城市轨道交通安检系统智能判图模式研究与设计

为提升城市轨道交通安检系统判图效率及准确率，设计一种新颖的判图模式，将 AI(人工智能)图像识别技术与人工集中判图技术有效结合。首先针对逢液必检现状引入液体检测算法，避免对安全液体的开包查验，其次按安检品风险等级进行分类处理，最后结合 AI 置信度判断、人工抽检或必检判图对安检品进行判定。本判图模式可根据不同阶段下 AI 图像识别准确度和判图需求灵活调节 AI 介入深度，随着 AI 图像识别准确度的不断提升，逐渐从 AI 为辅、人工为主判图模式向 AI 为主、人工为辅的判图模式平滑渐进迭代，最终实现完全智能判图模式。通过案例分析可知，在不降低城市轨道交通车站安检水平的前提下，本判图模式可以进一步达到前端过检增速、后端降本增效的目的。

Research and Design of an Intelligent Judgment Mode forUrban Rail Transit Security Inspection System

To improve the efficiency and accuracy of urban rail-transit security inspection systems, this paper designs a novelpattern-recognition mode to effectively combine artificial intelligence (AI) image recognition technology with manual centralizedpattern recognition. First, based on the current liquid inspection, a liquid detection algorithm is introduced to avoid the openinspection of safe liquids. Second, security products are classified according to their risk levels. Finally, AI confidence judgment,manual sampling, or necessary inspection charts are combined to determine the pattern recognition mode, which can flexiblyadjust the depth of the AI intervention according to the accuracy of the AI image recognition and the requirements for patternrecognition at different stages. With the continuous improvement in the accuracy of AI image recognition, it gradually changesfrom an AI-assisted manual-based pattern recognition mode to an AI-based manual-assisted pattern recognition mode and finallyachieves a fully intelligent pattern recognition mode. A case analysis reveals that the judgment graph model can further achieverapid security inspection, cost reduction, and efficiency increase without reducing the safety inspection level of urban rail transitstations.