Method and device for identifying dangerous area in underground coal mine

The embodiment of the invention discloses an underground coal mine dangerous area identification method and device. The method comprises the following steps: acquiring an underground coal mine image acquired by a camera arranged in an underground coal mine; the underground coal mine image is input into a preset dangerous area recognition model, a dangerous area in the underground coal mine image is recognized, and the dangerous area recognition model is obtained by adopting training data to train an image semantic segmentation algorithm model. According to the method, the dangerous area in the underground coal mine can be identified accurately in real time, and the safety of underground production of the coal mine can be improved. 本发明实施例公开了一种煤矿井下危险区域识别方法及装置，该方法包括：获取通过设置在煤矿井下的摄像头采集的煤矿井下图像；将所述煤矿井下图像输入到预设的危险区域识别模型中，识别出所述煤矿井下图像中的危险区域，其中，所述危险区域识别模型为采用训练数据对图像语义分割算法模型进行训练得到的。本发明实现了实时、准确的对煤矿井下危险区域进行识别，有助于提高煤矿井下生产的安全性。