Sidewalk detection method and device based on artificial intelligence, equipment and medium

Abstract

The invention relates to the technical field of artificial intelligence, and provides a sidewalk detection method and device based on artificial intelligence, equipment and a medium, which can more effectively fuse deep layer features and shallow layer features and reduce information loss, thereby improving the overall accuracy of sidewalk detection, increasing the loss of IOU, and so the positionand the size of the detection frame can be more accurately fitted; a non-sidewalk false recall phenomenon is effectively reduced, an integral sidewalk detection recall rate and accuracy are improved,a loss function can evaluate model loss from multiple aspects, the model effect is better, and high-efficiency and accurate automatic detection of the sidewalk is realized based on an artificial intelligence means. In addition, the invention can also be applied to smart traffic, thereby promoting the construction of smart cities. The invention further relates to a blockchain technology, and sidewalk detection results can be stored in blockchain nodes.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to an artificial intelligence-based sidewalk detection method, device, equipment and medium. Background technique [0002] The sidewalk is an important road safety facility, and the automatic detection and recognition of the sidewalk has become one of the important technical links in the intelligent transportation system, and has attracted more and more attention from researchers. [0003] The traditional sidewalk detection mainly uses the target detection algorithm. Although the algorithm can guarantee a certain accuracy rate under normal circumstances, the recall rate is not high and needs to be further improved. [0004] In addition, in special environments such as smog, rainy days, and blurred sidewalk zebra crossings, the accuracy of sidewalk detection based on target detection algorithms is low, and it is difficult to meet the detection needs. Content...

AI Technical Summary

Problems solved by technology

[0003] The traditional sidewalk detection mainly uses the target detection algorithm. Although the algorithm can guarantee a certain accuracy rate under normal circumstances, the recall rate is not high and needs to be further improved.

[0004] In addition, in special environments such as smog, rainy days, and blurred sidewalk zebra crossings, the accuracy of sidewalk detection based on target detection algorithms is low, and it is difficult to meet the detection needs.

Images