LANE CLASSIFICATION METHOD USING A STATISTICAL MODEL OF HIS(HUE, SATURATION, INTENSITY) COLOR INFORMATION, PARTICULARLY FOR EFFICIENTLY RECOGNIZING COLOR INFORMATION FOR A LANE BY USING AN HIS COLOR SPACE COORDINATE

摘要

PURPOSE: A lane classification method using a statistical model of HIS(Hue, Saturation, Intensity) color information is provided to shorten the processing time of an image information by reducing the amount of calculation and improve the reliability in lane detection by accurately recognizing a lane regardless of a state of the lane.;CONSTITUTION: A lane classification method using a statistical model of HIS(Hue, Saturation, Intensity) color information comprises the steps of: estimating a horizontal vanishing line from a focus distance and an inclination angle parameter in advance; extracting an ROI-LB(Region Of Interest for Lane Boundary) formed in plural blocks by optimally selecting a processing region necessary to the detection of a lane from a vehicle to the horizontal vanishing line region; detecting each lane in the right ROI-LB by using an HT(Hough Transform) and reinforcing an edge after detecting an edge from an image within the ROI-LB; converting a color image of an RGB method into an HIS color model and calculating the converted HIS color model to classify the lane; comparing the threshold value and the average and the variable for each statistical region; setting a clustering region by using the lane detection information processed in gray scale; and displaying the kind of lanes.;COPYRIGHT KIPO 2011