二维最大熵图像分割下的煤矿充水采空区AR检测技术

AR detection technology of coal mine water-filled goaf under two-dimensional maximum entropy image segmentation

传统煤矿采空区检测技术受到二维图像特征与井下环境因素的影响,在图像采集与分析精度方面存在较大偏差,无法精准还原矿井下真实环境,严重影响矿区井下环境勘测进度。为了解决这一问题,引入AR检测技术,对传统检测方法做如下优化:首先,对煤矿充水采空区的二维最大熵进行确定;然后,对图像信号进行AR初始检测模型构建及图像信号小波处理;最后,完成二维最大熵图像分割下的煤矿充水采空区识别模型建立;通过实验对比发现,采用提出检测方法所获得的检测结果与实际结果之间的偏差最小,且检测效果稳定性与可靠性好,应用要求低,实现技术门槛低,适合大面积推广应用。

The detection technology of traditional coal mines is affected by the two-dimensional image characteristics and underground environmental factors,and there is a significant deviation in the accuracy of image acquisition and analysis,which cannot accurately restore the natural environment under the mine and seriously affects the progress of underground environment survey in the mining area.In order to solve this problem,AR detection technology is introduced to optimize the traditional detection method as follows:firstly,the 2 D maximum entropy of the coal mine is determined;then,the initial image signal construction and the image signal wavelet processing;finally,through the experimental comparison,the deviation between the detection results and the actual results is minimum,and the detection effect has good stability and reliability,the application requirement is low,realizing low technical threshold and suitable for extensive area application.