基于人工靶向免疫疗法的红外手部痕迹目标提取

红外手部热痕迹图像是特殊的模糊图像，该文提出一种人工靶向免疫疗法对其进行手部目标提取。首先依据序列图像中像素灰度的变化趋势设计了先天性免疫识别进行初分割；然后借鉴免疫的提呈机制，根据热扩散特性定义同心圆模板提取特征；基于模板特征对模糊像素集适应性免疫识别；最后，指尖指谷病变检测分析，实施靶向治疗，保证了手的形态特征。与分水岭、SOM网络以及近几年研究成果进行了比较，表明提出的算法在目标提取率、绝对误差率均优于现有算法，提取结果更符合手的形态，同时扩展了应用热痕迹信息的时间跨度。

Target Extraction of Hand Infrared Trace Images Based on Artificial Targeting Immunotherapy

Hand infrared trace images can not clearly reflect the original contact contour of hand, which belong to a special kind of infrared blurred images. Inspired by biological immune, an artificial targeting immunotherapy is proposed to extract the hand target. Firstly, according to the feature of temporal correlation, the innate immune recognition is designed to preliminary segmentation. Secondly, according to the immune presentation, a concentric circles template based on thermal diffusion is defined to extract features. Then adaptive immune recognition is applied to the fuzzy pixels set based on the obtained template features. Finally, for the detected finger valley and fingertips lesions, targeted therapy is implemented to keep the shape of the hand. The proposed algorithm is compared with watershed method, SOM network and recent research achievements. Experimental results show that the proposed algorithm exhibits better extraction performance, meanwhile the application time of thermal trace images is extended.