 multiplied by 256, and for each particle, the two-dimensional value of the position of the foam image is respectively substituted into the sum of the corresponding two-dimensional threshold values, and calculating the fitness value of each particle by applying a particle swarm algorithm, namely
(1)
Wherein the fitness of the first particle in the antibody population is represented as two-dimensional local entropy
Wherein
Figure 675122DEST_PATH_IMAGE031
，
，
Figure 808480DEST_PATH_IMAGE033
，
Figure 618490DEST_PATH_IMAGE035
，
Figure 988292DEST_PATH_IMAGE036
For joint frequency:
Figure 675625DEST_PATH_IMAGE037
FIG. 1 is a schematic diagram of a two-dimensional histogram, abscissaIs a pixel pointGray value of (d), ordinate
Figure 520587DEST_PATH_IMAGE040
Is a pixel point
Figure 531268DEST_PATH_IMAGE041
Gray value, vector at right neighborhood point
Figure 30383DEST_PATH_IMAGE042
The two-dimensional histogram matrix is divided into four regions, which are a, B, C, and D regions, by two thresholds for image segmentation.
Figure 126515DEST_PATH_IMAGE043
Representing simultaneous satisfaction of the current point gray scale of
Figure 369277DEST_PATH_IMAGE044
The gray scale of the right neighborhood point is
Figure 500044DEST_PATH_IMAGE045
All of the total pairs of gray levels satisfying the condition,，for the number of lines of the image,
Figure 3204DEST_PATH_IMAGE048
the number of columns of the image, since the pixel size of the flotation froth image is 256 x 256, here
Figure 519636DEST_PATH_IMAGE049
。
3. For each particle, determining the best position of each particle based on its fitness value
Figure 829394DEST_PATH_IMAGE050
And the current global best position
Figure 634539DEST_PATH_IMAGE051
(ii) a Each particleIs the initial position value of the particle,
Figure 58885DEST_PATH_IMAGE051
initial value is all particles
Figure 70703DEST_PATH_IMAGE053
Is measured.
4. Adjusting the speed and the position of the particles according to the formulas (2) and (3);
Figure 628723DEST_PATH_IMAGE054
(2)
(3)
wherein,
Figure 711266DEST_PATH_IMAGE056
，
Figure 893985DEST_PATH_IMAGE057
is the total number of particles in the population,
Figure 673723DEST_PATH_IMAGE058
is the velocity of the particle;
Figure 68932DEST_PATH_IMAGE059
is the best position of each particle at present;is the best position experienced by all particles in the population;is a random number between 0 and 1,
Figure 707221DEST_PATH_IMAGE062
called the inertia factor, using a linear decreasing weight strategy, i.e.
Figure 234017DEST_PATH_IMAGE063
(4)
Wherein
Figure 902896DEST_PATH_IMAGE064
And
Figure 427418DEST_PATH_IMAGE065
is that
Figure 181747DEST_PATH_IMAGE066
Taking the maximum and minimum values of
Figure 184338DEST_PATH_IMAGE067
，
Figure 238882DEST_PATH_IMAGE068
；
Figure 403147DEST_PATH_IMAGE069
And
Figure 175931DEST_PATH_IMAGE070
respectively, a current iteration number and a maximum iteration number, wherein
Figure 716634DEST_PATH_IMAGE071
Get it，
Figure 429692DEST_PATH_IMAGE073
Is the first
Figure 893034DEST_PATH_IMAGE074
The current position of the individual particles is,
Figure 237428DEST_PATH_IMAGE075
and
Figure 797722DEST_PATH_IMAGE076
is a learning factor, selects
Figure 569369DEST_PATH_IMAGE077
。
5. The new fitness for each microparticle is recalculated and updated.
6. For each particle, the best position to pass byIf the fitness is better than the fitness
Figure 933671DEST_PATH_IMAGE079
Then it is taken as the best position currently
Figure 286155DEST_PATH_IMAGE080
(ii) a Find the best
Figure 963124DEST_PATH_IMAGE081
I.e. is up to date
Figure 401059DEST_PATH_IMAGE082
Make it update the global best position。
7. Order to
Figure 622142DEST_PATH_IMAGE084
Checking whether an end condition is reached, the end condition beingTo achieve
Figure 395243DEST_PATH_IMAGE086
Or two successive generations of the population
Figure 619551DEST_PATH_IMAGE087
Is less than or equal to
Figure 274523DEST_PATH_IMAGE088
(ii) a If the end condition is not met, turning to 4; if the end condition is satisfied, the process is executed,i.e. the optimal solution, whereinFor the maximum number of iterations, choose
Figure 236980DEST_PATH_IMAGE090
，
Figure DEST_PATH_IMAGE091
。
8. Obtained by
Figure 264979DEST_PATH_IMAGE092
As a two-dimensional threshold vector
Figure DEST_PATH_IMAGE093
And performing double-threshold binary segmentation on the image, extracting a maximum value region of the binary image, and taking the result as an identification point through morphological corrosion operation.
And clustering the flotation froth images by adopting a one-dimensional histogram weighted fuzzy C-means clustering algorithm, extracting a maximum value area of the clustered images, and obtaining identification points.
9. Determining the number of classes clustered by a one-dimensional histogram weighted fuzzy C-means clustering algorithm
Figure 977720DEST_PATH_IMAGE094
Due to the fact thatThe foam image is roughly divided into three categories according to the gray level, namely the gray level of the bubble top, the bubble surface and the background, so the category of the clusters is taken
Figure DEST_PATH_IMAGE095
。
10. Determining initial values of cluster centers
Figure 964130DEST_PATH_IMAGE096
Clustering center，
Figure 598374DEST_PATH_IMAGE098
Is the number of clusters. Get
Figure DEST_PATH_IMAGE099
The initial cluster center is
Figure 695643DEST_PATH_IMAGE100
Wherein
Figure 997311DEST_PATH_IMAGE007
Is a gray-scale value of the image,
Figure DEST_PATH_IMAGE101
is the minimum value of the gray scale of the image,
Figure 154623DEST_PATH_IMAGE102
is the median value of the gray levels of the image,
Figure DEST_PATH_IMAGE103
is the maximum value of the gray scale of the image,
Figure 807321DEST_PATH_IMAGE104
，
Figure DEST_PATH_IMAGE105
take a random number of 0-1.
11. Calculating a membership matrix according to equation (5)
Figure 708281DEST_PATH_IMAGE106
；
Figure 864456DEST_PATH_IMAGE107
（5）
For the ensemble of samples to be cluster-analyzed, where each object is
Figure 661511DEST_PATH_IMAGE108
，
Figure 801505DEST_PATH_IMAGE109
Is a natural number. For the gray image with 8bits of flotation froth image, sample taking is carried out
Figure 709418DEST_PATH_IMAGE110
Is a gray scale, here
Figure 219214DEST_PATH_IMAGE112
256 samples in total; clustering center，
Figure 26950DEST_PATH_IMAGE098
For clustering the number, take
Figure 423296DEST_PATH_IMAGE114
(ii) a Get
Figure 562154DEST_PATH_IMAGE115
；
Figure 676740DEST_PATH_IMAGE116
First fingerjThe individual sample is subordinate toiMembership of each cluster, and satisfy
Figure 192035DEST_PATH_IMAGE117
，
Figure 911730DEST_PATH_IMAGE118
Is as followsiThe center of each cluster andja sample
Figure 752647DEST_PATH_IMAGE119
The euclidean distance between them.
12. Calculating an objective function from the clustered objective function equation (6)；
Figure 876777DEST_PATH_IMAGE121
(6)
WhereinThe frequency of occurrence of each gray level; let the size of the image be
Figure 728376DEST_PATH_IMAGE123
，
Figure 286396DEST_PATH_IMAGE124
(7)
In the formulaExpress gray scale asThe number of times a pixel of (a) appears in the image,
Figure 20500DEST_PATH_IMAGE127
is a gray scale, and is a gray scale,
Figure 331395DEST_PATH_IMAGE128
；
Figure 726605DEST_PATH_IMAGE129
representing the size of the image, since the pixel size of the flotation froth image is 256 x 256, here
Figure 72135DEST_PATH_IMAGE130
And satisfy
Figure 894598DEST_PATH_IMAGE131
。
13. Updating clustering centers by formula (8)
Figure 161631DEST_PATH_IMAGE132
；
Figure 891690DEST_PATH_IMAGE133
(8)
14. If it isIteration stops, fetch(ii) a Output fuzzy partition matrix
Figure 42682DEST_PATH_IMAGE136
And a cluster center
Figure 779694DEST_PATH_IMAGE137
(ii) a Otherwise make
Figure 834238DEST_PATH_IMAGE138
And returns to 11.
15. And clustering the images of the coal slime flotation foam gray level images by a fuzzy C mean value clustering algorithm of one-dimensional histogram weighting.
16. And extracting a local maximum area from the clustered images by adopting a local maximum method, and extracting identification points through morphological corrosion operation.
17. And performing superposition operation on the identification points extracted by the 8 and 16 to serve as the internal marker of the final watershed operation.

Claims (3)

1. A method for extracting an internal marker in coal slime flotation froth image segmentation is characterized by comprising the following steps: the extraction algorithm is characterized in that a particle swarm optimization algorithm and a one-dimensional histogram weighted fuzzy C-means clustering algorithm are fused, internal markers obtained by the two algorithms are superposed to serve as internal markers for image segmentation, the internal markers are used for extracting conventional marker points, capturing bright points with darker bubbles and removing false bright points.
2. The method for extracting the internal marker in the coal slime flotation froth image segmentation as claimed in claim 1, wherein: the particle swarm optimization algorithm is characterized in that a two-dimensional threshold is used as the position of a particle in the particle swarm optimization algorithm, the local entropy of the two-dimensional maximum entropy is used as a fitness function of the particle swarm optimization algorithm, the two-dimensional threshold which enables the two-dimensional local entropy to obtain the maximum is obtained through iterative updating of the algorithm and is used as a two-dimensional threshold vector, the image is subjected to double-threshold binary segmentation, the maximum area of the binary image is extracted, and the result is used as an internal identification point through morphological corrosion operation.
3. The method for extracting the internal marker in the coal slime flotation froth image segmentation as claimed in claim 1, wherein: the gray level of the coal slime flotation froth image is obtained by clustering the image through a fuzzy C-means clustering algorithm weighted by a one-dimensional histogram, clustering the coal slime flotation image through a local maximum value method for the clustered image, extracting a local maximum value area from the clustered image, and extracting identification points through morphological corrosion operation.
CN201910667822.8A 2019-07-23 2019-07-23 Method for extracting internal marker in coal slime flotation foam image segmentation Pending CN110647887A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910667822.8A CN110647887A (en) 2019-07-23 2019-07-23 Method for extracting internal marker in coal slime flotation foam image segmentation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910667822.8A CN110647887A (en) 2019-07-23 2019-07-23 Method for extracting internal marker in coal slime flotation foam image segmentation

Publications (1)

Publication Number Publication Date
CN110647887A true CN110647887A (en) 2020-01-03

Family

ID=69009427

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910667822.8A Pending CN110647887A (en) 2019-07-23 2019-07-23 Method for extracting internal marker in coal slime flotation foam image segmentation

Country Status (1)

Country Link
CN (1) CN110647887A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112200813A (en) * 2020-09-30 2021-01-08 中国矿业大学(北京) Coal and gangue identification method and system considering illumination factor
CN113763404A (en) * 2021-09-24 2021-12-07 湖南工业大学 Foam image segmentation method based on optimization mark and edge constraint watershed algorithm
CN113837193A (en) * 2021-09-23 2021-12-24 中南大学 Zinc flotation froth image segmentation algorithm based on improved U-Net network

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101036904A (en) * 2007-04-30 2007-09-19 中南大学 Flotation froth image recognition device based on machine vision and the mine concentration grade forecast method
US20150356350A1 (en) * 2014-06-05 2015-12-10 Mohamad Mustafa Awad unsupervised non-parametric multi-component image segmentation method
US10115197B1 (en) * 2017-06-06 2018-10-30 Imam Abdulrahman Bin Faisal University Apparatus and method for lesions segmentation
CN109146864A (en) * 2018-08-10 2019-01-04 华侨大学 The method that galactophore image is split based on the differential evolution algorithm of fuzzy entropy
CN109269951A (en) * 2018-09-06 2019-01-25 山西智卓电气有限公司 Floating tail-coal ash content, concentration, coarse granule detection method of content based on image

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101036904A (en) * 2007-04-30 2007-09-19 中南大学 Flotation froth image recognition device based on machine vision and the mine concentration grade forecast method
US20150356350A1 (en) * 2014-06-05 2015-12-10 Mohamad Mustafa Awad unsupervised non-parametric multi-component image segmentation method
US10115197B1 (en) * 2017-06-06 2018-10-30 Imam Abdulrahman Bin Faisal University Apparatus and method for lesions segmentation
CN109146864A (en) * 2018-08-10 2019-01-04 华侨大学 The method that galactophore image is split based on the differential evolution algorithm of fuzzy entropy
CN109269951A (en) * 2018-09-06 2019-01-25 山西智卓电气有限公司 Floating tail-coal ash content, concentration, coarse granule detection method of content based on image

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
MU-LING TIAN 等: "Improved Extraction Algorithm of Internal Markers for Forth Image Wastershed Segmentation of Coal Flotation", 《INTERNATIONAL JOURNAL OF SIMULATION》 *
TIAN MU-LING 等: "Pre-processing of Froth Image of Coal Flotation Based on Weighted Fuzzy C-Mean Clustering by One-dimensional Histogram", 《2012 INTERNATIONAL CONFERENCE ON COMPUTING, MEASUREMENT, CONTROL AND SENSOR NETWORK》 *
周开军 等: "矿物浮选泡沫图像形态特征提取方法与应用", 《中国硕士论文全文数据库信息科技辑》 *
林小竹 等: "煤泥浮选泡沫图像分割与特征提取", 《煤炭学报》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112200813A (en) * 2020-09-30 2021-01-08 中国矿业大学(北京) Coal and gangue identification method and system considering illumination factor
CN112200813B (en) * 2020-09-30 2024-02-06 中国矿业大学(北京) Coal gangue identification method and system considering illumination factors
CN113837193A (en) * 2021-09-23 2021-12-24 中南大学 Zinc flotation froth image segmentation algorithm based on improved U-Net network
CN113837193B (en) * 2021-09-23 2023-09-01 中南大学 Zinc flotation froth image segmentation method based on improved U-Net network
CN113763404A (en) * 2021-09-24 2021-12-07 湖南工业大学 Foam image segmentation method based on optimization mark and edge constraint watershed algorithm
CN113763404B (en) * 2021-09-24 2023-06-06 湖南工业大学 Foam image segmentation method based on optimization mark and edge constraint watershed algorithm

Similar Documents

Publication Publication Date Title
CN104792792B (en) A kind of road surface crack detection method of Stepwise Refinement
WO2024021413A1 (en) Image segmentation method combining super-pixels and multi-scale hierarchical feature recognition
WO2016172827A1 (en) Stepwise-refinement pavement crack detection method
CN105139395B (en) SAR image segmentation method based on small echo pond convolutional neural networks
CN110648342B (en) Foam infrared image segmentation method based on NSST significance detection and image segmentation
CN110647887A (en) Method for extracting internal marker in coal slime flotation foam image segmentation
CN108830870A (en) Satellite image high-precision field boundary extracting method based on Multi-scale model study
CN103473785B (en) A kind of fast multi-target dividing method based on three-valued image clustering
CN104794502A (en) Image processing and mode recognition technology-based rice blast spore microscopic image recognition method
CN114067114B (en) Area nuclear segmentation counting method based on algae cell morphology
CN104463199A (en) Rock fragment size classification method based on multiple features and segmentation recorrection
CN104616308A (en) Multiscale level set image segmenting method based on kernel fuzzy clustering
CN114694144B (en) Intelligent identification and rating method for non-metallic inclusions in steel based on deep learning
CN115131561A (en) Potassium salt flotation froth image segmentation method based on multi-scale feature extraction and fusion
CN109871900A (en) The recognition positioning method of apple under a kind of complex background based on image procossing
CN114170511A (en) Pavement crack disease identification method based on Cascade RCNN
CN115661072A (en) Disc rake surface defect detection method based on improved fast RCNN algorithm
CN115880594A (en) Intelligent dam crack detection method based on unmanned aerial vehicle visual perception and deep learning
Zeng et al. Steel sheet defect detection based on deep learning method
CN117853722A (en) Steel metallographic structure segmentation method integrating superpixel information
CN111091071A (en) Underground target detection method and system based on ground penetrating radar hyperbolic wave fitting
CN113077438B (en) Cell nucleus region extraction method and imaging method for multi-cell nucleus color image
CN116486273B (en) Method for extracting water body information of small sample remote sensing image
CN111524143A (en) Processing method for foam adhesion image area segmentation
CN115909072A (en) Improved YOLOv4 algorithm-based impact point water column detection method

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
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20200103