{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,2]],"date-time":"2026-05-02T07:21:13Z","timestamp":1777706473846,"version":"3.51.4"},"reference-count":33,"publisher":"SAGE Publications","issue":"5","license":[{"start":{"date-parts":[[2022,6,2]],"date-time":"2022-06-02T00:00:00Z","timestamp":1654128000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["Journal of Intelligent &amp; Fuzzy Systems: Applications in Engineering and Technology"],"published-print":{"date-parts":[[2022,9,22]]},"abstract":"<jats:p>Geometric invariant feature representation plays an indispensable role in the field of image processing and computer vision. Recently, convolution neural networks (CNNs) have witnessed a great research progress, however CNNs do not excel at dealing with geometrically transformed images. Existing methods enhancing the ability of CNNs learning invariant feature representation rely partly on data augmentation or have a relatively weak generalization ability. This paper proposes orientation adaptive kernels (OA kernels) and orientation adaptive max pooling (OA max pooling) that comprise a new topological structure, orientation adaptive neural networks (OACNNs). OA kernels output the orientation feature maps which encode the orientation information of images. OA max pooling max-pools the orientation feature maps by automatically rotating the pooling windows according to their orientation. OA kernels and OA max pooling together allow for the eight orientation response of images to be computed, and then the max orientation response is obtained, which is proved to be a robust rotation invariant feature representation. OACNNs are compared with state-of-the-art methods and consistently outperform them in various experiments. OACNNs demonstrate a better generalization ability, yielding a test error rate 3.14 on the rotated images but only trained on \u201cup-right\u201d images, which outperforms all state-of-the-art methods by a large margin.<\/jats:p>","DOI":"10.3233\/jifs-213051","type":"journal-article","created":{"date-parts":[[2022,6,3]],"date-time":"2022-06-03T12:03:03Z","timestamp":1654257783000},"page":"5749-5758","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":0,"title":["OACNNs: Orientation adaptive convolutional neural networks"],"prefix":"10.1177","volume":"43","author":[{"given":"Xiang","family":"Ye","sequence":"first","affiliation":[{"name":"Beijing University of Posts and Telecommunications"}]},{"given":"Zihang","family":"He","sequence":"additional","affiliation":[{"name":"Beijing University of Posts and Telecommunications"}]},{"given":"Bohan","family":"Li","sequence":"additional","affiliation":[{"name":"Beijing University of Posts and Telecommunications"}]},{"given":"Yong","family":"Li","sequence":"additional","affiliation":[{"name":"Beijing University of Posts and Telecommunications"}]}],"member":"179","published-online":{"date-parts":[[2022,6,2]]},"reference":[{"issue":"2012","key":"e_1_3_1_2_2","first-page":"4","article-title":"Imagenet classificationwith deep convolutional neural networks","volume":"1","author":"Krizhevsky A.","unstructured":"KrizhevskyA., SutskeverI. and HintonG.E., Imagenet classificationwith deep convolutional neural networks, In NIPS 1(2012), 4.","journal-title":"In NIPS"},{"key":"e_1_3_1_3_2","unstructured":"Karen Simonyan and Andrew Zisserman Very deep con-volutional networks for large-scale image recognition In Computer Vision and Pattern Recognition 10 2014."},{"key":"e_1_3_1_4_2","doi-asserted-by":"crossref","unstructured":"HeK. ZhangX. RenS. and SunJ. Deep residual learning for image recognition IVPR pages 770\u2013778 2016.","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_3_1_5_2","doi-asserted-by":"crossref","unstructured":"Yanteng Zhang Qizhi Teng Linbo Qing Yan Liu and Xiao-hai He Lightweight deep residual network for alzheimer\u2019s disease classification using smri slices Journal of Intelligent and Fuzzy Systems 2021.","DOI":"10.3233\/JIFS-211247"},{"key":"e_1_3_1_6_2","doi-asserted-by":"crossref","unstructured":"Runze Song Zhaohui Liu and Chao Wang End-to-end dehazing of traffic sign images using reformulated atmospheric scattering model Journal of Intelligent and Fuzzy Systems 2021.","DOI":"10.3233\/JIFS-210733"},{"key":"e_1_3_1_7_2","doi-asserted-by":"crossref","unstructured":"Ross Girshick Jeff Donahue Trevor Darrell and Jitendra Malik Rich feature hierarchies for accurate object detection and semantic segmentation In Computer Vision and Pattern Recognition 10 2014.","DOI":"10.1109\/CVPR.2014.81"},{"key":"e_1_3_1_8_2","unstructured":"Shaoqing Ren Kaiming He Ross GirshickB. and Jian Sun Faster r-cnn: Towards real-time object detection with region proposal networks IEEE Transactions On Pattern Analysis And Machine Intelligence 2015."},{"key":"e_1_3_1_9_2","doi-asserted-by":"crossref","unstructured":"Wei Liu Dragomir Anguelov Dumitru Erhan Christian Szegedy Scott Reed Cheng-Yang Fu and Alexander BergC. Ssd: Single shot multibox detector In ECCV 12 2016.","DOI":"10.1007\/978-3-319-46448-0_2"},{"key":"e_1_3_1_10_2","doi-asserted-by":"crossref","unstructured":"Hei Law and Jia Deng Cornernet: Detecting objects as paired keypoints In ECCV 8 2018.","DOI":"10.1007\/978-3-030-01264-9_45"},{"key":"e_1_3_1_11_2","doi-asserted-by":"crossref","unstructured":"Sam Hallman and Charless FowlkesC. Oriented edge forests for boundary detection In Computer Vision and Pattern Recognition pages 1732\u20131740 2015.","DOI":"10.1109\/CVPR.2015.7298782"},{"key":"e_1_3_1_12_2","doi-asserted-by":"crossref","unstructured":"Gong Cheng Peicheng Zhou and Junwei Han Rifd-cnn: Rotation-invariant and fisher discriminative convolutional neural networks for object detection In Computer Vision and Pattern Recognition pages 2884\u20132893 2016.","DOI":"10.1109\/CVPR.2016.315"},{"key":"e_1_3_1_13_2","unstructured":"Max Jaderberg Karen Simonyan Andrew Zisserman and Koray Kavukcuoglu Spatial transformer networks In Computer Vision and Pattern Recognition 4 2016."},{"key":"e_1_3_1_14_2","doi-asserted-by":"crossref","unstructured":"Yanzhao Zhou QixiangYe Qiang Qiu and Jianbin Jiao Orientated response networks In Computer Vision and Pattern Recognition pages 4961\u20134970 Honolulu Hawaii USA 2017.","DOI":"10.1109\/CVPR.2017.527"},{"key":"e_1_3_1_15_2","doi-asserted-by":"crossref","unstructured":"David LoweG. Distinctive image features from scale-invariant keypoints International Journal of Computer Vision pages 91\u2013110 12 2004.","DOI":"10.1023\/B:VISI.0000029664.99615.94"},{"key":"e_1_3_1_16_2","first-page":"959","article-title":"Semi-local affine parts for object recognition","volume":"2","author":"Cordelia Schmid","year":"2010","unstructured":"Cordelia Schmid and Jean Ponce, Semi-local affine parts for object recognition, Proc.british Machine Vision Conf 2 (2010), 959\u2013968.","journal-title":"Proc.british Machine Vision Conf"},{"key":"e_1_3_1_17_2","doi-asserted-by":"crossref","unstructured":"Muhammed Jamshed Shahnaj Parvin and Subrina Akter Significant hog-histogram of oriented gradient feature selection for human detection International Journal of Computer Applications 132 2015.","DOI":"10.5120\/ijca2015907704"},{"key":"e_1_3_1_18_2","doi-asserted-by":"crossref","unstructured":"Yucel Uzun Muhammet Balcilar Khudaydad Mahmoodi Feruz Davletov Fatih AmasyaliM. and SirmaYavuz Usage of hog (histograms of oriented gradients) features for victim detection at disaster areas In International Conference on Electrical and Electronics Engineering pages 535\u2013538 2013.","DOI":"10.1109\/ELECO.2013.6713903"},{"key":"e_1_3_1_19_2","doi-asserted-by":"publisher","DOI":"10.4028\/www.scientific.net\/AMR.542-543.937"},{"key":"e_1_3_1_20_2","doi-asserted-by":"crossref","unstructured":"Xu Shen Xinmei Tian Anfeng He Shaoyan Sun and DachengTao Transform-invariant convolutional neural net-works for image classification and search In ACM on Multimedia Conference pages 1345\u20131354 2016.","DOI":"10.1145\/2964284.2964316"},{"key":"e_1_3_1_21_2","unstructured":"Sergey Demyanov James Bailey Ramamohanarao Kotagiri and Christopher Leckie Invariant backpropagation: how to train a transformation-invariant neural network Computer Science 2016."},{"key":"e_1_3_1_22_2","first-page":"2393","article-title":"A two-stage neural network for translation, rotation and size-invariant visual pattern recognition","volume":"4","author":"Ravichandran A.","year":"1991","unstructured":"RavichandranA. and YegnanarayanaB., A two-stage neural network for translation, rotation and size-invariant visual pattern recognition, In International Conference on Acoustics, Speech and Signal Processing 4(1991), 2393\u20132396.","journal-title":"In International Conference on Acoustics, Speech and Signal Processing"},{"key":"e_1_3_1_23_2","doi-asserted-by":"publisher","DOI":"10.1080\/02664760021745"},{"key":"e_1_3_1_24_2","unstructured":"Tak YanW. and Hector Garcia-Molina Sift: a tool for wide-area information dissemination In Usenix Technical Conference 1995."},{"key":"e_1_3_1_25_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2005.188"},{"key":"e_1_3_1_26_2","doi-asserted-by":"crossref","unstructured":"Fabio Bellavia Domenico Tegolo and Emanuele Trucco Improving sift-based descriptors stability to rotations In 2010 20th International Conference on Pattern Recognition pages 3460\u20133463 2010.","DOI":"10.1109\/ICPR.2010.845"},{"key":"e_1_3_1_27_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.jsb.2006.09.007"},{"key":"e_1_3_1_28_2","doi-asserted-by":"crossref","unstructured":"Dmitry Laptev Nikolay Savinov Joachim BuhmannM. and Marc Pollefeys Ti-pooling:transformation-invariant pooling for feature learning in covolutional neural networks In Computer Vision and Pattern Recognition 12 2016.","DOI":"10.1109\/CVPR.2016.38"},{"key":"e_1_3_1_29_2","unstructured":"Luke Taylor and Geoff Nitschke Improving deep learning using generic data augmentation CoRR abs\/1708.06020 2017."},{"key":"e_1_3_1_30_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.actamat.2020.08.046"},{"key":"e_1_3_1_31_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.matchar.2021.111213"},{"key":"e_1_3_1_32_2","doi-asserted-by":"crossref","unstructured":"Chunxiao Fan Yang Li Guijin Wang and Yong Li Learning transformation-invariant representations for image recognition with drop transformation networks 6 (2018) 73357\u201373369.","DOI":"10.1109\/ACCESS.2018.2850965"},{"key":"e_1_3_1_33_2","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2016.2601622"},{"key":"e_1_3_1_34_2","unstructured":"Kaiming He Xiangyu Zhang Shaoqing Ren and Jian Sun Delving deep into rectifiers: Surpassing human-level performance on imagenet classification CoRR abs\/1502.01852 2015."}],"container-title":["Journal of Intelligent &amp; Fuzzy Systems: Applications in Engineering and Technology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.3233\/JIFS-213051","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.3233\/JIFS-213051","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.3233\/JIFS-213051","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T09:46:56Z","timestamp":1777456016000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.3233\/JIFS-213051"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,6,2]]},"references-count":33,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2022,9,22]]}},"alternative-id":["10.3233\/JIFS-213051"],"URL":"https:\/\/doi.org\/10.3233\/jifs-213051","relation":{},"ISSN":["1064-1246","1875-8967"],"issn-type":[{"value":"1064-1246","type":"print"},{"value":"1875-8967","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,6,2]]}}}