CN108537124A - A kind of cervical cancer cell recognition methods based on cascade multiple Classifiers Combination - Google Patents
A kind of cervical cancer cell recognition methods based on cascade multiple Classifiers Combination Download PDFInfo
- Publication number
- CN108537124A CN108537124A CN201810203250.3A CN201810203250A CN108537124A CN 108537124 A CN108537124 A CN 108537124A CN 201810203250 A CN201810203250 A CN 201810203250A CN 108537124 A CN108537124 A CN 108537124A
- Authority
- CN
- China
- Prior art keywords
- cell
- cancer cell
- cervical
- cervical cancer
- recognition methods
- 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
- G06V20/695—Preprocessing, e.g. image segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
-
- 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
- G06V20/698—Matching; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of cervical cancer cell recognition methods based on cascade multiple Classifiers Combination, it is characterised in that:Based on the cervical cancer cell recognition methods of cascade multiple Classifiers Combination, the cell characteristic of cervical cell image is extracted first;Then the feature for utilizing feature selecting algorithm optimization extraction, finds out the feature set that can most distinguish normal cell and cancer cell;Finally cell is identified using cascade Combining Multiple Classifiers, improves the discrimination of cervical cancer cell.
Description
Technical field
The present invention relates to image characteristics extraction, the identification of feature selecting and image is specifically a kind of to utilize the more classification of cascade
The method of device fusion identifies cervical cancer cell.
Background technology
As the development of medical technology and computer technology is with maturation, the cell image recognition technology that the two is combined is met the tendency of
And give birth to, and cause extensive concern.Wherein, it be used to detect cervical cell in picture with automanual detection method automatically
Profile, and therefrom select abnormal cell.Cervical cell image recognition technology is that a kind of new cervical cell of rising in recent years is known
Other method.The method overcome traditional artificial interpretation screening mode there are of high cost, heavy workload, reliability and accuracy by
To doctor's professional technique and subjective emotion influence the problems such as.The purpose of cervical cell image recognition technology research is to identify
It whether there is lesion epithelial cell in cervical cell image, reduce the workload of doctor and reduce and is existing when cervical cell identification
False positive and false negative.
Image recognition is briefly sought to a kind of research object, is identified and is classified according to its certain feature.
It is believed that carrying out difference classification to digital picture, it is substantially exactly to carry out pattern-recognition to image.This identification probably already exists
In people’s lives practice.However, with the expansion of practical activity, deep and socialization needs, people not only need to identify
The many things of classification number, and identified contents of object also becomes increasingly complex.Especially because scientific and technological level carries
It is high so that a variety of different research objects " image conversion " or " digitlization " can be used certain technology and the object of investigation are converted into
Picture, oscillogram and several data, these data can represent studied object.But for pattern-recognition,
Either data, signal or flat image or stereo landscape are all to remove their physical content and find out their general character,
The one kind that is classified as with same general character, and it is classified as with another general character person another kind of.The purpose of image steganalysis is exactly
It develops and uses certain instrument or equipment, automatically process certain information, the task of classification and identification is completed instead of people, and can be fast
Speed and accurately carry out figure identification.
Single grader the problems such as there are one-sidedness and poor generalization abilities, cascade Combining Multiple Classifiers can improve figure
As discrimination.
Invention content
The purpose of the present invention is to provide a kind of cervical cancer cell recognition methods based on cascade multiple Classifiers Combination, with solution
Certainly the problems mentioned above in the background art.
To achieve the above object, the present invention provides the following technical solutions:A kind of uterine neck based on cascade multiple Classifiers Combination
Cancer cell identification method, includes the following steps:
(1), cervical cell image characteristics extraction:The color characteristic of cervical cell image is extracted, shape feature and texture are special
Sign;
(2), cervical cell image feature selection:Go out the high feature of Category Relevance with recursion elimination algorithms selection;
(3), cervical cancer cell identifies:First order grader uses KNN, and random forest grader and C4.5 graders are arranged side by side
It merges, second level grader uses LR graders;Recognition methods based on cascade multiple Classifiers Combination can improve uterine neck
The discrimination of cancer cell.
Compared with prior art, the beneficial effects of the invention are as follows:People can be helped to complete cervical cancer cell classification and distinguish
The task of knowledge, and can quickly and accurately carry out the identification of cervical cancer cell has good robustness, accuracy and low
Complexity, and the result identified is fine.
Description of the drawings
Fig. 1 is the flow chart of the present invention;
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
It is as follows based on cascade Combining Multiple Classifiers:
A, the first order uses parallel multi-categorizer:K- Nearest Neighbor Classifiers, random forest grader and C4.5 classification are used first
Device is classified, and then ballot method is used to carry out parallel sorting device fusion;
B, the second level uses LR graders.
K- Nearest Neighbor Classifiers are a kind of efficient and simple sorting techniques;Random forest grader has accuracy rate height, Shandong
The advantages that stick is good, easy to use;The classifying rules that C4.5 graders generate is it can be readily appreciated that accuracy rate is higher;LR graders pair
Big data has the characteristics that model of fit accuracy rate is high, processing speed is fast, is divided into cervical cell normally carefully using LR graders
Two class of born of the same parents and abnormal cell.
As shown in Figure 1, a kind of cervical cancer cell recognition methods based on cascade multiple Classifiers Combination, includes the following steps:
A, cervical cell image characteristics extraction:Extract the color characteristic of cervical cell image, shape feature and textural characteristics;
B, cervical cell image feature selection:Go out the high feature of Category Relevance with recursion elimination algorithms selection;
C, cervical cancer cell identifies:First order grader uses KNN, and random forest grader and C4.5 graders melt side by side
It closes, second level grader uses LR graders;Recognition methods based on cascade multiple Classifiers Combination can improve cervical carcinoma
The discrimination of cell.
In conclusion cervical cancer cell recognition methods accuracy rate provided by the invention is high, image classification can be effectively improved
Accuracy and efficiency;People can be helped to complete the task of cervical cancer cell classification and identification, and can be quickly and accurately
The identification of cervical cancer cell is carried out, there is good robustness, accuracy and low complex degree.
It although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, can be with
Understanding without departing from the principles and spirit of the present invention can carry out these embodiments a variety of variations, modification, replace
And modification, the scope of the present invention is defined by the appended.
Claims (2)
1. a kind of cervical cancer cell recognition methods based on cascade multiple Classifiers Combination, it is characterised in that:Include the following steps:
A, cervical cell image characteristics extraction;
B, cervical cell image feature selection;
C, cervical cancer cell is identified in cascade classifier.
2. a kind of cervical cancer cell recognition methods based on cascade multiple Classifiers Combination according to claim 1, feature
It is:The step C cascades multi-categorizer cancer cell identification method includes the following steps:
A, the first order use parallel multi-categorizer, first use K- Nearest Neighbor Classifiers, random forest grader and C4.5 graders into
Row classification, then uses ballot method to carry out parallel sorting device fusion;
B, the second level uses LR graders, is a kind of multivariate statistical method being widely used, and has model of fit to big data
The features such as accuracy rate is high, processing speed is fast, is divided into two class of normal cell and abnormal cell using LR graders by cervical cell.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810203250.3A CN108537124A (en) | 2018-03-13 | 2018-03-13 | A kind of cervical cancer cell recognition methods based on cascade multiple Classifiers Combination |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810203250.3A CN108537124A (en) | 2018-03-13 | 2018-03-13 | A kind of cervical cancer cell recognition methods based on cascade multiple Classifiers Combination |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108537124A true CN108537124A (en) | 2018-09-14 |
Family
ID=63484441
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810203250.3A Pending CN108537124A (en) | 2018-03-13 | 2018-03-13 | A kind of cervical cancer cell recognition methods based on cascade multiple Classifiers Combination |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108537124A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110647945A (en) * | 2019-09-27 | 2020-01-03 | 杭州智团信息技术有限公司 | Liquid-based cervical cell smear classification method, system and implementation device |
CN113255718A (en) * | 2021-04-01 | 2021-08-13 | 透彻影像科技(南京)有限公司 | Cervical cell auxiliary diagnosis method based on deep learning cascade network method |
-
2018
- 2018-03-13 CN CN201810203250.3A patent/CN108537124A/en active Pending
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110647945A (en) * | 2019-09-27 | 2020-01-03 | 杭州智团信息技术有限公司 | Liquid-based cervical cell smear classification method, system and implementation device |
CN110647945B (en) * | 2019-09-27 | 2022-11-22 | 杭州智团信息技术有限公司 | Liquid-based cervical cell smear classification method, system and implementation device |
CN113255718A (en) * | 2021-04-01 | 2021-08-13 | 透彻影像科技(南京)有限公司 | Cervical cell auxiliary diagnosis method based on deep learning cascade network method |
CN113255718B (en) * | 2021-04-01 | 2022-07-01 | 透彻影像科技(南京)有限公司 | Cervical cell auxiliary diagnosis method based on deep learning cascade network method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106803247B (en) | Microangioma image identification method based on multistage screening convolutional neural network | |
TWI767506B (en) | Image recognition method, training method and equipment of recognition model | |
CN106248559B (en) | A kind of five sorting technique of leucocyte based on deep learning | |
Putzu et al. | Leucocyte classification for leukaemia detection using image processing techniques | |
Carmichael et al. | Shape-based recognition of wiry objects | |
WO2019114036A1 (en) | Face detection method and device, computer device, and computer readable storage medium | |
CN104778470B (en) | Text detection based on component tree and Hough forest and recognition methods | |
CN104881671B (en) | A kind of high score remote sensing image Local Feature Extraction based on 2D Gabor | |
CN103810503A (en) | Depth study based method for detecting salient regions in natural image | |
CN104408449A (en) | Intelligent mobile terminal scene character processing method | |
CN105760858A (en) | Pedestrian detection method and apparatus based on Haar-like intermediate layer filtering features | |
CN109635669A (en) | Image classification method, the training method of device and disaggregated model, device | |
WO2019119515A1 (en) | Face analysis and filtering method, device, embedded apparatus, dielectric and integrated circuit | |
CN112464797A (en) | Smoking behavior detection method and device, storage medium and electronic equipment | |
Zhu et al. | Detecting natural scenes text via auto image partition, two-stage grouping and two-layer classification | |
Wu et al. | Natural scene text detection by multi-scale adaptive color clustering and non-text filtering | |
CN110084232B (en) | Method and device for recognizing Chinese characters in license plate and terminal equipment | |
Dong et al. | A self-adaptive approach for white blood cell classification towards point-of-care testing | |
CN108537124A (en) | A kind of cervical cancer cell recognition methods based on cascade multiple Classifiers Combination | |
CN115294377A (en) | System and method for identifying road cracks | |
CN108073940A (en) | A kind of method of 3D object instance object detections in unstructured moving grids | |
CN107818341A (en) | A kind of color extraction method based on improvement K means algorithms | |
CN108846364B (en) | FPGA-based video feature detection method and system | |
Vajravelu et al. | Machine learning techniques to detect bleeding frame and area in wireless capsule endoscopy video | |
CN110866547A (en) | Automatic classification system and method for traditional Chinese medicine decoction pieces based on multiple features and random forest |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180914 |
|
WD01 | Invention patent application deemed withdrawn after publication |