CN105761260B - A kind of skin image affected part dividing method - Google Patents

A kind of skin image affected part dividing method Download PDF

Info

Publication number
CN105761260B
CN105761260B CN201610085988.5A CN201610085988A CN105761260B CN 105761260 B CN105761260 B CN 105761260B CN 201610085988 A CN201610085988 A CN 201610085988A CN 105761260 B CN105761260 B CN 105761260B
Authority
CN
China
Prior art keywords
area
image
skin color
skin
component
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.)
Expired - Fee Related
Application number
CN201610085988.5A
Other languages
Chinese (zh)
Other versions
CN105761260A (en
Inventor
王昇
刘开华
马永涛
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 University
Original Assignee
Tianjin University
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 University filed Critical Tianjin University
Priority to CN201610085988.5A priority Critical patent/CN105761260B/en
Publication of CN105761260A publication Critical patent/CN105761260A/en
Application granted granted Critical
Publication of CN105761260B publication Critical patent/CN105761260B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30088—Skin; Dermal

Landscapes

  • Image Analysis (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

The present invention relates to a kind of skin image affected part dividing methods, including：Input skin image is pre-processed, RGB component value is extracted, image is converted into HSV color spaces；Area of skin color and non-area of skin color are detected, internal filling is carried out to area of skin color；Area of skin color H component maps to obtaining in 3) use the new area of skin color H component maps of Otsu threshold methods into row pixel value statistics with histogram；To area of skin color H component map binaryzations, Morphological scale-space.The segmentation in affected part of the present invention suitable for multiple dermatosis image has the speed of service fast, the high feature of accuracy rate.

Description

A kind of skin image affected part dividing method
Technical field
The present invention relates to field of medical image processing, more particularly to a kind of skin image affected part dividing method.
Background technology
As image processing techniques becomes the heat studied in the extensive use of medical domain, computer diagnosis and auxiliary diagnosis Door.Skin disease is a kind of common disease, since its affected part can observe directly, so can realize skin by image processing techniques The computer diagnosis and analysis of skin disease.The first step of skin disease computer diagnosis is exactly by skin disease affected part region from collecting Skin picture in split, each of affected part could further be analyzed to the accurate and efficient segmentation in affected part by only realizing Kind feature.Current main difficulty has：(1) because skin is affected by various factors, dermopathic form and color distortion It is larger；(2) often occurs heat flush region around affected part so that the interface edge of affected part and skin is not obvious, with based on ladder Spend the partitioning algorithm effect not rationality at edge；It (3) may be to segmentation if there are the regions except skin, this region in photo Generate interference.
Invention content
The object of the present invention is to provide a kind of speed of service is fast, the high skin image affected part dividing method of accuracy rate.This hair Bright technical solution is as follows：
1. a kind of skin image affected part dividing method, includes the following steps：
1) input skin image is pre-processed, extracts RGB component value, image is converted in HSV color spaces, and To the H component maps of normalized value.
2) area of skin color and non-area of skin color are detected using the skin color detection algorithm based on rgb color space, to colour of skin area Domain carries out internal filling；
3) it is zero by the H values of the corresponding H components of non-area of skin color, obtains area of skin color H component of the skin by internal filling Figure；
4) to the area of skin color H component maps that obtain in 3) into row pixel value statistics with histogram, the H component values are enabled to be in histogram The mean value that zero point ordinate value is equal to front and back ordinate value is calculated by histogram to right translation 0.6 using Otsu threshold methods The optimal threshold T of H histogram of component；
5) new area of skin color H component maps are obtained to left T+0.6 to the H component values that obtain in 3)；
6) 5) the area of skin color H component map binaryzations that will be obtained in, white area are affected part region, and black region is non-trouble Locate region, white area interior void is eliminated using the closed operation in Morphological scale-space to the image after binaryzation and keeps edge flat It is sliding；
7) closed operation treated bianry image edge, the marker edge in original image in extraction 6), and export image.
Input picture is transformed into HSV color spaces by rgb color space first and obtains normalization H components by the present invention, so It is gone using skin color detection algorithm afterwards unless area of skin color interference, calculates H histogram of component threshold values using Otsu threshold methods, make H points Amount moves to left the size of the threshold value, finally obtains accurate segmentation result with closing operation of mathematical morphology, is suitable for multiple dermatosis image In affected part segmentation, have the speed of service it is fast, the high feature of accuracy rate.
Description of the drawings
Fig. 1 is flow diagram of the present invention.
Fig. 2 is present invention specific implementation example figure.
Fig. 3 is H component exemplary plots.
Fig. 4 is Face Detection and internal filling result exemplary plot.
Fig. 5 is that H components are gone unless flesh tone portion exemplary plot.
Fig. 6 is exemplary plot after the translation of H components.
Fig. 7 is closing operation of mathematical morphology result exemplary plot.
Fig. 8 is affected part segmentation result output example image.
Specific implementation mode
In order to further illustrate the present invention, the implementation example of 1 flow diagram and attached drawing 2-8 are given with figure below in conjunction with the accompanying drawings Go out a specific example.
Fig. 2 show a typical affected skin image.Wherein there is several piece affected part region, in addition to this also two parts Background area except skin interferes.Part affected part in figure is not obvious with normal skin boundary.
Flow shown in reference block Fig. 1, the specific skin image affected part cutting procedure of the present invention are described as follows：
Step 1：Noise suppression preprocessing is carried out to input picture shown in Fig. 2.Image is transformed into HSV color spaces, is obtained To normalized H component maps I1, i.e., the value of H components is between 0 to 1.The formula of normalized H components is calculated by R, G, B component For：
Wherein max=max { R, G, B }, min=min { R, G, B }.Fig. 3 show the H component maps after normalization.
Step 2：Using in the skin color detection algorithm detection image based on rgb color space area of skin color and the non-colour of skin Region.Face Detection discrimination formula is as follows：
[yj(1)>95&&yj(2)>40&&yj(3)>20&&yj(1)-yj(3)>15&&yj(1)-yj(2)>15]
||[yj(1)>200&&yj(2)>210&&yj(1)>170&&abs(yj(1)-yj(3)]
<=[15＆＆yj(1)>yj(3)&&yj(2)>yj(3)]
Wherein yj(1), yj(2), yj(3) be respectively each image pixel R, G, B color component value.Meet above-mentioned differentiation The pixel of formula is determined as skin area, and 1 is assigned a value of in binary map, is otherwise non-skin region, is assigned a value of in binary map 0.Skin area inside is filled, output binary map I2.Fig. 4 show the binary map after Face Detection and internal filling
Step 3：From I2The position that statistical value is 0, in I1It is 0 that middle corresponding position, which enables the H values of this subregion, obtains I3, To go unless the influence of area of skin color.Fig. 5 show I3。
Step 4：To I3In H component values carry out statistics with histogram, it is 0 point ordinate value that H component values are enabled in histogram Equal to the mean value of front and back ordinate value.Due to the color character of the colour of skin and affected part, it is 0- that the distribution of H histogram of component, which concentrates on value, Between 0.35 and 0.87-1, when calculating threshold value using Otsu threshold methods, need to be adjusted histogram distribution.In view of skin Skin image is 0.6 or so without pixel distribution in H components, by histogram to right translation 0.6.H components are calculated using Otsu threshold methods The optimal threshold T of histogram.The computational methods of threshold value T are T values when seeking the maximum of following object function.
Wherein w (T) is H component values 0 to the probability of occurrence between T, the H component average values that μ (T) is threshold value when being T, μ It is the H component average values of general image.
Step 5：To I3In H component values to left T+0.6, obtain new H component maps I4.It is flat that Fig. 6 show H components Image after shifting.
Step 6：By I4Binaryzation obtains I5, white area is affected part region, and black region is non-affected part region.In order to disappear Except white area interior void and make edge-smoothing, to the image after binaryzation using the closed operation in Morphological scale-space, i.e., first Expand post-etching：
Wherein B (x) is structural element.It finally obtains bianry image and is denoted as I6.Fig. 7 show the two-value after Morphological scale-space Image.
Step 7：Extract I6Edge, the mark edge in original image, and export image.Fig. 8 show output result Image, affected part segmenting edge is in figure with green wire tag.

Claims (1)

1. a kind of skin image affected part dividing method, includes the following steps：
1) input skin image is pre-processed, extracts RGB component value, image is transformed into HSV color spaces, and is obtained The H component maps of normalized value；
2) using the skin color detection algorithm detection area of skin color and non-area of skin color based on rgb color space, to area of skin color into The internal filling of row；
3) the H values of the corresponding H components of non-area of skin color are set as zero, obtain the area of skin color H component maps by inside filling；
4) to the area of skin color H component maps that obtain in 3) into row pixel value statistics with histogram, it is zero that H component values are enabled in histogram Point ordinate value is equal to the mean value of front and back ordinate value, and by histogram to right translation 0.6, H points are calculated using Otsu threshold methods Measure the optimal threshold T of histogram；
5) new area of skin color H component maps are obtained to left T+0.6 to the H component values that obtain in 3)；
6) 5) the area of skin color H component map binaryzations that will be obtained in, white area are affected part region, and black region is non-affected part area Domain eliminates white area interior void using the closed operation in Morphological scale-space to the image after binaryzation and makes edge-smoothing；
7) closed operation treated bianry image edge, the marker edge in original image in extraction 6), and export image.
CN201610085988.5A 2016-02-15 2016-02-15 A kind of skin image affected part dividing method Expired - Fee Related CN105761260B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610085988.5A CN105761260B (en) 2016-02-15 2016-02-15 A kind of skin image affected part dividing method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610085988.5A CN105761260B (en) 2016-02-15 2016-02-15 A kind of skin image affected part dividing method

Publications (2)

Publication Number Publication Date
CN105761260A CN105761260A (en) 2016-07-13
CN105761260B true CN105761260B (en) 2018-09-28

Family

ID=56330773

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610085988.5A Expired - Fee Related CN105761260B (en) 2016-02-15 2016-02-15 A kind of skin image affected part dividing method

Country Status (1)

Country Link
CN (1) CN105761260B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3291173A1 (en) 2016-09-02 2018-03-07 Casio Computer Co., Ltd. Diagnosis assisting device, image processing method in diagnosis assisting device, and program
CN106875391A (en) * 2017-03-02 2017-06-20 深圳可思美科技有限公司 The recognition methods of skin image and electronic equipment
CN107392904A (en) * 2017-07-28 2017-11-24 陆杰 A kind of partitioning algorithm of the medical image based on mathematical morphology
CN108961295B (en) * 2018-07-27 2022-01-28 重庆师范大学 Purple soil image segmentation and extraction method based on normal distribution H threshold
CN109035289B (en) * 2018-07-27 2021-11-12 重庆师范大学 Purple soil image segmentation and extraction method based on Chebyshev inequality H threshold
CN110490844A (en) * 2019-07-24 2019-11-22 广州三得医疗科技有限公司 A kind of recognition methods, system, device and the therapeutic equipment of electromagnetic therapeutic apparatus tank print
CN110766713A (en) * 2019-10-30 2020-02-07 上海微创医疗器械（集团）有限公司 Lung image segmentation method and device and lung lesion region identification equipment
CN112037235B (en) * 2020-08-27 2023-01-10 平安科技（深圳）有限公司 Injury picture automatic auditing method and device, electronic equipment and storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102479322A (en) * 2010-11-30 2012-05-30 财团法人资讯工业策进会 System, apparatus and method for analyzing facial defect by facial image with angle
CN103632132A (en) * 2012-12-11 2014-03-12 广西工学院 Face detection and recognition method based on skin color segmentation and template matching
CN103646398A (en) * 2013-12-04 2014-03-19 山西大学 Demoscopy focus automatic segmentation method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102479322A (en) * 2010-11-30 2012-05-30 财团法人资讯工业策进会 System, apparatus and method for analyzing facial defect by facial image with angle
CN103632132A (en) * 2012-12-11 2014-03-12 广西工学院 Face detection and recognition method based on skin color segmentation and template matching
CN103646398A (en) * 2013-12-04 2014-03-19 山西大学 Demoscopy focus automatic segmentation method

Also Published As

Publication number Publication date
CN105761260A (en) 2016-07-13

Similar Documents

Publication Publication Date Title
CN105761260B (en) A kind of skin image affected part dividing method
CN107578035B (en) Human body contour extraction method based on super-pixel-multi-color space
CN105787929B (en) Skin rash point extracting method based on spot detection
CN103679173B (en) Method for detecting image salient region
CN105139383B (en) Medical image cutting method based on definition circle hsv color space
Duan et al. A WBC segmentation methord based on HSI color space
CN106096491B (en) Automatic identification method for microaneurysms in fundus color photographic image
Sarrafzadeh et al. A simple and accurate method for white blood cells segmentation using K-means algorithm
Meskini et al. A new algorithm for skin lesion border detection in dermoscopy images
CN108805865B (en) Bone marrow leukocyte positioning method based on saturation clustering
CN114596329A (en) Gas image enhancement and gas leakage detection method and system
Rosebrock et al. Quantitative analysis of TDLUs using adaptive morphological shape techniques
Mehena et al. Brain tumor segmentation and extraction of MR images based on improved watershed transform
Zhou et al. A novel adaptive local thresholding approach for segmentation of HEp-2 cell images
KR20140094975A (en) Method and Apparatus of Skin Pigmentation Detection Using Projection Transformed Block Coefficient
CN110458042B (en) Method for detecting number of probes in fluorescent CTC
Choukikar et al. Segmenting the optic disc in retinal images using thresholding
Ibrahim et al. Texture and pixel intensity characterization-based image segmentation with morphology and watershed techniques
Aramesh et al. A new method for segmentation of retinal blood vessels using Morphological image processing technique
Sanmiguel et al. Pixel-based colour contrast for abandoned and stolen object discrimination in video surveillance
Lv et al. Acquisition of fruit region in green apple image based on the combination of segmented regions
Zhou et al. Medical diagnosis algorithm based on tongue image on mobile device
Li et al. Automatic detection of leukocytes for cytometry with color decomposition
Gangwar Study on different Skin lesion Segmentation techniques and their comparisons
KR100927642B1 (en) A face edge extraction method using visual division histogram analysis

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
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20180928

Termination date: 20200215

CF01 Termination of patent right due to non-payment of annual fee