Research On Three Dimensional Measurement Of Coal Quantity Of Main Conveyor Belt Based On Binocular Vision

In stereo matching,image edge information is one of the most widely used features,but there is a low precision of edge matching in stereo matching,which has always been the bottleneck of stereo matching.It is the key to find the edge detection algorithm which can locate the edge effectively.In addition,binocular vision has some research on the measurement of irregular object volume.In the complex coal mine environment,how to obtain accurate coal production data is still an important problem.In order to solve these problems,we introduce binocular vision technology to the volume measurement of irregular objects such as coal.In this paper,the problem of low matching precision and the measurement error of coal quantity of main transportation surface of belt conveyor in coal mine is very low.A stereo matching method combining edge features and a dynamic irregular object weight measurement method based on parallax map are proposed.This paper mainly studies the three-dimensional matching algorithm and the coal quantity measurement research method,as shown below:(1)Aiming at the problem of low edge matching accuracy in stereo matching,a RBF Harris sift image stereo matching algorithm is proposed based on Harris sift image stereo matching algorithm.Based on Harris sift,radial basis function is introduced to analyze and extract the edge features of image.First of all,the algorithm uses RBF algorithm to extract the image edge contour information.Because of the small amount of calculation and fast speed of radial basis function,it can quickly realize the feature extraction of the image edge.Then,Harris algorithm is used to detect the feature points of the image edge obtained by RBF algorithm,and the main direction of those feature points is calculated,and the SIFT feature descriptor is generated Finally,the left and right edge detection is completed Corners in the view match.Through the comparative analysis of experiments,it is concluded that the shortcomings of edge extraction method in stereo matching are solved,and the accuracy of stereo matching is improved.(2)Aiming at the problem of coal measurement weight measurement in coal mines,the calculation method of the average height of the coal block makes the calculation error of the coal amount larger.This paper restores the depth information of each point in the three-dimensional space corresponding to the two-dimensional image,and calculates the three-dimensional information of the corresponding point;Based on the three-dimensional information,a dynamic irregular volume measurement method based on pixels is used to calculate the volume of raw coal on the surface of the belt conveyor.When calculating the volume,the height of the pixel is used to indicate the actual height of the coal.The method takes the disparity map as input,first uses the watershed segmentation algorithm to achieve the segmentation of the disparity map,then uses the stereo imaging method to achieve three-dimensional reconstruction of the target(raw coal)area,and then uses the improved pixel-based volume calculation method to achieve the volume of the raw coal Calculate,and finally realize coal quantity measurement in combination with coal density.It can be seen from the result the difference between the actual weight and the measured weight is less than 10%,which proposes an effective scheme for the measurement of underground coal quantity in coal mines,which can meet the estimated demand of actual coal underground coal production.