Research on Deviation Detection of Belt Conveyor Based on Inspection Robot and Deep Learning

Algorithm 1

Conveyor belt edge detection algorithm based on the Hough line transform.

	Input: ROI image ROI_O or ROI_I.
	Output: Coordinates of the start and end points of the proposal line.
(1)	Convert ROI_O or ROI_I into gray image;
(2)	Denoise the gray image by using Gaussian filter;
(3)	Obtain the edge map by using Canny edge detector [29];
(4)	Map the edge points in to the parameter space by using Hough line transform, then set the accumulators;
(5)	Extract the plausible lines by setting a proper accumulator threshold;
(6)	Extract the proposal line from the plausible lines by the prior knowledge of the conveyor belt edge.