Research On Detection Technology Of Passable Area For Auxiliary Transportation Vehicle In Underground Coal Mine

Auxiliary transportation vehicles play an important role in the transportation of personnel,materials,gangue and equipment in underground coal mine.Due to the narrow space and complex environment of underground tunnel in coal mine,together with the change of light intensity,it is easy to cause driver fatigue,unclear observation,manual operation errors and other driving safety accidents.With the continuous development of China’s smart mine construction,it is of great practical significance to improve the safety and intelligence of auxiliary transportation vehicles to realize the minimization and unmanned coal mine.This thesis is based on the theory of machine vision to study the detection of passable areas for auxiliary transportation vehicle in underground coal mine.The main work of this thesis is as follows:(1)Research on rail detection method based on switchable models.Enhance and denoise the underground images of coal mines,and use edge detection,morphological top hat transformation,and eight connected area filtering to obtain binary images of railway tracks;Divide the parts below the vanishing point of the image into close and far range regions,and fit the straight rail model and curved rail model respectively.The straight rail model uses asymptotic probability Hough transform to fit the straight rail,and the curved rail model uses Catmull Rom curve with improved feature point selection to fit the curved rail.Verify the advantages of this method in terms of detection accuracy and robustness through rail detection experiments.(2)Research on detection method of driving area based on fusion of pipeline and rail.Based on the common color attributes of underground tunnel wall pipelines in coal mines,a pipeline feature detection strategy is constructed through HSV color space transformation,edge detection,and least squares fitting;Using camera calibration to obtain camera parameters,convert rail and pipeline feature points to the world coordinate system;Research on an auxiliary method for extracting the centerline trajectory of transportation vehicles based on weighted fusion of pipeline and rail feature points;Considering both the presence and absence of railway tracks in the tunnel,the rules for generating driving area for underground auxiliary transportation vehicles in coal mines are studied;Build a simulated tunnel and conduct experimental research,and the experimental results show that the proposed method can achieve the detection of driving areas.(3)Research on detection method of passable area based on euclidean distance.In order to solve the problem of over segmentation in watershed algorithm,the minimum labeling method is used to segment obstacles and backgrounds,and the Graph Cut algorithm is used to detect obstacles;Extract pedestrian HOG features and use PCA dimensionality reduction to serially fuse the reduced HOG features with LBP features,and use SVM classifier to train the fused features;By calculating the Euclidean distance between the driving area and the obstacle,the positional relationship between the obstacle and the driving area is judged,and the passable area is obtained.Experiments show that the algorithm in this thesis can achieve obstacle detection and passable area detection.(4)Interface design for detecting the passable area of auxiliary transportation vehicles.Design an auxiliary transportation vehicle passable area detection software based on the GUI platform of MATLAB;Based on the running video of underground electric locomotive in coal mine,rail detection,pipeline detection,obstacle detection,passable area detection and software application experiments are carried out.The experimental results show that the method in this thesis is suitable for the underground environment in coal mine and convenient for the driver to monitor the detection situation in real time.