CN103955940B - A kind of detection method of the human body cache based on X ray backscatter images - Google Patents

A kind of detection method of the human body cache based on X ray backscatter images Download PDF

Info

Publication number
CN103955940B
CN103955940B CN201410209538.3A CN201410209538A CN103955940B CN 103955940 B CN103955940 B CN 103955940B CN 201410209538 A CN201410209538 A CN 201410209538A CN 103955940 B CN103955940 B CN 103955940B
Authority
CN
China
Prior art keywords
threshold
image
carried out
msup
mrow
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201410209538.3A
Other languages
Chinese (zh)
Other versions
CN103955940A (en
Inventor
戴维迪
王诗瑶
何吉元
贾重
王玉川
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.)
TIANJIN CHONGFANG TECHNOLOGY Co Ltd
Original Assignee
TIANJIN CHONGFANG 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 TIANJIN CHONGFANG TECHNOLOGY Co Ltd filed Critical TIANJIN CHONGFANG TECHNOLOGY Co Ltd
Priority to CN201410209538.3A priority Critical patent/CN103955940B/en
Publication of CN103955940A publication Critical patent/CN103955940A/en
Application granted granted Critical
Publication of CN103955940B publication Critical patent/CN103955940B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Apparatus For Radiation Diagnosis (AREA)
  • Image Analysis (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)
  • Image Processing (AREA)

Abstract

The invention discloses a kind of detection method of the human body cache based on X ray backscatter images, it is related to computer image processing technology field, it is characterised in that：Comprise the following steps：X ray backscatter images are gathered, the image obtained is pre-processed；The threshold value calculation method that global threshold is combined with dynamic threshold is carried out to the image after pretreatment, and the gray scale discontinuity to caused by dynamic threshold is handled using smoothing technique；Image segmentation is carried out using above-mentioned obtained threshold value, and by the use of fuzziness, region inner homogeneous and these three evaluation indexes of interregional contrast as feedback, adaptively obtains the best dynamic threshold of segmentation effect and finally obtains preferable segmentation effect.The cache being partitioned into is extracted using connected component labeling method on this basis.This method has the characteristics of operating efficiency is high, and testing result is accurate.

Description

A kind of detection method of the human body cache based on X ray backscatter images
Technical field
The present invention relates to computer image processing technology field, more particularly to a kind of based on X ray backscatter images The detection method of human body cache.
Background technology
More and more deep with the research of the detection for dangerous matter sources, assessment and emergency plan both at home and abroad, the X ray back of the body dissipates Penetrate imaging technique and have been used for field of safety check so as to carry out the detection that human body carries cache.Wherein for the image of scan image Segmentation is the most basic link of X ray backscatter images human body detection.It is aimed at from X ray back scattering scan image Cache is extracted from human body.Automatically segmentation carries complete effective human body cache for the feature of later stage cache Take, Classification and Identification and harmful grade differentiate etc. play an important roll.The scan image that the said equipment is obtained at present relies primarily on Inspection personnel carries out artificial interpretation, very few for the dangerous material Automatic Measurement Technique achievement in research of such image, in addition, by The shadow of the factor such as position, clothing materials, human posture and illumination when X ray back scattering imaging by object on human body Ring so that fast and accurately human body cache segmentation difficulty increase.Therefore realize computer to human body in X ray backscatter images The automatic detection of hiding article, and then identify all kinds of dangerous material, so as to avoid expending a large amount of manpowers and the time have it is particularly significant Research Significance.
The content of the invention
The technical problem to be solved in the present invention is：A kind of inspection of the human body cache based on X ray backscatter images is provided Survey method.
The present invention is adopted the technical scheme that to solve technical problem present in known technology：
A kind of detection method of the human body cache based on X ray backscatter images, comprises the following steps：
X ray backscatter images are pre-processed, obtain pretreatment image by S101, collection X ray backscatter images；
S102, the threshold calculations that global threshold is combined with dynamic threshold are carried out to pretreatment image, and utilize smooth skill Art gray scale discontinuity to caused by dynamic threshold is handled；Detailed process is：
S1021, global threshold calculating, acquisition image overall threshold value t1 are carried out to pretreatment image using otsu algorithms；
S1022, pretreatment image is divided into a series of subgraphs, and otsu local thresholds are carried out to each frame subgraph Calculate, obtain the local threshold t2 of each frame subgraph, Taxonomic discussion is then carried out according to the inter-class variance under the local threshold： If inter-class variance is less than the threshold constant threshold of user's self-defining, the global threshold obtained in S1021 is utilized T1, according to formula
T=(1- α) t1+ α t2 obtain new threshold value t；Wherein, α is inter-class variance；Otherwise：New threshold value t=t2；
S1023, new threshold value t is stored in matrix M, matrix M is smoothed, is specially：
Each new threshold value t in matrix M is weighted with each threshold element present in 4 neighborhoods around it and is added, and New threshold value t is replaced with this；The weights in wherein 4 fields are respectively 0.1；New threshold matrix M2 is obtained after the completion of smooth work；
S103, image segmentation, and utilization fuzziness, region inner homogeneous and region are carried out using above-mentioned obtained threshold value Between contrast these three evaluation indexes as feedback, adaptively obtain the best dynamic threshold of segmentation effect and complete image two-value Change；Detailed process is：
S1031, using the threshold element in threshold matrix M2 to pretreatment image carry out binarization segmentation processing；
S1032, using fuzziness, region inner homogeneous and interregional contrast to above-mentioned binarization segmentation processing after Segmentation result carries out overall merit；
S1033, step S1022, S1023, S1031, S1032 are repeated in, until obtaining evaluation of estimate highest segmentation feelings Condition；
S1034, the threshold calculations that global threshold is combined with dynamic threshold are carried out using the segmentation situation of above-mentioned acquisition, it is complete Into image binaryzation；
S104, connected component labeling is carried out to the binary image obtained, the number of pixels included using object is existed This condition obtains the extraction of final cache as screening in a certain scope.
Further：Pretreatment in the step S101 is specially：
S1011, image gray processing
Entered using the rgb pixel value of DIB structure extraction X ray backscatter images, and by three components with different weights Row weighted average, wherein red pixel weight are 30%, green pixel 59%, and blue pixel is 11%；
S1012, image denoising
Denoising is carried out to image using gaussian filtering, the concrete operations of gaussian filtering are：With a template scanning figure Each pixel as in, go to substitute convolution central pixel point with the weighted average gray value of pixel in the neighborhood of template determination Value；The position of the centre of neighbourhood is more proximate in Gaussian template, its weights is higher；The meaning for being arranged weights is to image While details carries out smooth, it can more retain the overall gray distribution features of image；Above-mentioned template size is 3*3, mould Coordinate is for the numerical computational formulas of (x, y) in plate：
Wherein：σ is scale parameter, for determining the smoothness of filtering.Using template center as origin, x, y are that template is each Position relative to origin coordinate；Gaussian template is built according to above formula, and denoising is carried out to image.
Further：The step S104 is specially：
S1041, connected component labeling is carried out to the binary image obtained, obtain multiple connected regions；
S1042, calculate the pixel number that each connected region is included；
S1043, cross more than or less than user-defined threshold value when pixel number, then it is assumed that be unlikely to be cache from And filter out final cache
The present invention has the advantages and positive effects of：
The present invention realizes the automatic detection for the cache that computer carries to human body in X ray backscatter images, allows work Personnel free from heavy visual tasks, and improve the accuracy of detection to improve safety check speed and quality, the party Method can reach following effect：Solve existing global OTSU algorithms and can not take image local details into account to cause segmentation effect not The problem of good, dynamic threshold is adaptively calculated using overall merit, is combined with global threshold so as to complete to split.And for Dynamic threshold according to actual conditions smoothly so as to reach to improve partitioning algorithm and complete the hiding analyte detection of human body know the real situation；Together When this method advantage be when image target object and background gray scale relatively or background and object grey scale change compared with Local detail can be taken into full account so as to obtain satisfied segmentation result when big.
Brief description of the drawings
Fig. 1 is the flow chart of the detection method of the human body cache based on X ray backscatter images in the present invention；
Fig. 2 is the first preferred embodiments design sketch of the human body cache based on X ray backscatter images in the present invention；
Fig. 3 is the second preferred embodiments design sketch of the human body cache based on X ray backscatter images in the present invention.
Embodiment
In order to further understand the content, features and effects of the present invention, enumerating following examples herein, and coordinate attached Figure describes in detail as follows：
Referring to Fig. 1, a kind of detection method of the human body cache based on X ray backscatter images, by adaptively Dynamic threshold calculates with global threshold and completes segmentation task, and detailed step is：
S101：X ray backscatter images are gathered, X ray backscatter images are pre-processed, obtain pretreatment image；
1) image gray processing
This specific embodiment use DIB structure extraction X ray backscatter images rgb pixel value, and by three components with Different weights are weighted averagely, and wherein red pixel weight is 30%, green pixel 59%, and blue pixel is 11%；
2) image denoising
This specific embodiment carries out denoising using gaussian filtering to image.The concrete operations of gaussian filtering are：With one Each pixel in individual template scan image, go to substitute convolution with the weighted average gray value of pixel in the neighborhood of template determination The value of central pixel point.The simple smooth processing of image is different from, the position of the centre of neighbourhood is more proximate in Gaussian template, it is weighed Value is higher.Be arranged weights meaning be to image detail carry out it is smooth while, it is total can more to retain image The gray distribution features of body.For 3*3, the calculation formula of template is the template size that this specific embodiment uses：
Gaussian template is built according to above formula, and denoising is carried out to image.
S102：The threshold calculations that global threshold is combined with dynamic threshold are carried out to the image after pretreatment, and are utilized Smoothing technique gray scale discontinuity to caused by dynamic threshold is handled：
1) global threshold calculates
During sheet, global threshold calculating is carried out using otsu algorithms.This method is also referred to as maximum variance between clusters.It is right In image I (x, y), the segmentation threshold of foreground and background is denoted as t1, belong to prospect pixel number account for entire image ratio note For ω0, its average gray μ0；The ratio that background pixel points account for entire image is ω1, its average gray is μ1, image it is total flat Equal gray scale is designated as μ, and inter-class variance is designated as g.
Assuming that the size of image is M × N, number of pixels of the gray value of pixel less than threshold value T is denoted as N0, pixel in image Number of pixels of the gray scale more than threshold value T is denoted as N1, then has：
ω0=N0/M × N (1)
ω1=N1/M × N (2)
N0+N1=M × N (3)
ω0+ω1=1 (4)
μ=ω0*μ0+ω1*μ1 (5)
G=ω0(μ0-μ)^2+ω1(μ1-μ)^2 (6)
Formula (5) is substituted into formula (6), obtains equivalence formula:
G=ω0ω1(μ0-μ1)^2 (7)
This is inter-class variance
The threshold value for making inter-class variance g maximum is obtained using the method for traversal, is required.
2) dynamic threshold calculates
Piecemeal is carried out to image and obtains a series of subgraphs, and local threshold is carried out to each frame subgraph office is calculated Portion threshold value t2Otsu algorithms obtain threshold value in (each frame subgraph uses 1)), and divided according to the inter-class variance under the threshold value Class discussion：If inter-class variance σ is less than threshold value threshold, 1) the middle global threshold t obtained is utilized1, according to t=(1- α) t1+ αt2Principle is weighted, and tries to achieve each piece of new threshold value t, otherwise new threshold value t=t2。
3) threshold smoothing
By in the new threshold value deposit matrix M of 2) each subgraph of gained, matrix M is smoothed：Will be each new in M Threshold element is weighted with each threshold element present in 4 neighborhoods around it and is added, and replaces original threshold value with this. The weights in wherein 4 fields are respectively 0.1.New threshold matrix M2 is obtained after the completion of smooth work.
S103：Image segmentation is carried out using above-mentioned obtained threshold value, and utilizes fuzziness, region inner homogeneous and region Between contrast these three evaluation indexes as feedback, adaptively obtain the best dynamic threshold of segmentation effect and complete image two-value Change.
1) thresholding segmentation index calculates
Tied using fuzziness, region inner homogeneous and interregional contrast to splitting corresponding to piecemeal situation in process 2 Fruit carries out overall merit.
Wherein fuzziness refers to image transforming from a spatial domain to fuzzy quality domain, and point of image is weighed using fuzzy measurement Cut FUZZY MAPPING in the effect present invention use a kind of Nonlinear Mapping model for：
Wherein XijFor the gray value at (i, j) place in original image, UijFor the pixel fuzzy quality domain is mapped to from spatial domain Institute
Corresponding fuzzy value；α is fuzzy factor, 0≤α≤1, t=Xij/Xmax。
Nonlinear function
G (t)=(1-e-t)/(1+e-t)
In addition, interregional contrast GLC and gradation uniformity UM calculation formula are as follows：
Wherein, f1And f2Represent the average gray value in adjacent two region
Gradation uniformity
UM value is bigger, and the uniformity characterized in segmentation figure inside each region is better, that is, it is higher to split quality.Wherein C is Normalization factor, RiIt is i-th piece of cut zone, f (x, y) is the gray value of pixel (x, y), AiIt is region RiPixel count.
2) block count is iterated to calculate
2) -3 in repeating 102) and process three in 1) step, obtain evaluation of estimate highest piecemeal situation.The present invention is implemented Be applied in example method etc. be known technology in data processing method, the embodiment of the present invention will not be described here.
S104：Connected component labeling is carried out to the binary image obtained, the number of pixels included using object is existed This condition obtains the extraction of final cache as screening in a certain scope.
Fig. 2 and Fig. 3 are referred to, wherein, Fig. 2 a and 3a are X ray backscatter images, and Fig. 2 b and 3b have this hair in being The binary map after the Threshold segmentation accessed by method in bright；Fig. 2 c and 3c are the image after cache extraction.
Embodiments of the invention are described in detail above, but the content is only presently preferred embodiments of the present invention, It is not to be regarded as the practical range for limiting the present invention.All equivalent changes made according to the present patent application scope and improvement, Should still it belong within the patent covering scope of the present invention.

Claims (3)

  1. A kind of 1. detection method of the human body cache based on X ray backscatter images, it is characterised in that：Comprise the following steps：
    X ray backscatter images are pre-processed, obtain pretreatment image by S101, collection X ray backscatter images；
    S102, the threshold calculations that global threshold is combined with dynamic threshold are carried out to pretreatment image, and utilize smoothing technique pair Gray scale discontinuity is handled caused by dynamic threshold；Detailed process is：
    S1021, global threshold calculating, acquisition image overall threshold value t1 are carried out to pretreatment image using otsu algorithms；
    S1022, pretreatment image is divided into a series of subgraphs, and otsu local threshold meters are carried out to each frame subgraph Calculate, obtain the local threshold t2 of each frame subgraph, Taxonomic discussion is then carried out according to the inter-class variance under the local threshold：If Inter-class variance is less than the threshold constant threshold of user's self-defining, then using the global threshold t1 obtained in S1021, According to formula
    T=(1- α) t1+ α t2 obtain new threshold value t；Wherein, α is inter-class variance；Otherwise：New threshold value t=t2；
    S1023, new threshold value t is stored in matrix M, matrix M is smoothed, is specially：
    Each new threshold value t in matrix M is weighted with each threshold element present in 4 neighborhoods around it and is added, and with this To replace new threshold value t；The weights in wherein 4 fields are respectively 0.1；New threshold matrix M2 is obtained after the completion of smooth work；
    S103, carry out image segmentation using above-mentioned obtained threshold value, and utilize fuzziness, region inner homogeneous and interregional right Than these three evaluation indexes of degree as feedback, adaptively obtain the best dynamic threshold of segmentation effect and complete image binaryzation； Detailed process is：
    S1031, using the threshold element in threshold matrix M2 to pretreatment image carry out binarization segmentation processing；
    S1032, using fuzziness, region inner homogeneous and interregional contrast to above-mentioned binarization segmentation processing after segmentation As a result overall merit is carried out；
    S1033, step S1022, S1023, S1031, S1032 are repeated in, until obtaining evaluation of estimate highest segmentation situation；I.e. The threshold calculations being combined using the segmentation situation progress global threshold of above-mentioned acquisition with dynamic threshold, complete image binaryzation；
    S104, connected component labeling is carried out to the binary image obtained, the number of pixels included using object is a certain In the range of this condition as screening conditions, obtain the cache finally extracted.
  2. 2. detection method according to claim 1, it is characterised in that：Pretreatment in the step S101 is specially：
    S1011, image gray processing
    Added using the rgb pixel value of DIB structure extraction X ray backscatter images, and by three components with different weights Weight average, wherein red pixel weight are 30%, green pixel 59%, and blue pixel is 11%；
    S1012, image denoising
    Denoising is carried out to image using gaussian filtering, the concrete operations of gaussian filtering are：With in a template scan image Each pixel, with template determine neighborhood in pixel weighted average gray value go substitute convolution central pixel point value； The position of the centre of neighbourhood is more proximate in Gaussian template, its weights is higher；Be arranged weights meaning be it is thin to image While section carries out smooth, it can more retain the overall gray distribution features of image；Above-mentioned template size is 3*3, template Middle coordinate is that the numerical computational formulas of (x, y) is：
    <mrow> <mi>G</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>;</mo> <mi>y</mi> <mo>,</mo> <mi>&amp;sigma;</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <mn>2</mn> <msup> <mi>&amp;pi;&amp;sigma;</mi> <mn>2</mn> </msup> </mrow> </mfrac> <msup> <mi>exp</mi> <mrow> <mo>-</mo> <mrow> <mo>(</mo> <msup> <mi>x</mi> <mn>2</mn> </msup> <mo>+</mo> <msup> <mi>y</mi> <mn>2</mn> </msup> <mo>)</mo> </mrow> <mo>/</mo> <mn>2</mn> <msup> <mi>&amp;sigma;</mi> <mn>2</mn> </msup> </mrow> </msup> </mrow>
    Wherein：σ is scale parameter, and for determining the smoothness of filtering, using template center as origin, x, y are template each position Relative to the coordinate of origin；Gaussian template is built according to above formula, and denoising is carried out to image.
  3. 3. detection method according to claim 1 or 2, it is characterised in that：The step S104 is specially：
    S1041, connected component labeling is carried out to the binary image obtained, obtain multiple connected regions；
    S1042, calculate the pixel number that each connected region is included；
    S1043, it is more than or less than user-defined threshold value when the pixel number that connected region includes, then it is assumed that be unlikely to be Cache is so as to filtering out final cache.
CN201410209538.3A 2014-05-16 2014-05-16 A kind of detection method of the human body cache based on X ray backscatter images Active CN103955940B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410209538.3A CN103955940B (en) 2014-05-16 2014-05-16 A kind of detection method of the human body cache based on X ray backscatter images

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410209538.3A CN103955940B (en) 2014-05-16 2014-05-16 A kind of detection method of the human body cache based on X ray backscatter images

Publications (2)

Publication Number Publication Date
CN103955940A CN103955940A (en) 2014-07-30
CN103955940B true CN103955940B (en) 2018-01-16

Family

ID=51333209

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410209538.3A Active CN103955940B (en) 2014-05-16 2014-05-16 A kind of detection method of the human body cache based on X ray backscatter images

Country Status (1)

Country Link
CN (1) CN103955940B (en)

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2530252B (en) * 2014-09-10 2020-04-01 Smiths Heimann Sas Determination of a degree of homogeneity in images
JP6588102B2 (en) * 2015-03-20 2019-10-09 ベンタナ メディカル システムズ， インコーポレイテッド System and method for image segmentation
CN105678773B (en) * 2016-01-12 2018-10-26 西北工业大学 A kind of soft image dividing method
CN106023166B (en) * 2016-05-12 2018-06-29 深圳市太赫兹科技创新研究院 The detection method and device of dangerous object hidden by human body in microwave image
CN106920398A (en) * 2017-04-27 2017-07-04 深圳大图科创技术开发有限公司 A kind of intelligent vehicle license plate recognition system
CN107316318B (en) * 2017-05-26 2020-06-02 河北汉光重工有限责任公司 Air target automatic detection method based on multi-subregion background fitting
CN110544227A (en) * 2018-05-29 2019-12-06 中国科学院电子学研究所 Passive terahertz human body security inspection image target detection method
CN110335666A (en) * 2019-05-22 2019-10-15 平安国际智慧城市科技股份有限公司 Medical image appraisal procedure, device, computer equipment and storage medium
CN111182173B (en) * 2019-11-27 2021-10-01 绍兴柯桥浙工大创新研究院发展有限公司 Image transmission processing method and system
CN112001333A (en) * 2020-08-27 2020-11-27 中广核贝谷科技有限公司 Intelligent identification method based on container X-ray image
CN112308823B (en) * 2020-10-14 2024-03-15 杭州三坛医疗科技有限公司 Method and device for positioning region of interest in medical image
CN114638851B (en) * 2022-05-17 2022-09-27 广州优刻谷科技有限公司 Image segmentation method, system and storage medium based on generation countermeasure network

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102542570A (en) * 2011-12-30 2012-07-04 北京华航无线电测量研究所 Method for automatically detecting dangerous object hidden by human body in microwave image
CN102956035A (en) * 2011-08-25 2013-03-06 深圳市蓝韵实业有限公司 Preprocessing method and preprocessing system used for extracting breast regions in mammographic images
CN103279960A (en) * 2013-06-18 2013-09-04 天津大学 Human body hidden thing image segmentation method based on X-ray back scattering image

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1213592C (en) * 2001-07-31 2005-08-03 佳能株式会社 Adaptive two-valued image processing method and equipment

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102956035A (en) * 2011-08-25 2013-03-06 深圳市蓝韵实业有限公司 Preprocessing method and preprocessing system used for extracting breast regions in mammographic images
CN102542570A (en) * 2011-12-30 2012-07-04 北京华航无线电测量研究所 Method for automatically detecting dangerous object hidden by human body in microwave image
CN103279960A (en) * 2013-06-18 2013-09-04 天津大学 Human body hidden thing image segmentation method based on X-ray back scattering image

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
《Computerized Detection of Masses;Bin Zheng等;《Acad Radiol 1995》;19951130;第2卷(第11期);全文 *
《一种全局和局部相结合的二值化方法研究》;孟立娜等;《计算机技术与发展》;20121130;第22卷(第11期);全文 *
《机械工件数字识别的二值化算法研究》;高潮等;《光电工程》;20100630;第37卷(第6期);全文 *

Also Published As

Publication number Publication date
CN103955940A (en) 2014-07-30

Similar Documents

Publication Publication Date Title
CN103955940B (en) A kind of detection method of the human body cache based on X ray backscatter images
CN108154168B (en) Comprehensive cargo inspection system and method
Zou et al. CrackTree: Automatic crack detection from pavement images
Sommer et al. Learning-based mitotic cell detection in histopathological images
CN109472822A (en) Dimension of object measurement method based on depth image processing
CN109447065A (en) A kind of method and device of breast image identification
CN106780485A (en) SAR image change detection based on super-pixel segmentation and feature learning
WO2017010397A1 (en) Image analysis device, image analysis method, image analysis system, image analysis program, and recording medium
Tan et al. Segmentation and counting algorithm for touching hybrid rice grains
CN110148130A (en) Method and apparatus for detecting part defect
CN104282008B (en) The method and apparatus that Texture Segmentation is carried out to image
CN109738137A (en) Earth and rockfill dam leakage real-time monitoring and fast diagnosis method based on image comparison
CN109363699A (en) A kind of method and device of breast image lesion identification
CN105069818A (en) Image-analysis-based skin pore identification method
CN109191424A (en) A kind of detection of breast lump and categorizing system, computer readable storage medium
CN103218832A (en) Visual saliency algorithm based on overall color contrast ratio and space distribution in image
CN107229910A (en) A kind of remote sensing images icing lake detection method and its system
Wang et al. A hybrid method for the segmentation of a ferrograph image using marker-controlled watershed and grey clustering
CN110021023A (en) A kind of electronics cloth defect segmentation method
CN109363697A (en) A kind of method and device of breast image lesion identification
Ojeda-Magaña et al. Identification of pore spaces in 3D CT soil images using PFCM partitional clustering
WO2022267270A1 (en) Crack characteristic representation method and system based on multi-fractal spectrum
CN105046265B (en) Iris image intestines ring region area detecting method based on texture difference
CN103279960B (en) A kind of image partition method of human body cache based on X-ray backscatter images
CN105488798B (en) SAR image method for measuring similarity based on point set contrast

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
PP01 Preservation of patent right

Effective date of registration: 20181012

Granted publication date: 20180116

PP01 Preservation of patent right