Rural road recognition based on semantic segmentation
In allusion to the problem of insufficient recognition accuracy of surrounding features when intelligent agricultural machinery drives in complex rural environments,an improved PP-LiteSeg model is proposed based on rural road scenes as the research object.The STDC is used to extract features from the image,which can ensure the completeness of the feature information while ensuring the lightweight.The strip pooling is introduced into a simple pyramid module to enhance feature extraction capabilities.The coordinate attention is integrated into the unified attention fusion module to further enhance the fusion of multi-scale features and capture richer information,thereby improving the accuracy of the model in recognizing complex rural scenes.The experiments show that the model can realize better segmentation results in different scenes,and the accuracy rate of individual categories such as buildings,asphalt roads,and obstacles can reach more than 80%,which has can effectively segment the rural road scene.The improved model can provide technical references for the intelligent agricultural machine to drive safely in the rural road scene.
semantic segmentationrural roadfeature recognitionstrip poolingcoordinate attentionscene classificationimage process
万方数据
维普
