Image generation-based confrontation contrast learning palm vein recognition method

• Main Authors: YANG KAIYUAN, WANG SHANSHAN, WU YIQUAN, QIN HUAFENG, RUAN DONGMEI, XIONG ZHIPENG, CHEN JING
• Format: Patent
• Language: Chinese; English
• Published: 14.05.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention provides an image generation-based confrontation contrast learning palm vein recognition method, which comprises the following steps of: acquiring an interested image of a palm vein, and performing data enhancement on the interested image to obtain a data enhanced image; inputting the data enhanced image into a UNet network to obtain a generated image, and inputting the generated image into a contrast learning model for contrast learning to obtain a trained contrast learning model; and obtaining a target vein recognition model by combining the trained contrast learning model and a UNet network in an adversarial mode, learning the target vein recognition model by adopting adversarial contrast learning loss, and recognizing a vein image to be recognized based on the trained target vein recognition model to obtain a recognition result. The method can achieve the expansion of the data volume of the vein data set, and improves the vein recognition precision. 本发明提供一种基于图像生成的对抗对比学习手掌静脉识别方法，包括：获取手掌静脉的感兴趣