Research on automatic evaluation method of millimeter wave human image effectiveness

Aiming at the problem of unstable quality and easy interference of millimeter wave human image, this paper proposes a practical method to classify millimeter wave human body images based on imaging characteristics and image features, and then evaluate the effectiveness of the images. By analyzing the causes of abnormal images, the optimized depth residual network model is used to classify the images, and the image effectiveness evaluation problem is transformed into image classification problem. This method fundamentally simplifies the problem of effectiveness evaluation, so that the indexes of effectiveness evaluation can be quantified, so as to realize the dynamic monitoring of human image quality and scanner status in the process of security check. The automatic evaluation method eliminates the hidden security hazards in the automatic mode of the millimeter wave body scanner, and can effectively improve the stability and applicability of the scanner.