Method and device for detecting lane through classification of lane candidate pixels

Abstract

A learning method for detecting at least one lane based on a convolutional neural network (CNN) is provided. The learning method includes steps of: (a) a learning device obtaining encoded feature maps, and information on lane candidate pixels in a input image; (b) the learning device, classifying a first parts of the lane candidate pixels ,whose probability scores are not smaller than a predetermined threshold, as strong line pixels, and classifying the second parts of the lane candidate pixels, whose probability scores are less than the threshold but not less than another predetermined threshold, as weak lines pixels; and (c) the learning device, if distances between the weak line pixels and the strong line pixels are less than a predetermined distance, classifying the weak line pixels aspixels of additional strong lines, and determining that the pixels of the strong line and the additional correspond to pixels of the lane.

Description

technical field [0001] The present invention relates to a CNN (Convolutional Neural Network)-based learning method and learning device for detecting at least one vehicle lane and a testing method and testing device using the same, and specifically relates to the following learning method and learning device as well as a testing method based on it and test device. According to the learning method for detecting at least one vehicle lane based on the CNN (Convolutional Neural Network), it is characterized in that it includes: (a) when the input image is obtained, the learning device makes the coding layer applying at least one convolution operation to obtain at least one encoded feature map, causing the decoding layer to apply at least one deconvolution operation to the specific encoded feature map output from the encoding layer to obtain at least one The step of segmenting the result of the information of the supplementary pixels of the lane; (b) the learning device compares ea...

AI Technical Summary

Problems solved by technology

As described above, the predetermined supplementary pixels that are judged to be one of the lanes with a low probability are not recognized as lanes, so there is a problem that many broken lanes occur and it is difficult to obtain a long line.

Images