A Non-contact Method for Measuring the Size Distribution of Rock Blocks Based on Image Processing Technology

Due to the unique volcanic geological conditions in the Changbai Mountain area, debris flow disasters occur frequently and show an increasing trend, greatly threatening the safety of nearby people’s lives and property. The size distribution of larger rock blocks carried by debris flows directly determines the destruction scale of debris flow. In this paper, a non-contact method based on image processing technology for measuring the size distribution of rock blocks is proposed. It employs the K-way Normalized Cuts algorithm in combination with top-hat and bottom-hat transform to segment the adjacent rock blocks in the image. Based on segmentation results, the sizes of rock blocks are automatically calculated by the conversion of the pixel area of each partitioned region through the image scale, then generating the cumulative size distribution curves. The median of the cumulative size distribution curve is chosen as a single statistical parameter to characterize the rock blocks group. By comparing the median from the non-contact method with the manual field measurement method, the results are basically compatible and the standard deviation is less than 3%, indicating the effectiveness and reliability of the non-contact method. This automated non-contact method provides a new approach for rapidly analyzing the size distribution of rock blocks in the debris flow deposits, which can not only significantly reduce physical labor but also guide the works of debris flow prevention in a timely manner based on the real-time data.