基于多语义情景协同感知的煤矿执法系统研究

Research on coal mine law enforcement system based on multi-semantic scene collaborative perception

研发了基于多语义情景协同感知的煤矿执法系统,系统包括矿山安全生产执法平台、移动执法设备和煤矿安全管控云平台;通过面向多语义情景协同感知的多模态融合技术,融合文本、图像、语音、视频、空间位置等多元信息,构建多模态多情景决策模型;通过信息转换和融合,实现移动执法的情景识别、业务自动导航;构建了闭环执法链路,通过多模态数据融合方法自动识别执法场景,并触发执法任务、执法流程、执法方法,实现井下精准执法,执法数据可传输到煤矿安全管控云平台中,极大提高了监察执法的高效性和客观性。

A coal mine law enforcement system based on multi-semantic scenario collaborative perception is developed, which includes mine safety production law enforcement platform, mobile law enforcement equipment and mine safety control cloud platform. Through the multi-modal fusion technology for multi-semantic scene collaborative perception, the multi-modal and multi situation decision-making model is constructed by fusing multiple information such as text, image, voice, video, spatial location and so on. Through information transformation and fusion, mobile law enforcement scene recognition and business automatic navigation can be realized. A closed-loop law enforcement link is constructed to automatically identify law enforcement scenarios through multimodal data fusion methods, and trigger law enforcement tasks, law enforcement processes, and law enforcement methods, so as to achieve precise law enforcement underground. Law enforcement data can be transmitted to the coal mine safety management and control cloud platform, which greatly improves the efficiency, objectivity of supervision and law enforcement.