Research on Image Context Anomaly Detection Based on Optimized CNN_SVM

Context image anomaly detection, this paper puts forward a kind of based on convolution neural network(CNN) and support vector machine (SVM) anomaly detection methods. Firstly, the image object recognition technology is used to detect the target image and extract the context difference features. Secondly, CNN and SVM are used to classify context anomalies and detect targets. At the same time, the differential evolution algorithm (DE) is used to optimize the parameters of the cnn method. The analysis results show that the improved CNN_SVM model based on DE has high performance and accuracy of context anomaly feature extraction.