Image Anomaly Detection Method Based on Variational Autoencoder

Abstract

The invention discloses an image abnormality detection method based on variational self-encoder, in particular an abnormality detection method described by fusion of variational self-encoder and support vector data, which solves the two stages of abnormality detection and feature extraction in the prior art separation, limited anomaly detection performance, and the inability of traditional anomaly detection to cope with high-dimensional, large-scale anomaly detection tasks. The implementation steps include: image data acquisition; data set division and data preprocessing; constructing an anomaly detection model based on a variational autoencoder; training an anomaly detection model; calculating a threshold for distinguishing normal and abnormal image data according to the trained model; Use the trained model to determine whether the image to be tested is an abnormal image. The invention adopts the support vector data description to constrain the distance of the features extracted by the variational autoencoder, the extracted features are more suitable for abnormal detection, the memory complexity is low, and it can be applied to high-dimensional and large-scale abnormal detection tasks.

Description

technical field [0001] The invention belongs to the technical field of deep learning, and mainly relates to image anomaly detection, in particular to an image anomaly detection method based on a variational autoencoder, which is used for image anomaly detection combined with support data vector description. Background technique [0002] The task of anomaly detection is to identify data that is inconsistent with expectations, which is usually defined as an anomaly. Anomaly detection has an important responsibility in many fields. Among the anomaly detection technologies that have been researched and developed, applications such as credit card fraud detection, network intrusion detection, and medical diagnosis are involved. The present invention addresses the problem of anomaly detection with respect to images. [0003] The methods applied to image anomaly detection are divided into traditional anomaly detection techniques and deep anomaly detection techniques. Traditional ...

AI Technical Summary

Problems solved by technology

proposed a support vector machine method (OC-SVM) in the paper "Estimating the Support of a High Dimensional Distribution", which can construct a hyperplane model of positive data and divide the data on the other side of the hyperplane into Anomaly classes are used to obtain ideal anomaly detection results. The disadvantage of this method is that the result depends largely on the selection of regularization parameters and kernel functions. When the amount of data is huge, the memory complexity is very high, which limits this method. Applications in Large-Scale Image Anomaly Detection Tasks

The disadvantage of this type of method is that because the general loss function is used to replace the customized anomaly detection target, the target of anomaly detection cannot affect the hidden layer features extracted by the network at all, making this method often suboptimal

[0005] In summary, the traditional anomaly detection method applied in the image field will not have a good anomaly detection effect due to the large amount of data in the image dataset. The deep anomaly detection method represented by the encoder and the variational autoencoder is used in the anomaly detection task. When , due to the lack of customized anomaly detection targets or the lack of tolerance for the diversity of normal samples, the performance of the deep model is often not fully utilized

Images