Abstract In the field of computer science, in order to realize the classification of images, a naive Bayes method based on multiple features has been proposed. This method is widely used and rich. It uses four features, such as extracting gray histogram features, SIFT features, SURF features, and reducing the dimensions of the dataset from data image sets. Naive Bayesian methods are used to obtain The accuracy, recall, and F1 value of the image for each feature. This paper analyzes the application and feature comparison of Naive Bayes method in images, and shows that image representation using SURF feature description can achieve better classification results.

Abstract With the development of technology, IC has entered the era of VLSI. The internal circuit network structure is gradually complex, and the external threats are more and more.We can abstract the circuit in FPGA into complex network .Then characteristic is studied and the invulnerability is measured to optimize the circuit.