Parasite egg identifying method based on sparse representation

Abstract

The invention belongs to the technical field of image identification, and particularly relates to a parasite egg identifying method based on sparse representation. The method comprises that steps of establishing an initial dictionary; using a K-SVD algorithm to study the dictionary; processing an input image; calculating a reconstructed error matrix; obtaining a candidate image block; identifying the candidate image block. The parasite egg identifying method based on sparse representation introduces a sorting algorithm based on sparse representation, reinforces robustness of the whole parasite egg algorithm to various interfering factors, introduces the Batch-OMP algorithm which is used in the process of large-scale sparse representation, improves identification efficiency, introduces a method that a dictionary is directly established after a sample is subjected to gaussian pyramid dimensionality reduction, avoids the step that egg target characteristics are abstracted, enables the identification process to be simple and convenient, introduces the method for establishing an error matrix and solving the smallest local value of the error matrix, and avoids the situation that different image blocks containing the same target are obtained in the preliminary locating process.

Description

technical field [0001] The invention belongs to the technical field of image recognition, and in particular relates to a method for recognizing parasite eggs based on sparse representation. Background technique [0002] The key to the automatic identification of parasite eggs based on computer image processing and medical microscopy technology is to design a fast and effective image recognition algorithm. In the past, the automatic identification method of parasite eggs based on images mainly relied on the separation of egg targets first. , and then extract various features of the target, and finally combine a classifier to complete the recognition. Give two examples of methods that are more relevant to the present invention: (1) Derya Avci et al. combined 7 invariant moments and support vector machines of Hu in 2009 in the document "An expert diagnosis system for classification of human parasite eggs based on multi-class In "SVM", 16 kinds of human parasite eggs are identi...

AI Technical Summary

Problems solved by technology

Give two examples of methods that are more relevant to the present invention: (1) Derya Avci et al. combined 7 invariant moments and support vector machines of Hu in 2009 in the document "An expert diagnosis system for classification of human parasite eggs based on multi-class In "SVM", 16 kinds of human parasite eggs are identified, although a high recognition rate is obtained, but it can only be achieved under the premise that the image is relatively ideal, and the situation when there are many interference factors is not considered; (2) Chinese patent CN201110022426 .3 A method combined with the edge histogram of parasite eggs is proposed to recognize the shape of human parasite eggs, which can overcome the influence of weak boundaries and improve the reliability of recognition. However, for parasites with similar shapes There are still deficiencies in shape recognition

Judging from the existing methods, there are many types of features. In addition to the features described in the above methods, they also include color, shape, size, texture, etc. The quality of feature selection largely determines the final recognition rate. And the steps of extracting features are also difficult to complete accurately

There are also a variety of classifiers, including Bayesian classifiers, linear discriminant analysis, support vector machines, neural networks, minimum distance, etc., because these classifiers are sensitive to features, so which feature to choose is optimal for the classifier It is often difficult to determine, and these classifiers are less robust to interference factors such as noise, occlusion, and impurities

Images