Deep learning algorithm for segmenting abnormal region in image

Abstract

The invention relates to the technical field of unsupervised image abnormal segmentation, in particular to a deep learning algorithm for segmenting an abnormal region in an image, which comprises the following steps of: S1, firstly, performing positive sample image feature extraction, and achieving high algorithm accuracy by designing the deep learning algorithm for segmenting the abnormal region in the image; reasoning time and memory occupancy are constant orders of magnitude and are irrelevant to the number of positive sample reference pictures. The algorithm process is simple, only the optimal single-level feature representation needs to be selected, the effect of multi-level feature fusion is not needed, meanwhile, the segmentation threshold can be selected in different modes according to the actual situation, the algorithm is good in generalization ability and easy to reuse, the problems that an existing unsupervised abnormal segmentation algorithm is not high in accuracy, long in reasoning time, large in memory occupation and the like are solved. Therefore, the abnormal region can be segmented accurately and efficiently, and the current optimal effect can be achieved without multi-level feature fusion.

Description

technical field [0001] The invention relates to the technical field of unsupervised image abnormal segmentation, in particular to a deep learning algorithm for segmenting abnormal regions in images. Background technique [0002] There are two main existing unsupervised anomaly segmentation algorithms: 1. Based on the reconstruction method, the positive sample image is reconstructed through the autoencoder (AutoEncoder). Due to the large difference between the abnormal sample before and after reconstruction, the abnormal region is segmented. This method model The amount of parameters is large, the training time is long, the reconstruction effect of complex patterns is not good, and the segmentation accuracy is not high; 2. Based on the pre-trained network to extract feature representation, and use knn and other methods for image retrieval, the image to be tested is the closest to the feature space The positive sample reference picture is compared to segment the abnormal area....

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a deep learning algorithm for segmenting abnormal regions in images. In order to overcome the problems of the existing unsupervised abnormal segmentation algorithms such as low accuracy, long reasoning time, and large memory usage, the present invention provides a new The solution can accurately and efficiently segment abnormal areas, and can achieve the current optimal effect without multi-level feature fusion

Images