Semantic segmentation method for street images with multi-dimensional features
Image Processing|更新时间:2024-07-26
|
Semantic segmentation method for street images with multi-dimensional features
“In the field of deep learning semantic segmentation, this paper proposes a street view image semantic segmentation network MDFNet that integrates multidimensional features. Through the target area enhancement module, feature pyramid grid, and dual solution dock technology, the segmentation accuracy of complex street view images is significantly improved, with an average intersection to union ratio of 80.11%, providing a new solution for street view image understanding.”
Chinese Journal of Liquid Crystals and DisplaysVol. 39, Issue 7, Pages: 980-989(2024)
作者机构:
西安工程大学 电子信息学院, 陕西 西安 710600
作者简介:
基金信息:
National Natural Science Foundation of China(61971339);Shaanxi Province Key R&D Program(2019GY-113);Shaanxi Provincial Natural Science Basic Research Program(2019JQ-361)
ZHU Lei, CHE Chenjie, YAO Tongyu, et al. Semantic segmentation method for street images with multi-dimensional features[J]. Chinese journal of liquid crystals and displays, 2024, 39(7): 980-989.
DOI:
ZHU Lei, CHE Chenjie, YAO Tongyu, et al. Semantic segmentation method for street images with multi-dimensional features[J]. Chinese journal of liquid crystals and displays, 2024, 39(7): 980-989. DOI: 10.37188/CJLCD.2023-0208.
Semantic segmentation method for street images with multi-dimensional features