Improvement method for terahertz imaging quality

Terahertz detection and recognition of suspicious objects hidden in the human body based on DeepLabV3+ deep learning model

“…The automatic detection and recognition method adopts the traditional detection process of image segmentation, positioning, and recognition, which is difficult to meet the requirements of fast security inspection and high-precision resolution. Up to now, deep learning methods have made breakthrough progress in object visual detection [7][8][9][10][11][12]. But, the research on object detection and recognition in terahertz images mainly focuses on traditional methods.…”

R-PCNN Method to Rapidly Detect Objects on THz Images in Human Body Security Checks

“…In [6], to improve the image object location detection, threshold segmentation was applied to segment objects from the background; the LOG operator was used to extract edge features, and gun detection in the terahertz image was achieved. In [7], by analyzing the gray histogram and fitting the curve, the seed point and growth threshold were used to grow the image region, and object segmentation and detection in the human terahertz image were achieved.…”