Picture anomaly detection method based on deep learning

Abstract

The invention discloses a picture data set-oriented picture anomaly detection method based on deep learning. According to the method, normal picture category data is used as input data; a picture anomaly detection model based on deep learning is constructed and includes two submodules, one representation module is used for learning features of a picture data set of a normal category; one detectionmodule is used for predicting the probability that the input picture belongs to an abnormal picture category; in addition, confidence estimation is used for improving prediction accuracy, the two modules adopt an adversarial training method, the representation module can better learn features of a normal category picture data set, and the detection module can give a prediction result with higherconfidence and higher accuracy. For four common data sets in the field of anomaly detection, the method overcomes the problem that abnormal types of pictures are various in types and difficult to collect, only normal types of pictures are needed as training data, and meanwhile, the effect is remarkably superior to that of other existing anomaly detection methods for picture data sets.

Description

technical field [0001] The invention belongs to the field of artificial intelligence, and specifically relates to a picture data set data, a method for analyzing and detecting pictures of abnormal categories after defining normal picture categories. Background technique [0002] Anomaly detection aims to identify rare and unusual instances (anomaly classes) that differ from the majority (normal class) patterns in a dataset. In recent years, many researchers in the direction of multimedia and computer vision have carried out research in the direction of detection and classification. They have conducted in-depth research on the detection of multimedia such as images, videos, and audios. For example, they study audio detection in acoustic scenes and events, object detection in vehicular and pedestrian videos, and action recognition detection in videos. Anomaly detection, which has just emerged recently, plays an important role in the fields of multimedia and computer vision, ...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to solve the problem of lack of end-to-end detection model and low confidence estimation in the existing picture anomaly detection method, and propose a picture anomaly detection method based on deep learning. The present invention utilizes the method based on deep learning to innovate A classification of image anomaly detection method is proposed, which improves the effect of image anomaly detection

Images