Heterogeneous Gray-Temperature Fusion-Based Deep Learning Architecture for Far Infrared Small Target Detection

Figure 9

Comparison of the results of proposed deep learning based detector, conventional CFAR detector, and HB-based detector. The proposed detector is based on the fusion method using two sets of radiometric temperature data that showed the best performance and normalization method with a value between -1 and 1. In case of HB, the threshold parameter for detection at 208m was applied to 321m as it is.

(a)

(b)

(c)