Weapon Detector System by Using X-ray Image Processing Technique

摘要

The images in the research were obtained from the internet because there is no x-ray camera. After preparing images, the next step is to test that the evaluation used to analyze the suspicious images. The images were detected by Canny Edge Detector algorithm that consists of four steps that started by smoothing the images with a Gaussian filter to eliminate interfering signals. Then, it calculated magnitude and orientation of gradient. Non-maxima Suppression and Gradient Magnitude were used to create thinner edges. The last step is double threshold algorithm to identify edge pixels and connect the edges by optimizing image recognition technique. When detecting a suspicious object, the system would send the notification. The result of this project is that the template matching algorithm was 80% accurate.