Hand target extraction from infrared images based on the danger theory

The gray distribution of infrared thermal image is uneven, which brings difficulties to image segmentation. In this paper, based on the “the danger theory” of immunology, a method for hand segmentation is proposed. By using the dual-threshold of immune genetic algorithm, the image is segmented into a target pixel set, a background pixel set and a blurred pixel set. According to the timing characteristics of trace images, the 3D template is designed to extract features. According to the template characteristics, the danger signal is defined and the danger theory is used to identify the blurred pixel set. The proposed algorithm is compared with watershed method, active contour model and recent research achievements. Experimental results show that the proposed algorithm has a better performance of target extraction rate, coincidence with reference standard, and less absolute error rate.