CN104899584A - Stained section identification method based on fuzzy thought - Google Patents
Stained section identification method based on fuzzy thought Download PDFInfo
- Publication number
- CN104899584A CN104899584A CN201510381236.9A CN201510381236A CN104899584A CN 104899584 A CN104899584 A CN 104899584A CN 201510381236 A CN201510381236 A CN 201510381236A CN 104899584 A CN104899584 A CN 104899584A
- Authority
- CN
- China
- Prior art keywords
- stained
- gray
- scale map
- recognition methods
- image
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a stained section identification method based on fuzzy thought, which is suitable for separating tissues from a background in a section preview. The method comprises the steps of preparing a preview of the stained section, converting the preview to be a grey-scale image, wherein the grey-scale image includes the tissues and the background image; dividing the grey-scale image into a plurality of small blocks, judging whether each small block belongs to the tissue or the background image according to a standard that the grey value of all pixels in one small block is identical, carrying out the binarization for the small block, and obtaining a block image; dynamically processing the block image, corroding, then swelling, filtering independent impurity small blocks, and obtaining a final result image. According to the technical scheme, the tissues are separated from the background, so that a digital section scanning system has better adaptability to the sections of different types.
Description
Technical field
The invention belongs to image processing field, relate to a kind of stained recognition methods, particularly relate to a kind of stained recognition methods based on obscure idea.
Background technology
In the last few years, the raising required medical skill along with society, various new Medical Devices emerged in an endless stream, and the development of digital slices scanner also becomes a wherein considerable ring.Digital slices scanning system can, by microslide perfect information, comprehensive rapid scanning, make the microslide of conventional matter become new-generation digital pathological section, is realize epoch-making change to pathological diagnosis technology.
In scanning system, an important step is to section preview, and utilizes image processing techniques by the tissue identification in preview graph out, with background separation, to carry out subsequent treatment.Traditional recognition methods is by laggard for preview graph gray processing row threshold division, realizes tissue and is separated with the binaryzation of background.
But the tissue in common section can effectively identify by traditional recognition methods, but universality is not high, once run into the tissue of color lighter (tissue color is close with background color), algorithm is difficult to good dividing tissue and background.This is because after gray processing, tissue is very close with the gray-scale value of background, is difficult to determine that effective threshold value goes dividing tissue and background, causes occurring in result that tissue identification is incomplete, and the situation that a large amount of impurity is mistaken as tissue occurs.
Summary of the invention
In view of this, the invention provides a kind of stained recognition methods based on obscure idea, the tissue in stained is effectively separated with background, retain as much as possible and organize and remove impurity.
For achieving the above object, concrete technical scheme is as follows:
Based on a stained recognition methods for obscure idea, be applicable to, by the tissue in section preview graph and background separation, comprise the following steps:
Step 1, the preview graph of preparation stained, and transferred to gray-scale map, described gray-scale map comprises tissue and background image;
Step 2, is divided into several fritters by gray-scale map image, and the standard identical according to all grey scale pixel values in a fritter judges that fritter belongs to tissue or background image, by fritter binaryzation, obtains block figure;
Step 3, carries out Morphological scale-space by block figure, first corrodes and expands afterwards, independently will filter by impurity fritter, and obtain net result figure.
Preferably, the threshold value of the binaryzation in described step 2 is the average of gray-scale map.
Preferably, the square tiles of to be several length of sides the be a of the fritter in described step 2.
Preferably, the stained in described step 1 comprises common dyeing section or light stained.
Preferably, described step 1 comprises:
Step 1.1, prepares the preview graph of light stained and blank section, and subtracts each other after transferring both to gray-scale map, obtain the gray-scale map subtracted each other;
All grey scale pixel values of the gray-scale map subtracted each other are multiplied by coefficient μ, obtain final gray-scale map by step 1.2;
In step 2, final gray-scale map image is divided into several fritters.
Preferably, in described step 1.2, coefficient μ determines according to the grey level histogram of the gray-scale map subtracted each other.
Preferably, described method is applicable in digital slices scanning system.
Relative to prior art, technical scheme of the present invention solves the problem of the light stained poor effect of common algorithm identification, pass through obscure idea, well tissue is separated with background, positive effect is produced to digital slices scanning system, makes it have better adaptability to dissimilar section.
Accompanying drawing explanation
The accompanying drawing forming a part of the present invention is used to provide a further understanding of the present invention, and schematic description and description of the present invention, for explaining the present invention, does not form inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 is the schematic flow sheet of the embodiment of the present invention;
Fig. 2 is the preview graph of the light stained of the embodiment of the present invention;
Fig. 3 is the preview graph of the blank section of the embodiment of the present invention;
Fig. 4 is the gray-scale map subtracted each other of the embodiment of the present invention;
Fig. 5 is the final gray-scale map of the embodiment of the present invention;
Fig. 6 is the block figure of the embodiment of the present invention;
Fig. 7 is the result figure of the embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
It should be noted that, when not conflicting, the embodiment in the present invention and the feature in embodiment can combine mutually.
Below with reference to accompanying drawing, concrete explaination is done to embodiments of the invention.
A kind of stained recognition methods based on obscure idea of embodiments of the invention as shown in Figure 1, concrete steps are:
Step 1.1, background subtraction.As shown in Figures 2 and 3, if directly preview graph is not made any pre-service Direct Recognition, poor effect, therefore first prepare a blank section preview graph to scheme as a setting, subtract each other after transferring two figure to gray-scale map, the gray-scale map subtracted each other as shown in Figure 4, is designated as sub_img, and at this moment tissue does not more deal with the difference of background and improves a lot.
Step 1.2, grey level enhancement.As shown in Figure 5, because light stained tissue color is shallow, in order to obtain better segmentation effect, all grey scale pixel values of sub_img are multiplied by a coefficient μ, obtain final gray-scale map, be designated as ex_img, to strengthen the gray scale difference between tissue and background, the value of μ needs the grey level histogram according to sub_img and determines, excessive too small be all unfavorable for after process.
Step 2, fritter binaryzation.
Core concept of the present invention is: the difference of machine and human eye is, tiny difference more easily discovered by machine, and these differences are often ignored by human eye.The sensitive of machine sometimes can have a negative impact to result on the contrary, and the identification of light stained tissue is exactly an example.In light stained, although tissue color is light, human eye very naturally can regard as tissue the place of joining together, and for machine, machine can judge that each pixel belongs to tissue or background successively, is therefore difficult to obtain the result that we want.
Resolving ideas of the present invention is obfuscation, namely machine is not allowed to go to go into seriously each pixel, but be divided into by image many length of sides to be the square tiles of a, machine only needs to judge that fritter is tissue or background image according to all grey scale pixel values are identical in a fritter, so more meets the criterion of human eye vision.
As shown in Figure 6, by little for ex_img blocking and binaryzation, obtain block figure, be designated as block_img, the threshold value of binaryzation is the average of figure ex_img, and experimental result repeatedly shows to adopt average to be feasible.
Step 3, block_img is basic by tissue identification out but to be had much impurity and is also identified into, as shown in Figure 7, carries out Morphological scale-space here to block_img, first corrodes and expand afterwards, independently will filter by impurity fritter, and obtain net result figure.As can be seen from final result figure, organize substantially identified, although there is still some impurity not to be removed, this is because impurity is larger, considers the possibility that fritter tissue exists, does not remove here to larger impurity.
Embodiments of the invention solve the problem of the light stained poor effect of common algorithm identification, pass through obscure idea, well tissue is separated with background, positive effect is produced to digital slices scanning system, make it have better adaptability to dissimilar section.
Be described in detail specific embodiments of the invention above, but it is just as example, the present invention is not restricted to specific embodiment described above.To those skilled in the art, any equivalent modifications that the present invention is carried out and substituting also all among category of the present invention.Therefore, equalization conversion done without departing from the spirit and scope of the invention and amendment, all should contain within the scope of the invention.
Claims (7)
1., based on a stained recognition methods for obscure idea, be applicable to, by the tissue in section preview graph and background separation, it is characterized in that, comprise the following steps:
Step 1, the preview graph of preparation stained, and transferred to gray-scale map, described gray-scale map comprises tissue and background image;
Step 2, is divided into several fritters by gray-scale map image, and the standard identical according to all grey scale pixel values in a fritter judges that fritter belongs to tissue or background image, by fritter binaryzation, obtains block figure;
Step 3, carries out Morphological scale-space by block figure, first corrodes and expands afterwards, independently will filter by impurity fritter, and obtain net result figure.
2., as claimed in claim 1 based on the stained recognition methods of obscure idea, it is characterized in that, the threshold value of the binaryzation in described step 2 is the average of gray-scale map.
3. as claimed in claim 2 based on the stained recognition methods of obscure idea, it is characterized in that, the square tiles of to be several length of sides the be a of the fritter in described step 2.
4. as claimed in claim 1 based on the stained recognition methods of obscure idea, it is characterized in that, the stained in described step 1 comprises common dyeing section or light stained.
5., as claimed in claim 4 based on the stained recognition methods of obscure idea, it is characterized in that, described step 1 comprises:
Step 1.1, prepares the preview graph of light stained and blank section, and subtracts each other after transferring both to gray-scale map, obtain the gray-scale map subtracted each other;
All grey scale pixel values of the gray-scale map subtracted each other are multiplied by coefficient μ, obtain final gray-scale map by step 1.2;
In step 2, final gray-scale map image is divided into several fritters.
6., as claimed in claim 5 based on the stained recognition methods of obscure idea, it is characterized in that, in described step 1.2, coefficient μ determines according to the grey level histogram of the gray-scale map subtracted each other.
7., as claimed in claim 1 based on the stained recognition methods of obscure idea, it is characterized in that, described method is applicable in digital slices scanning system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510381236.9A CN104899584A (en) | 2015-06-29 | 2015-06-29 | Stained section identification method based on fuzzy thought |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510381236.9A CN104899584A (en) | 2015-06-29 | 2015-06-29 | Stained section identification method based on fuzzy thought |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104899584A true CN104899584A (en) | 2015-09-09 |
Family
ID=54032238
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510381236.9A Pending CN104899584A (en) | 2015-06-29 | 2015-06-29 | Stained section identification method based on fuzzy thought |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104899584A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105344620A (en) * | 2015-10-14 | 2016-02-24 | 合肥安晶龙电子股份有限公司 | Color sorting method based on material shapes |
| CN106815849A (en) * | 2017-01-18 | 2017-06-09 | 宁波江丰生物信息技术有限公司 | A kind of method for recognizing biopsy tissues |
| CN111784630A (en) * | 2020-05-18 | 2020-10-16 | 广州信瑞医疗技术有限公司 | Method and device for segmenting components of pathological image |
| CN111784698A (en) * | 2020-07-02 | 2020-10-16 | 广州信瑞医疗技术有限公司 | Image self-adaptive segmentation method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101877074A (en) * | 2009-11-23 | 2010-11-03 | 常州达奇信息科技有限公司 | Tubercle bacillus target recognizing and counting algorithm based on diverse characteristics |
| US20110286654A1 (en) * | 2010-05-21 | 2011-11-24 | Siemens Medical Solutions Usa, Inc. | Segmentation of Biological Image Data |
| CN104361340A (en) * | 2014-11-04 | 2015-02-18 | 西安电子科技大学 | SAR image target fast detecting method based on significance detecting and clustering |
-
2015
- 2015-06-29 CN CN201510381236.9A patent/CN104899584A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101877074A (en) * | 2009-11-23 | 2010-11-03 | 常州达奇信息科技有限公司 | Tubercle bacillus target recognizing and counting algorithm based on diverse characteristics |
| US20110286654A1 (en) * | 2010-05-21 | 2011-11-24 | Siemens Medical Solutions Usa, Inc. | Segmentation of Biological Image Data |
| CN104361340A (en) * | 2014-11-04 | 2015-02-18 | 西安电子科技大学 | SAR image target fast detecting method based on significance detecting and clustering |
Non-Patent Citations (2)
| Title |
|---|
| 张立兰: "基于数学形态学的组织切片细胞分割算法的研究", 《中国优秀硕士论文全文数据库信息科技辑》 * |
| 张雅兰: "图像的二值化处理", 《广西工学院学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105344620A (en) * | 2015-10-14 | 2016-02-24 | 合肥安晶龙电子股份有限公司 | Color sorting method based on material shapes |
| CN106815849A (en) * | 2017-01-18 | 2017-06-09 | 宁波江丰生物信息技术有限公司 | A kind of method for recognizing biopsy tissues |
| CN111784630A (en) * | 2020-05-18 | 2020-10-16 | 广州信瑞医疗技术有限公司 | Method and device for segmenting components of pathological image |
| CN111784698A (en) * | 2020-07-02 | 2020-10-16 | 广州信瑞医疗技术有限公司 | Image self-adaptive segmentation method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104463161B (en) | The color document images segmentation repaired using automated graphics and binaryzation | |
| CN104899584A (en) | Stained section identification method based on fuzzy thought | |
| EP1587295B1 (en) | Boundary extracting method, program and device using the same | |
| EP1526481A3 (en) | Object extraction based on color and visual texture | |
| JP6318755B2 (en) | Filamentous bacteria detection apparatus and filamentous bacteria detection method | |
| CN104156951B (en) | A kind of white blood cell detection method for BAL fluid smear | |
| CN107122597B (en) | Intelligent diagnosis system for corneal damage | |
| CN105447489B (en) | A kind of character of picture OCR identifying system and background adhesion noise cancellation method | |
| CN106485696B (en) | A kind of detection method of the explosive dangerous material stitch defect based on machine vision | |
| CN109544583B (en) | Method, device and equipment for extracting interested area of leather image | |
| CN108182671B (en) | Single image defogging method based on sky area identification | |
| KR101549495B1 (en) | An apparatus for extracting characters and the method thereof | |
| CN109785321A (en) | Meibomian gland method for extracting region based on deep learning and Gabor filter | |
| CN109325421B (en) | Eyelash removing method and system based on edge detection | |
| EP0828375A3 (en) | Image-region discriminating method and image-processing apparatus | |
| KR101875891B1 (en) | apparatus and method for face detection using multi detection | |
| CN107174232A (en) | A kind of electrocardiographic wave extracting method | |
| CN106886779A (en) | A kind of adaptive threshold method of fluorescence microscope images binaryzation | |
| WO2020130799A1 (en) | A system and method for licence plate detection | |
| CN107886493B (en) | Method for detecting conductor stranding defects of power transmission line | |
| CN107491714B (en) | Intelligent robot and target object identification method and device thereof | |
| CN111080723A (en) | Image element segmentation method based on Unet network | |
| CN102968616A (en) | Light yellow tobacco shred identification and deepening method in cigarette loose-end detection process | |
| CN115994921A (en) | Mature cherry fruit image segmentation method integrating HSV model and improving Otsu algorithm | |
| CN109146871B (en) | Crack identification method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150909 |