{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,28]],"date-time":"2025-10-28T18:41:51Z","timestamp":1761676911559,"version":"build-2065373602"},"reference-count":25,"publisher":"Institution of Engineering and Technology (IET)","issue":"1","license":[{"start":{"date-parts":[[2020,12,8]],"date-time":"2020-12-08T00:00:00Z","timestamp":1607385600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["ietresearch.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["IET Image Processing"],"published-print":{"date-parts":[[2021,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Nowadays, numerous semantic segmentation techniques were used to complex scenes such as urban streets. However, speed issues are not considered in most of these methods, and real\u2010time methods do not mainly include enough accuracy. In this paper, an efficient semantic segmentation method is proposed, using the feature extractor of a real\u2010time object detection model, Darknet53, as the backbone of DeepLabv3+. By the high accuracy of DeepLabv3+ structure and great efficiency of Darknet53, a mean intersection was obtained over union of 76.3% in Cityscapes test set, and fast inference speed simultaneously (0.178 s per frame on one GTX 1080Ti GPU). A huge imbalance of objects was noticed on Cityscapes dataset. To solve this problem, a Focal Loss like loss function was proposed to concentrate more on the hard difficult pixels. Moreover, an atrous convolution block was proposed to extract more high\u2010level features. Based on the experimental results, it is proved that these changes contribute to a better result on the Cityscapes test set (77.8% mean Intersection over Union) and faster inference speed (0.171 s per frame). Authors' model achieves state\u2010of\u2010art results on Cityscapes test set (79.1% mean Intersection over Union) after fine\u2010tuning on Cityscapes coarsely annotated\u00a0dataset.<\/jats:p>","DOI":"10.1049\/ipr2.12005","type":"journal-article","created":{"date-parts":[[2020,12,9]],"date-time":"2020-12-09T23:51:00Z","timestamp":1607557860000},"page":"57-64","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":12,"title":["An efficient semantic segmentation method based on transfer learning from object detection"],"prefix":"10.1049","volume":"15","author":[{"given":"Wei","family":"Yang","sequence":"first","affiliation":[{"name":"Institute of Optics and Electronics Chinese Academy of Science Chengdu People's Republic of China"},{"name":"University of Chinese Academy of Sciences Beijing People's Republic of China"}]},{"given":"Jianlin","family":"Zhang","sequence":"additional","affiliation":[{"name":"Institute of Optics and Electronics Chinese Academy of Science Chengdu People's Republic of China"}]},{"given":"Zhongbi","family":"Chen","sequence":"additional","affiliation":[{"name":"Institute of Optics and Electronics Chinese Academy of Science Chengdu People's Republic of China"}]},{"given":"Zhiyong","family":"Xu","sequence":"additional","affiliation":[{"name":"Institute of Optics and Electronics Chinese Academy of Science Chengdu People's Republic of China"}]}],"member":"265","published-online":{"date-parts":[[2020,12,8]]},"reference":[{"key":"e_1_2_5_2_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2016.2572683"},{"key":"e_1_2_5_3_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2016.2644615"},{"key":"e_1_2_5_4_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2017.2699184"},{"key":"e_1_2_5_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.660"},{"key":"e_1_2_5_6_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.549"},{"key":"e_1_2_5_7_1","unstructured":"SimonyanK. ZissermanA.:Very deep convolutional networks for large\u2010scale image recognition. Paper presented at international conference on learning representations (ICLR) San Diego 7\u20139 May 2015"},{"key":"e_1_2_5_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_2_5_9_1","doi-asserted-by":"publisher","DOI":"10.1109\/TITS.2017.2750080"},{"key":"e_1_2_5_10_1","unstructured":"PaszkeA. et\u00a0al.:Enet: A deep neural network architecture for real\u2010time semantic segmentation arXiv:1606.02147 2016"},{"key":"e_1_2_5_11_1","doi-asserted-by":"crossref","unstructured":"ZhaoH. et\u00a0al.:ICNet for real\u2010time semantic segmentation on high\u2010resolution images arXiv:1704.08545v1 2017","DOI":"10.1007\/978-3-030-01219-9_25"},{"key":"e_1_2_5_12_1","doi-asserted-by":"crossref","unstructured":"LiuW. et\u00a0al.:SSD: Single shot multibox detector In: Leibe B. Matas J. Sebe N. Welling M. (eds) Computer Vision ECCV 2016. ECCV 2016. Lecture Notes in Computer Science vol 9905. Springer Cham.https:\/\/doi.org\/10.1007\/978-3-319-46448-0_2 2016","DOI":"10.1007\/978-3-319-46448-0_2"},{"key":"e_1_2_5_13_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.91"},{"key":"e_1_2_5_14_1","unstructured":"RedmonJ. FarhadiA.:YOLOv3: An incremental improvement arXiv:1804.02767 2018"},{"key":"e_1_2_5_15_1","doi-asserted-by":"crossref","unstructured":"ChenL.C. et\u00a0al.:Encoder\u2010decoder with atrous separable convolution for semantic image segmentation In: Ferrari V. Hebert M. Sminchisescu C. Weiss Y. (eds) Computer Vision ECCV 2018. ECCV 2018. Lecture Notes in Computer Science vol 11211. Springer Cham.https:\/\/doi.org\/10.1007\/978-3-030-01234-2_49 2018","DOI":"10.1007\/978-3-030-01234-2_49"},{"key":"e_1_2_5_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"e_1_2_5_17_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"e_1_2_5_18_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.350"},{"key":"e_1_2_5_19_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.195"},{"key":"e_1_2_5_20_1","doi-asserted-by":"crossref","unstructured":"Bul\u00f3S.R. et\u00a0al.:In\u2010place activated BatchNorm for memory\u2010optimized training of DNNs arXiv:1712.02616 2017","DOI":"10.1109\/CVPR.2018.00591"},{"key":"e_1_2_5_21_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.324"},{"key":"e_1_2_5_22_1","doi-asserted-by":"crossref","unstructured":"RonnebergerO. et\u00a0al.:U\u2010Net: Convolutional networks for biomedical image segmentation In: Navab N. Hornegger J. Wells W. Frangi A. (eds) Medical Image Computing and Computer\u2010Assisted Intervention MICCAI 2015. MICCAI 2015. Lecture Notes in Computer Science vol 9351. Springer Cham.https:\/\/doi.org\/10.1007\/978-3-319-24574-4_28 2015","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_5_23_1","first-page":"448","volume-title":"JMLR International Conference on Machine Learning (ICML)","author":"Ioffe S.","year":"2016"},{"key":"e_1_2_5_24_1","doi-asserted-by":"crossref","unstructured":"SzegedyC. et\u00a0al.:Going deeper with convolutions IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Boston MA 2015 pp.1\u20139 https:\/\/doi.org\/10.1109\/CVPR.2015.7298594 2015","DOI":"10.1109\/CVPR.2015.7298594"},{"volume-title":"Proceedings of Advances in Neural Information Processing System (NIPS)","year":"2017","author":"Paszke A.","key":"e_1_2_5_25_1"},{"volume-title":"Proceedings of Advances in Neural Information Processing System (NIPS)","year":"2016","author":"Treml M.","key":"e_1_2_5_26_1"}],"container-title":["IET Image Processing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1049\/ipr2.12005","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1049\/ipr2.12005","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/ietresearch.onlinelibrary.wiley.com\/doi\/pdf\/10.1049\/ipr2.12005","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,27]],"date-time":"2025-10-27T12:31:47Z","timestamp":1761568307000},"score":1,"resource":{"primary":{"URL":"https:\/\/ietresearch.onlinelibrary.wiley.com\/doi\/10.1049\/ipr2.12005"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,12,8]]},"references-count":25,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2021,1]]}},"alternative-id":["10.1049\/ipr2.12005"],"URL":"https:\/\/doi.org\/10.1049\/ipr2.12005","archive":["Portico"],"relation":{},"ISSN":["1751-9659","1751-9667"],"issn-type":[{"type":"print","value":"1751-9659"},{"type":"electronic","value":"1751-9667"}],"subject":[],"published":{"date-parts":[[2020,12,8]]},"assertion":[{"value":"2018-08-14","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2020-07-27","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2020-12-08","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}