CN101334844A - Critical characteristic extraction method for flotation foam image analysis - Google Patents
Critical characteristic extraction method for flotation foam image analysis Download PDFInfo
- Publication number
- CN101334844A CN101334844A CNA2008100318095A CN200810031809A CN101334844A CN 101334844 A CN101334844 A CN 101334844A CN A2008100318095 A CNA2008100318095 A CN A2008100318095A CN 200810031809 A CN200810031809 A CN 200810031809A CN 101334844 A CN101334844 A CN 101334844A
- Authority
- CN
- China
- Prior art keywords
- area
- image
- foam
- froth
- flotation
- 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
Images
Landscapes
- Image Processing (AREA)
Abstract
The invention relates to a method used for extracting key characteristics in flotation froth image analysis. The method of the invention establishes a flotation froth image collecting platform, on the basis of acquiring froth images with an industrial camera, extracts color feature corresponding to red intensity and H intensity in RGB and HSV color spaces, combines morphologic operations and a watershed method to segment the froth images and extract size characteristics, calculates froth total reflection points to extract bearing capacity feature and adopts an image block correlation analyzing method to extract froth speed feature. The method of the invention can be used for analysis of flotation froth images and estimation of technological parameters and recovery percent in mineral floating process, thus realizing optimization of floatation production operation and reducing the waste of mineral resources.
Description
[technical field]
The present invention relates to the image processing method in the ore dressing process, be specially the critical characteristic extraction method of floatation foam image, particularly the foam characteristics of light metal flotation is extracted.
[background technology]
Flotation is most widely used a kind of beneficiation method in the mineral processing, and it relates to extremely complex physicochemical process.All the time, experienced operator regulates floating operation by observing the flotation cell foam characteristics, but since operating personnel by turns, the subjectivity of eye-observation, can not make accurate judgement to foam state, thereby cause floatation process can not be in the optimized operation state.
The flotation site environment is abominable, uneven illumination, and foam size is tiny, adhesion between the bubble, even packing phenomenon appears, and bubble edge and background area are not obvious, and this extracts for foam color and size characteristic and brings difficulty.In addition, because froth images does not have background, number of bubbles is many, flow velocity is fast and produce deformation, conventional method is difficult to be applicable to that velocity characteristic extracts.
[summary of the invention]
It is not obvious to the objective of the invention is to solve the flotation froth feature, and is difficult to the problem of quantitative description, and a kind of critical characteristic extraction method that is used for the froth images signature analysis is provided, for the graphical analysis and the index prediction of mineral floating process provides effective information.The present invention adopts video camera, light source, image pick-up card, computing machine and appurtenances construction system hardware platform thereof, obtains the flotation cell froth images thus, and extracts foam characteristics, and system software adopts the exploitation of C++ programming language.Main contents of the present invention are as follows:
At first pass through a series of hardware devices, as: computer PC, illuminator, CCD colour TV camera, image pick-up card and optical fiber make up froth images and obtain platform.The vision signal that video camera obtains, is converted to digital picture thereupon and is sent to computing machine to image pick-up card through Optical Fiber Transmission, by computing machine the froth images that collects is carried out signature analysis again and calculates.
At RGB and HSV spatial extraction foam color characteristic, the improvement conversion and the dividing ridge method that adopt h to back down reconstruct are cut apart froth images, and then according to pixel ratio extraction foaming size characteristic, follow the tracks of and correlation analysis extracts the foam velocity characteristic according to image block, utilize the foam exit point that is all-trans to extract foam bearing capacity feature at last.
By the foam characteristics that critical characteristic extraction method extracts, solved the problem that the flotation froth feature is difficult to quantitative description effectively, to floatation process parameter and index prediction the time, precision of prediction is respectively 97.5% and 95%.
[description of drawings]
Fig. 1 is a flotation foam image analysis system hardware structure synoptic diagram of the present invention;
Fig. 2 identification image of the present invention extracts flow process.
Fig. 3 froth images of the present invention is cut apart flow process.
Fig. 4 image block of the present invention is followed the tracks of and correlation analysis.
[enforcement embodiment]
Image characteristic extracting method is implemented as follows:
The foam color is extracted relative red component and at hsv color spatial extraction tone H, is represented the foam color in conjunction with two color values at the RGB color space.In the commercial Application, H represents different colors, and S span [0,1.0], correspondence never are saturated to saturated entirely (not having white); V span [0,1.0], corresponding color by dark to bright.At first extract relative red component at the RGB color space, according to formula (2) this color space conversion is become the HSV space then, extract tone H according to formula (1).
Bubble size adopts morphology opening operation and area reconstructed operation that image is carried out pre-service; Thereby using the improvement of backing down reconstruct based on h to be transformed to watershed transform provides identification point to finish cutting apart of froth images; Adopt the skeleton image of watershed algorithm mark bubble connected region, calculate the number of pixels of each connected region, obtain the size of froth images thus.
Give RGB different weights according to the importance of foam color component, obtain the weighted mean of gradation of image value, that is:
Grey=W
RR+W
GG+W
BB (3)
W wherein
R, W
G, W
BBe respectively the weights of RGB, get W
R=0.3, W
G=0.59, W
B=0.11.
Froth images f (x, tonal range y) is [a, b], through the image g of linear conversion (contrast expansion) (x, tonal range y) is [c, d], then the relation of the image after original image and the conversion is represented by formula (4):
Wherein a, b are tried to achieve by histogram, and c, d are the tonal range that is mapped to.
Top-Hat conversion and area form H top reconstructing method during combining form is learned, area of structure reconstruct H top conversion and improvement conversion thereof are for watershed transform provides identification information.The Top-Hat transformation operator is defined as:
HAT(f)=f-(fοB) (5)
" ο " expression opening operation operation in the formula, f is an original image, B is 3 * 3 ellipticity structural elements.
According to Top-hat conversion and area form H reconstruct, the area reconstruct H of definition gray level image backs down conversion:
Area threshold S=3 in the formula, utilize the antithesis characteristic can obtain corresponding area reconstruct H top closed operator, when getting parameter h=0 (f=g), area reconstruct H opening and closing operations will be reduced to area filtering opening and closing operations, and area reconstruct H top conversion is equal to the hat conversion based on area filtering.
It is as follows that (closing) conversion performing step is opened in the improvement of area reconstruct H top:
(1), opens the zone that (closing) operation filters out small size and low contrast with area reconstruct H according to designated parameter h and S.
(2) the filtering result is area reconstruct H and backs down (closing) conversion.
(3) set that all non-zero points in the transformation results are formed is exported as characteristic area.It is to the operation of image high luminance area that area reconstruct H improves on the top open transformation, and the closed operation of antithesis is to its dark region operation.
Adopt area reconstruct H top to improve open transformation and extract the watershed transform identification point, the result that floatation foam image is negated is the topology surface, and its watershed line is the bubble edge, detailed process is seen Fig. 2, wherein when asking for the topology surface of watershed transform, adopt deque's method, concrete steps are as follows:
(1) improves open transformation according to area reconstruct H top and give different numberings each zone, top among the floatation foam image counter-rotating figure.Note the gray-scale value of each edges of regions point, the line ordering of going forward side by side obtains ordered series of numbers L;
(2) minimum gradation value (L among the taking-up ordered series of numbers L
Min), get L
MinBe current search gray-scale value h, and the corresponding point in the image are joined formation K; Showing has a new seed points to participate in growth;
(3) compare L
MinWhether equal current search gray-scale value h, if equate to forward to step 2, otherwise execution in step 4;
(4) each point (K among the search queue K
p) eight neighborhood N
Kp(i), if N
Kp(i) gray scale is just given and K smaller or equal to h+1
pIdentical numbering is promptly thought and is put K
pBelong to same target,, just be decided to be frontier point if certain point is endowed different numberings simultaneously; As fruit dot K
pEight neighborhoods all numbered, put K so
pJust can be from shifting out to being listed as the K;
(5) the non-frontier point with new assignment joins formation K; Repeated execution of steps 4 adds formation K up to not new point;
(6) if the current search gray scale less than the highest gray-scale value among the figure, h=h+1 then, and return execution in step 3, otherwise finish search;
Improve the floatation foam image partitioning algorithm idiographic flow of conversion and watershed transform based on area reconstruct H top and see Fig. 3.Determine the pixel ratio by camera resolution, operating distance and visual field, can calculate the foaming size characteristic according to cut zone.
For a large amount of foams that move in floatation process, foam local deformation and cause moving with different speed adopts the correlation analysis of image block to detect the mean speed of whole froth images.The definition that foam speed is asked for principle is:
Wherein Δ t is the time interval of two continuous frames image, x
1, x
2, y
1And y
2Be respectively the foam coordinate position of Δ t front and back, u
xAnd u
yBe respectively the foam speed component.When Δ t foam displacement in enough hour also enough little, thereby the velocity limit value of measuring can well be similar to this time point foam speed.
In order to follow the tracks of same bubble area, adopt method as shown in Figure 4, at first the RGB image transitions is become gray level image, formula is: Grey=0.299 * R+0.587 * G+0.114 * B.Get the diagnostic window of one 32 * 32 pixel in gray level image arbitrarily, the adjacent bubbles in the zone has essentially identical motion vector, and bubble will keep essentially identical shape in the two continuous frames image.Respectively at T
0And T
1A diagnostic window W is respectively got at the same position place in the two continuous frames froth images constantly
1, W
2, the related coefficient of two windows of calculating, it is defined as:
F in the formula, g are respectively the intensity profile function of two windows, f
m, g
mBe respectively the average gray of two windows.The position of target is x in first frame
0, y
0, its gray-scale value is V, in second two field picture, at x
0, y
0The gray-scale value of 8 direction foams of Position Tracking, obtain the zone of maximum cross correlation coefficient value, be same foam regions.By the centre of form coordinate of two foam regions, can try to achieve the displacement of foam in the Δ t period in the time interval, and then obtain movement velocity.For avoiding the stochastic error of single-frame images, calculate the speed of 10 two field pictures continuously, the mean value of asking for these values is as final foam speed.
The foam bearing capacity, in froth images, the foam that the contains a large amount of mineral exit point that often is not all-trans, at first adopt the scanning labeling method that mutual disconnected bright spot (bubble center) in the froth images is carried out label, measure the area of each bright spot, its method is that the pixel to same numeral adds up, and obtains the pixel total area of each bright spot, and the ratio of calculating the bubble total area and froth images area according to formula (10) obtains the bubble bearing capacity then.
S in the formula
TrBe total reflection area, S
ImBe the froth images area, BL is the bubble bearing capacity.
Claims (3)
1. the critical characteristic extraction method that is used for flotation foam image analysis, it is characterized in that: video camera vertically is installed on the flotation cell top, altitude range is 80-120cm, with the horizontal range scope of video camera be 5-10cm, extract foam color, bubble size, foam speed and four key features of bearing capacity of foam, detailed process is as follows:
(1) extract relative red component at the RGB color space, and at hsv color spatial extraction H tone component, wherein extract relative red component equation and be:
Red in the formula
MeanBe red component mean value, Grey
MeanBe the mean value of gray-scale value, RGB is respectively the RGB component of froth images;
(2) according to following formula image transitions is become gray level image:
Grey=W
RR+W
GG+W
BB
W wherein
R, W
G, W
BBe respectively the weights of RGB, get W
R=0.3, W
G=0.59, W
B=0.11, image is carried out enhancement process, adopt Top-Hat conversion and area form H top reconstructing method in the morphology, area of structure reconstruct H improves conversion in the top, for watershed transform provides identification information, and then the result that floatation foam image is negated is the topology surface, and its watershed line is the bubble edge, wherein structural element is 3 * 3 ellipticity structures, and area threshold is 3;
(3) for a large amount of foams that move in floatation process, foam local deformation and cause moving with different speed, the correlation analysis of employing image block detects the mean speed of whole froth images, at first the RGB image transitions is become gray level image, formula is: Grey=0.299 * R+0.587 * G+0.114 * B, the diagnostic window size of pixel is 32 * 32 in gray level image, search for 8 directional image gray-scale values, obtain the zone of maximum cross correlation coefficient value, calculate 10 frame foam speed continuously, ask for average velocity;
(4) adopt the scanning labeling method that mutual disconnected bubble center bright spot in the froth images is carried out label, measure the area of each bright spot, pixel to same numeral adds up, obtain the pixel total area of each bright spot, the ratio of calculating reflective surface area and froth images area according to following formula obtains the bubble bearing capacity:
S in the formula
TrBe total reflection area, S
ImBe the froth images area, BL is the bubble bearing capacity.
2. the critical characteristic extraction method that is used for flotation foam image analysis according to claim 1 is characterized in that the concrete steps of described area of structure reconstruct H top improvement conversion are:
(1), filters out the zone of small size and low contrast with area reconstruct H ON operation according to designated parameter h and S;
(2) the filtering result is area reconstruct H and backs down conversion;
(3) set that all non-zero points in the transformation results are formed is exported as characteristic area.It is to the operation of image high luminance area that area reconstruct H improves on the top open transformation, and the closed operation of antithesis is to its dark region operation.
3. the critical characteristic extraction method that is used for flotation foam image analysis according to claim 1, it is characterized in that: described statistics bubble size scope is [2mm, 50mm], adopts and calculates 10 frames continuously as foam average velocity, the computing velocity scope is [20cm/s ,+20cm/s].
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2008100318095A CN101334844A (en) | 2008-07-18 | 2008-07-18 | Critical characteristic extraction method for flotation foam image analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2008100318095A CN101334844A (en) | 2008-07-18 | 2008-07-18 | Critical characteristic extraction method for flotation foam image analysis |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101334844A true CN101334844A (en) | 2008-12-31 |
Family
ID=40197433
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNA2008100318095A Pending CN101334844A (en) | 2008-07-18 | 2008-07-18 | Critical characteristic extraction method for flotation foam image analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101334844A (en) |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101620060B (en) * | 2009-08-13 | 2010-12-29 | 上海交通大学 | Automatic detection method of particle size distribution |
CN102519540A (en) * | 2011-12-10 | 2012-06-27 | 山东明佳包装检测科技有限公司 | Detecting method of beer foam compensating liquid level |
CN102680050A (en) * | 2012-04-24 | 2012-09-19 | 中南大学 | Sulfur flotation liquid level measuring method based on foam image characteristic and air volume |
CN103345636A (en) * | 2013-06-24 | 2013-10-09 | 中南大学 | Method for identifying foam working condition on copper flotation site based on wavelet multi-scale binaryzation |
CN103530653A (en) * | 2013-10-28 | 2014-01-22 | 中国矿业大学(北京) | Flotation froth tracking method and device |
CN103839057A (en) * | 2014-03-28 | 2014-06-04 | 中南大学 | Antimony floatation working condition recognition method and system |
CN104174505A (en) * | 2014-08-07 | 2014-12-03 | 玉溪矿业有限公司 | Method for quantificationally predicting theoretical ore dressing recovery rate and concentrate grade of ore floatation |
CN104408724A (en) * | 2014-11-27 | 2015-03-11 | 中南大学 | Depth information method and system for monitoring liquid level and recognizing working condition of foam flotation |
CN104657949A (en) * | 2015-02-12 | 2015-05-27 | 太原理工大学 | Method for optimizing structural elements during denoising of coal slime flotation froth image |
CN105300954A (en) * | 2015-09-02 | 2016-02-03 | 中南大学 | Method for Raman spectrum characterization of heterogeneous foam layer minerals in antimony ore flotation |
CN106228155A (en) * | 2016-10-11 | 2016-12-14 | 山东为华智能设备制造有限公司 | The technology that coal and gangue are identified by a kind of visual identity method |
CN107274403A (en) * | 2017-06-30 | 2017-10-20 | 长安大学 | A kind of evaluation method of flotation surface quality |
CN107392232A (en) * | 2017-06-23 | 2017-11-24 | 中南大学 | A kind of flotation producing condition classification method and system |
CN107480379A (en) * | 2017-08-17 | 2017-12-15 | 广东工业大学 | A kind of manufacture method of the evaluation optimization decorative panel based on picture structure similitude |
CN107705283A (en) * | 2017-06-14 | 2018-02-16 | 华北理工大学 | Particle and bubble hit detection method based on Otsu image segmentation |
CN108931621A (en) * | 2018-05-11 | 2018-12-04 | 中南大学 | A kind of zinc ore grade flexible measurement method of Kernel-based methods textural characteristics |
CN109410248A (en) * | 2018-10-23 | 2019-03-01 | 湖南科技大学 | A kind of flotation froth motion feature extracting method based on r-K algorithm |
CN109815963A (en) * | 2019-01-28 | 2019-05-28 | 东北大学 | Flotation tailing froth images target area automatic obtaining method based on sliding window technology |
CN110728676A (en) * | 2019-07-22 | 2020-01-24 | 中南大学 | Texture feature measurement method based on sliding window algorithm |
CN110728677A (en) * | 2019-07-22 | 2020-01-24 | 中南大学 | Texture roughness defining method based on sliding window algorithm |
CN111028193A (en) * | 2019-03-26 | 2020-04-17 | 桑尼环保(江苏)有限公司 | Real-time water surface data monitoring system |
CN115272327A (en) * | 2022-09-28 | 2022-11-01 | 南通西田环保科技有限公司 | Sewage multistage treatment method and system based on image treatment |
CN115861672A (en) * | 2022-12-20 | 2023-03-28 | 中南大学 | Froth flotation operation performance evaluation method based on image feature joint distribution |
CN116385455A (en) * | 2023-05-22 | 2023-07-04 | 北京科技大学 | Flotation foam image example segmentation method and device based on gradient field label |
-
2008
- 2008-07-18 CN CNA2008100318095A patent/CN101334844A/en active Pending
Cited By (40)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101620060B (en) * | 2009-08-13 | 2010-12-29 | 上海交通大学 | Automatic detection method of particle size distribution |
CN102519540A (en) * | 2011-12-10 | 2012-06-27 | 山东明佳包装检测科技有限公司 | Detecting method of beer foam compensating liquid level |
CN102680050A (en) * | 2012-04-24 | 2012-09-19 | 中南大学 | Sulfur flotation liquid level measuring method based on foam image characteristic and air volume |
CN102680050B (en) * | 2012-04-24 | 2014-03-05 | 中南大学 | Sulfur flotation liquid level measuring method based on foam image characteristic and air volume |
CN103345636A (en) * | 2013-06-24 | 2013-10-09 | 中南大学 | Method for identifying foam working condition on copper flotation site based on wavelet multi-scale binaryzation |
CN103530653A (en) * | 2013-10-28 | 2014-01-22 | 中国矿业大学(北京) | Flotation froth tracking method and device |
CN103530653B (en) * | 2013-10-28 | 2017-05-03 | 中国矿业大学(北京) | Flotation froth tracking method and device |
CN103839057A (en) * | 2014-03-28 | 2014-06-04 | 中南大学 | Antimony floatation working condition recognition method and system |
CN103839057B (en) * | 2014-03-28 | 2017-03-15 | 中南大学 | A kind of antimony flotation operating mode's switch method and system |
CN104174505A (en) * | 2014-08-07 | 2014-12-03 | 玉溪矿业有限公司 | Method for quantificationally predicting theoretical ore dressing recovery rate and concentrate grade of ore floatation |
CN104408724A (en) * | 2014-11-27 | 2015-03-11 | 中南大学 | Depth information method and system for monitoring liquid level and recognizing working condition of foam flotation |
CN104408724B (en) * | 2014-11-27 | 2017-12-01 | 中南大学 | Froth flotation level monitoring and operating mode's switch method and system based on depth information |
CN104657949A (en) * | 2015-02-12 | 2015-05-27 | 太原理工大学 | Method for optimizing structural elements during denoising of coal slime flotation froth image |
CN104657949B (en) * | 2015-02-12 | 2017-05-31 | 太原理工大学 | A kind of method of structural element optimization in coal slime flotation froth images denoising |
CN105300954B (en) * | 2015-09-02 | 2018-04-13 | 中南大学 | A kind of Raman Characterization method of the heterogeneous froth bed mineral of antimony ore flotation |
CN105300954A (en) * | 2015-09-02 | 2016-02-03 | 中南大学 | Method for Raman spectrum characterization of heterogeneous foam layer minerals in antimony ore flotation |
CN106228155A (en) * | 2016-10-11 | 2016-12-14 | 山东为华智能设备制造有限公司 | The technology that coal and gangue are identified by a kind of visual identity method |
CN107705283A (en) * | 2017-06-14 | 2018-02-16 | 华北理工大学 | Particle and bubble hit detection method based on Otsu image segmentation |
CN107705283B (en) * | 2017-06-14 | 2020-11-17 | 华北理工大学 | Particle and bubble collision detection method based on Otsu image segmentation |
CN107392232B (en) * | 2017-06-23 | 2020-09-29 | 中南大学 | Flotation working condition classification method and system |
CN107392232A (en) * | 2017-06-23 | 2017-11-24 | 中南大学 | A kind of flotation producing condition classification method and system |
CN107274403A (en) * | 2017-06-30 | 2017-10-20 | 长安大学 | A kind of evaluation method of flotation surface quality |
CN107274403B (en) * | 2017-06-30 | 2020-12-18 | 长安大学 | Evaluation method for flotation surface quality |
CN107480379A (en) * | 2017-08-17 | 2017-12-15 | 广东工业大学 | A kind of manufacture method of the evaluation optimization decorative panel based on picture structure similitude |
CN108931621A (en) * | 2018-05-11 | 2018-12-04 | 中南大学 | A kind of zinc ore grade flexible measurement method of Kernel-based methods textural characteristics |
CN108931621B (en) * | 2018-05-11 | 2020-10-02 | 中南大学 | Zinc ore grade soft measurement method based on process texture characteristics |
CN109410248B (en) * | 2018-10-23 | 2021-07-20 | 湖南科技大学 | Flotation froth motion characteristic extraction method based on r-K algorithm |
CN109410248A (en) * | 2018-10-23 | 2019-03-01 | 湖南科技大学 | A kind of flotation froth motion feature extracting method based on r-K algorithm |
CN109815963A (en) * | 2019-01-28 | 2019-05-28 | 东北大学 | Flotation tailing froth images target area automatic obtaining method based on sliding window technology |
CN111028193A (en) * | 2019-03-26 | 2020-04-17 | 桑尼环保(江苏)有限公司 | Real-time water surface data monitoring system |
CN111028193B (en) * | 2019-03-26 | 2020-09-04 | 三明市润泽环保科技有限公司 | Real-time water surface data monitoring system |
CN110728676A (en) * | 2019-07-22 | 2020-01-24 | 中南大学 | Texture feature measurement method based on sliding window algorithm |
CN110728677B (en) * | 2019-07-22 | 2021-04-02 | 中南大学 | Texture roughness defining method based on sliding window algorithm |
CN110728677A (en) * | 2019-07-22 | 2020-01-24 | 中南大学 | Texture roughness defining method based on sliding window algorithm |
CN110728676B (en) * | 2019-07-22 | 2022-03-15 | 中南大学 | Texture feature measurement method based on sliding window algorithm |
CN115272327A (en) * | 2022-09-28 | 2022-11-01 | 南通西田环保科技有限公司 | Sewage multistage treatment method and system based on image treatment |
CN115861672A (en) * | 2022-12-20 | 2023-03-28 | 中南大学 | Froth flotation operation performance evaluation method based on image feature joint distribution |
CN115861672B (en) * | 2022-12-20 | 2023-09-19 | 中南大学 | Foam flotation operation performance evaluation method based on image feature joint distribution |
CN116385455A (en) * | 2023-05-22 | 2023-07-04 | 北京科技大学 | Flotation foam image example segmentation method and device based on gradient field label |
CN116385455B (en) * | 2023-05-22 | 2024-01-26 | 北京科技大学 | Flotation foam image example segmentation method and device based on gradient field label |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101334844A (en) | Critical characteristic extraction method for flotation foam image analysis | |
CN106651872B (en) | Pavement crack identification method and system based on Prewitt operator | |
CN109902600B (en) | Road area detection method | |
CN100544446C (en) | The real time movement detection method that is used for video monitoring | |
CN106056155B (en) | Superpixel segmentation method based on boundary information fusion | |
CN101334837B (en) | Multi-method integrated license plate image positioning method | |
CN109583425A (en) | A kind of integrated recognition methods of the remote sensing images ship based on deep learning | |
CN101430195B (en) | Method for computing electric power line ice-covering thickness by using video image processing technology | |
CN103646400B (en) | Multi-scale segmentation parameter automatic selecting method in object-oriented remote sensing images analysis | |
CN103630496B (en) | Based on the traffic video visibility detecting method of road surface apparent brightness and least square method | |
CN102044151A (en) | Night vehicle video detection method based on illumination visibility identification | |
CN101527043B (en) | Video picture segmentation method based on moving target outline information | |
CN110060508B (en) | Automatic ship detection method for inland river bridge area | |
US20170132764A1 (en) | Image inpainting system and method for using the same | |
CN102073852B (en) | Multiple vehicle segmentation method based on optimum threshold values and random labeling method for multiple vehicles | |
CN103903278A (en) | Moving target detection and tracking system | |
CN105608429B (en) | Robust method for detecting lane lines based on difference excitation | |
CN105654091A (en) | Detection method and apparatus for sea-surface target | |
CN102663357A (en) | Color characteristic-based detection algorithm for stall at parking lot | |
CN104156731A (en) | License plate recognition system based on artificial neural network and method | |
CN105069801A (en) | Method for preprocessing video image based on image quality diagnosis | |
CN101216943B (en) | A method for video moving object subdivision | |
CN109255350A (en) | A kind of new energy detection method of license plate based on video monitoring | |
CN110210451A (en) | A kind of zebra line detecting method | |
CN112287838B (en) | Cloud and fog automatic identification method and system based on static meteorological satellite image sequence |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20081231 |