Image-based Identification Technique With Timeliness For Particle Sizes Of Coarse-grained Soils

The traditional method for analyzing the particle size of coarse-grained soil samples and obtaining the particle gradation curve is the sieving method.Although the particle size analysis using the sieving method is simple and intuitive,the operation is cumbersome and the particle gradation information of the outdoor undisturbed soil cannot be obtained immediately.In order to solve this problem,a research on coarse-grained soil gradation information recognition technology based on digital image processing was carried out.The main research contents and results are as follows:(1)Perform research and comparison of commonly used image analysis methods,use the local adaptive threshold method to binarize the image,and use the watershed algorithm to segment the bonded part of the image,while retaining more details,Better segmentation of the bonded part of the particles,and analysis of different types of images with fewer and more particles,respectively,to verify the reliability of the results;(2)Technology Digital image processing technology has developed a strong and time-sensitive coarse particle size analysis system,which can obtain image photos through mobile phone cameras,and then obtain the particle gradation curve of soil particles through analysis and identification of the image,and according to the particle gradation The curve can get other information of soil particles;(3)Apply the developed system to the slag monitoring of the TBM conveyor belt,which can realize s real-time analysis and processing of the slag image,and send out when the information about the poor surrounding rock grade in front and high knife consumption is obtained according to the analysis results Warning message.