Improve SegNet with feature pyramid for road scene parsing
Road scene parsing is a common task in semantic segmentation. Its images have characteristics of containing complex scene context and differing greatly among targets of the same category from different scales. To address these problems, we propose a semantic segmentation model combined with edge det...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
EDP Sciences
2021-01-01
|
| Series: | E3S Web of Conferences |
| Online Access: | https://www.e3s-conferences.org/articles/e3sconf/pdf/2021/36/e3sconf_aepee2021_03012.pdf |
| _version_ | 1818605869326139392 |
|---|---|
| author | Ai Xinbo Xie Yunhao He Yinan Zhou Yi |
| author_facet | Ai Xinbo Xie Yunhao He Yinan Zhou Yi |
| author_sort | Ai Xinbo |
| collection | DOAJ |
| description | Road scene parsing is a common task in semantic segmentation. Its images have characteristics of containing complex scene context and differing greatly among targets of the same category from different scales. To address these problems, we propose a semantic segmentation model combined with edge detection. We extend the segmentation network with an encoder-decoder structure by adding an edge feature pyramid module, namely Edge Feature Pyramid Network (EFPNet, for short). This module uses edge detection operators to get boundary information and then combines the multiscale features to improve the ability to recognize small targets. EFPNet can make up the shortcomings of convolutional neural network features, and it helps to produce smooth segmentation. After extracting features of the encoder and decoder, EFPNet uses Euclidean distance to compare the similarity between the presentation of the encoder and the decoder, which can increase the decoder’s ability to restore from the encoder. We evaluated the proposed method on Cityscapes datasets. The experiment on Cityscapes datasets demonstrates that the accuracies are improved by 7.5% and 6.2% over the popular SegNet and ENet. And the ablation experiment validates the effectiveness of our method. |
| first_indexed | 2024-12-16T14:01:49Z |
| format | Article |
| id | doaj.art-a027b23515ee4b2e8e732ee4e221027d |
| institution | Directory Open Access Journal |
| issn | 2267-1242 |
| language | English |
| last_indexed | 2024-12-16T14:01:49Z |
| publishDate | 2021-01-01 |
| publisher | EDP Sciences |
| record_format | Article |
| series | E3S Web of Conferences |
| spelling | doaj.art-a027b23515ee4b2e8e732ee4e221027d2022-12-21T22:29:04ZengEDP SciencesE3S Web of Conferences2267-12422021-01-012600301210.1051/e3sconf/202126003012e3sconf_aepee2021_03012Improve SegNet with feature pyramid for road scene parsingAi XinboXie YunhaoHe YinanZhou Yi0Beijing Academy of Safety Science and TechnologyRoad scene parsing is a common task in semantic segmentation. Its images have characteristics of containing complex scene context and differing greatly among targets of the same category from different scales. To address these problems, we propose a semantic segmentation model combined with edge detection. We extend the segmentation network with an encoder-decoder structure by adding an edge feature pyramid module, namely Edge Feature Pyramid Network (EFPNet, for short). This module uses edge detection operators to get boundary information and then combines the multiscale features to improve the ability to recognize small targets. EFPNet can make up the shortcomings of convolutional neural network features, and it helps to produce smooth segmentation. After extracting features of the encoder and decoder, EFPNet uses Euclidean distance to compare the similarity between the presentation of the encoder and the decoder, which can increase the decoder’s ability to restore from the encoder. We evaluated the proposed method on Cityscapes datasets. The experiment on Cityscapes datasets demonstrates that the accuracies are improved by 7.5% and 6.2% over the popular SegNet and ENet. And the ablation experiment validates the effectiveness of our method.https://www.e3s-conferences.org/articles/e3sconf/pdf/2021/36/e3sconf_aepee2021_03012.pdf |
| spellingShingle | Ai Xinbo Xie Yunhao He Yinan Zhou Yi Improve SegNet with feature pyramid for road scene parsing E3S Web of Conferences |
| title | Improve SegNet with feature pyramid for road scene parsing |
| title_full | Improve SegNet with feature pyramid for road scene parsing |
| title_fullStr | Improve SegNet with feature pyramid for road scene parsing |
| title_full_unstemmed | Improve SegNet with feature pyramid for road scene parsing |
| title_short | Improve SegNet with feature pyramid for road scene parsing |
| title_sort | improve segnet with feature pyramid for road scene parsing |
| url | https://www.e3s-conferences.org/articles/e3sconf/pdf/2021/36/e3sconf_aepee2021_03012.pdf |
| work_keys_str_mv | AT aixinbo improvesegnetwithfeaturepyramidforroadsceneparsing AT xieyunhao improvesegnetwithfeaturepyramidforroadsceneparsing AT heyinan improvesegnetwithfeaturepyramidforroadsceneparsing AT zhouyi improvesegnetwithfeaturepyramidforroadsceneparsing |