PE malicious software detection method and system based on comparative learning

• Main Authors: JIA YETAO, SU QIANYE, GU JIAXIANG, ZHUANG HONGLIN, MENG YANGYANG
• Format: Patent
• Language: Chinese; English
• Published: 15.03.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention provides a PE malicious software detection method and system based on comparative learning, and belongs to the technical field of artificial intelligence, and the method comprises the steps: obtaining a byte channel image, an operation code channel image and an API function channel image of each piece of malicious software of a known type; obtaining a color image set of each piece of malicious software of the known type; obtaining a first color image with a label and a second color image without a label; respectively inputting the first color image and the second color image into a comparative learning network for feature extraction training so as to obtain a feature coding module; constructing a deep neural classification network; inputting the first color image into a deep neural classification network, and performing classification fine tuning training on the deep neural classification network; and obtaining a color image corresponding to the unknown type of malicious software, and inputting