Research On Image Processing And Recognition Of Blasting Rock Fragmentation Based On Machine Vision

The blasting fragmentaion is an important index to measure the blasting effect.A fast and efficient blasting fragmentaion detection technology and identification method can be established to quickly quantify the feedback blasting effect,can reduce the blasting fragmentaion detection intensity of the detection technicians and influence on the loading work.To this end,the machine vision technology is introduced into the detection of the blasting fragmentaion information.And the large-scale blasting rock fragmentaions exists these characteristics,which are adhered to each other with the irregulared shape,the large fragmentaion differences and the adopted noisy image.The system development of the blasting fragmentaion image processing and engineering application is carried out for research projects.The specific research contents and main results are as follows.(1)Development of image processing and recognition system for blasting rock fragmentaions.According to the characteristics of the image,the rock scene image denoising method and analysis method are proposed,and BFAS of the image analysis software based on ImagePy is developed.Firstly,the bilateral filtering plug-in is re-development,and the selection of bilateral filtering parameters is discussed through experimental analysis.Then,the adaptive thresholding algorithm and the mathematical morphology method are applied to the secondary processing of the rock binary image.Finally,the evaluation indicators of rock mass image fragmentaion are used,which are the fragmentaion distribution histogram,the fragmentaion cumulative distribution curve,the large fragmentaion rate to quantitatively evaluate the blasting pile.(2)Collection design of large-scale soil-rock blasting images.According to the large-scale blasting pile shape,the loading process and the scene environment characteristics,the basic principles of blasting pile image acquisition are proposed.The blasting surface image acquisition scheme and the blasting space layered image acquisition scheme are designed respectively,which provides a basis for spatial distribution data collection.(3)Engineering applications.On-site image acquisition,image processing and quantitative evaluation were carried out on 5 blastings of a project,and the evaluation indexes such as large fragmentaion rate,fragmentaion distribution histogram and fragmentaion cumulative distribution curve were obtained.And the comprehensive blasting effect was analyzed and evaluated.BFAS of the image analysis software based on ImagePy development can solve the problem of fragmentaion efficient identification and analysis of the blasting pile rock fragmentaions.And it is of great significance to promote the localization of rock image fragmentaion analysis software and the intellectualization of rock fragmentaion detection technology.