{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T13:30:38Z","timestamp":1790688638318,"version":"4.1.0"},"reference-count":36,"publisher":"Wiley","license":[{"start":{"date-parts":[[2022,8,28]],"date-time":"2022-08-28T00:00:00Z","timestamp":1661644800000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Innovation Team Project of Guangdong Province General University","award":["2021KCXTD068"],"award-info":[{"award-number":["2021KCXTD068"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Computational Intelligence and Neuroscience"],"published-print":{"date-parts":[[2022,8,28]]},"abstract":"<jats:p>Foreign object intrusion is one of the main causes of train accidents that threaten human life and public property. Thus, the real-time detection of foreign objects intruding on the railway is important to prevent the train from colliding with foreign objects. Currently, the detection of railway foreign objects is mainly performed manually, which is prone to negligence and inefficient. In this study, an efficient two-stage framework is proposed for foreign object detection in railway images. In the first stage, a lightweight railway image classification network is established to classify any input railway images into one of two classes: normal or intruded. To enable real-time and accurate classification, we propose an improved inverted residual unit by introducing two improvements to the original inverted residual unit. First, the selective kernel convolution is used to dynamically select kernel size and learn multiscale features from railway images. Second, we employ a lightweight attention mechanism, called the convolutional block attention module, to exploit both spatial and channel-wise relationships between feature maps. In the second stage of our framework, the intruded image is fed to the foreign object detection network to further detect the location and class of the objects in the image. Experimental results confirm that the performance of our classification network is comparable to the widely used baselines, and it obtains outperforming efficiency. Moreover, the performances of the second-stage object detection are satisfying.<\/jats:p>","DOI":"10.1155\/2022\/3749635","type":"journal-article","created":{"date-parts":[[2022,8,28]],"date-time":"2022-08-28T23:20:07Z","timestamp":1661728807000},"page":"1-10","source":"Crossref","is-referenced-by-count":18,"title":["Foreign Object Detection in Railway Images Based on an Efficient Two-Stage Convolutional Neural Network"],"prefix":"10.1155","volume":"2022","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9134-6634","authenticated-orcid":true,"given":"Weixun","family":"Chen","sequence":"first","affiliation":[{"name":"Information Engineering Institute, Guangzhou Railway Polytechnic, Guangzhou 510430, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5500-192X","authenticated-orcid":true,"given":"Siming","family":"Meng","sequence":"additional","affiliation":[{"name":"Information Engineering Institute, Guangzhou Railway Polytechnic, Guangzhou 510430, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuelong","family":"Jiang","sequence":"additional","affiliation":[{"name":"Information Engineering Institute, Guangzhou Railway Polytechnic, Guangzhou 510430, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","reference":[{"key":"1","doi-asserted-by":"publisher","DOI":"10.1007\/s11042-015-2654-x"},{"key":"2","doi-asserted-by":"publisher","DOI":"10.37965\/jait.2020.0027"},{"key":"3","doi-asserted-by":"publisher","DOI":"10.1109\/tim.2021.3116315"},{"key":"4","doi-asserted-by":"publisher","DOI":"10.1109\/ACPR.2013.123"},{"key":"5","doi-asserted-by":"publisher","DOI":"10.3923\/itj.2014.2611.2618"},{"key":"6","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2009.167"},{"key":"7","doi-asserted-by":"publisher","DOI":"10.1109\/tnnls.2018.2876865"},{"key":"8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"9","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3059968"},{"key":"10","doi-asserted-by":"publisher","DOI":"10.1109\/access.2018.2879270"},{"key":"11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00060"},{"key":"12","first-page":"3","article-title":"Cbam: convolutional block attention module","author":"S. Woo"},{"key":"13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"14","article-title":"Yolov3: an Incremental Improvement","author":"J. Redmon","year":"2018"},{"key":"15","doi-asserted-by":"publisher","DOI":"10.1109\/tits.2014.2331347"},{"key":"16","first-page":"180","article-title":"An obstacle detection system for automated trains","author":"M. Ruder"},{"key":"17","doi-asserted-by":"publisher","DOI":"10.1109\/tie.2012.2236994"},{"key":"18","doi-asserted-by":"publisher","DOI":"10.1109\/ICCAS.2008.4694449"},{"key":"19","doi-asserted-by":"publisher","DOI":"10.1109\/ICIP.2016.7533104"},{"key":"20","doi-asserted-by":"publisher","DOI":"10.1109\/tits.2020.2969993"},{"key":"21","doi-asserted-by":"publisher","DOI":"10.1109\/CISP-BMEI51763.2020.9263662"},{"key":"22","first-page":"1007","article-title":"Real-time obstacle detection over rails using deep convolutional neural network","author":"Y. Xu"},{"key":"23","first-page":"21","article-title":"Ssd: single shot multibox detector","author":"W. Liu"},{"key":"24","doi-asserted-by":"publisher","DOI":"10.1109\/DICTA.2015.7371273"},{"key":"25","doi-asserted-by":"publisher","DOI":"10.3390\/s18061916"},{"key":"26","doi-asserted-by":"publisher","DOI":"10.1109\/access.2018.2883704"},{"key":"27","doi-asserted-by":"publisher","DOI":"10.3390\/s19112594"},{"key":"28","doi-asserted-by":"publisher","DOI":"10.3390\/s19143075"},{"key":"29","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.195"},{"key":"30","article-title":"Attention is all you need","volume":"30","author":"A. Vaswani","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"31","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2016.2577031"},{"key":"32","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.106"},{"key":"33","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"34","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-009-0275-4"},{"key":"35","article-title":"An image is 11 worth 16x16 words: transformers for image recognition at scale","author":"A. Dosovitskiy"},{"key":"36","first-page":"770","article-title":"Deep residual learning for image recognition","author":"K. He"}],"container-title":["Computational Intelligence and Neuroscience"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/downloads.hindawi.com\/journals\/cin\/2022\/3749635.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/downloads.hindawi.com\/journals\/cin\/2022\/3749635.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/downloads.hindawi.com\/journals\/cin\/2022\/3749635.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,28]],"date-time":"2022-08-28T23:20:14Z","timestamp":1661728814000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.hindawi.com\/journals\/cin\/2022\/3749635\/"}},"subtitle":[],"editor":[{"given":"Yaxiang","family":"Fan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2022,8,28]]},"references-count":36,"alternative-id":["3749635","3749635"],"URL":"https:\/\/doi.org\/10.1155\/2022\/3749635","relation":{},"ISSN":["1687-5273","1687-5265"],"issn-type":[{"value":"1687-5273","type":"electronic"},{"value":"1687-5265","type":"print"}],"subject":[],"published":{"date-parts":[[2022,8,28]]}}}