With the deepening of China's urbanization construction, public safety is also facing great challenges. Therefore, China has strengthened security measures for important public places, and established monitoring network and security inspection system. However, the current machine security inspection algorithm is not accurate enough for the identification of sensitive objects such as controlled knives, and it also depends on manual detection and confirmation, which leads to low work efficiency and often leads to low efficiency It is also not conducive to the prevention and control of the epidemic. This paper proposes an image segmentation method based on Improved level set to enhance fuzzy clustering. In this paper, the fuzzy clustering method and the improvement of the level set method, first of all, using morphological method image preprocessing algorithms, so as to realize the foreground and background of the area range, is advantageous for the next level set segmentation of the target outline, so as to determine the fuzzy clustering of heart, makes strengthening fuzzy clustering on its own the advantage of rapid more accurate. Through experiments, the effectiveness of the algorithm is verified, and the segmentation accuracy is higher than the traditional segmentation algorithm.