{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,7,30]],"date-time":"2025-07-30T16:05:09Z","timestamp":1753891509935,"version":"3.41.2"},"reference-count":37,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2024,5,16]],"date-time":"2024-05-16T00:00:00Z","timestamp":1715817600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurorobot."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>Accurately counting the number of dense objects in an image, such as pedestrians or vehicles, is a challenging and practical task. The existing density map regression methods based on CNN are mainly used to count a class of dense objects in a single scene. However, in complex traffic scenes, objects such as vehicles and pedestrians usually exist at the same time, and multiple classes of dense objects need to be counted simultaneously.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>To solve the above issues, we propose a new multiple types of dense object counting method based on feature enhancement, which can enhance the features of dense counting objects in complex traffic scenes to realize the classification and regression counting of dense vehicles and people. The counting model consists of the regression subnet and the classification subnet. The regression subnet is primarily used to generate two-channel predicted density maps, mainly including the initial feature layer and the feature enhancement layer, in which the feature enhancement layer can enhance the classification features and regression counting features of dense objects in complex traffic scenes. The classification subnet mainly supervises classifying dense vehicles and people into two feature channels to assist the regression counting task of the regression subnets.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Our method is compared on VisDrone+ datasets, ApolloScape+ datasets, and UAVDT+ datasets. The experimental results show that the method counts two kinds of dense objects simultaneously and outputs a high-quality two-channel predicted density map. The counting performance is better than the state-of-the-art counting network in dense people and vehicle counting.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>In future work, we will further improve the feature extraction ability of the model in complex traffic scenes to classify and count a variety of dense objects such as cars, pedestrians, and non-motor vehicles.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fnbot.2024.1383943","type":"journal-article","created":{"date-parts":[[2024,5,16]],"date-time":"2024-05-16T04:54:06Z","timestamp":1715835246000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Counting dense object of multiple types based on feature enhancement"],"prefix":"10.3389","volume":"18","author":[{"given":"Qiyan","family":"Fu","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Weidong","family":"Min","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Weixiang","family":"Sheng","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chunjiang","family":"Peng","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2024,5,16]]},"reference":[{"key":"ref1","first-page":"504","article-title":"Interactive object counting","author":"Arteta","year":"2014"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"210","DOI":"10.1016\/j.neucom.2019.11.064","article-title":"Crowd counting with crowd attention convolutional neural network","volume":"382","author":"Chen","year":"2020","journal-title":"Neurocomputing"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"306","DOI":"10.3390\/fi13120306","article-title":"An advanced deep learning approach for multi-object counting in urban vehicular environments","volume":"13","author":"Dirir","year":"2021","journal-title":"Future Internet"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"224","DOI":"10.1016\/j.neucom.2021.02.103","article-title":"A survey of crowd counting and density estimation based on convolutional neural network","volume":"472","author":"Fan","year":"2022","journal-title":"Neurocomputing"},{"key":"ref5","first-page":"2685","article-title":"Learning to count with regression Forest and structured labels","author":"Fiaschi","year":"2012"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"245","DOI":"10.1007\/s10707-023-00503-7","article-title":"MSCNet: dense vehicle counting method based on multi-scale dilated Convolution Channel-aware deep network","volume":"28","author":"Fu","year":"2023","journal-title":"GeoInformatica"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"81","DOI":"10.1016\/j.engappai.2015.04.006","article-title":"Fast crowd density estimation with convolutional neural networks","volume":"43","author":"Fu","year":"2015","journal-title":"Eng. App. Art. Int."},{"key":"ref8","doi-asserted-by":"publisher","first-page":"4803","DOI":"10.1109\/TNNLS.2021.3124272","article-title":"Domain-adaptive crowd counting via high-quality image translation and density reconstruction","volume":"34","author":"Gao","year":"2023","journal-title":"IEEE Trans. Neural Networks Learn. Syst."},{"key":"ref9","first-page":"4137","article-title":"Dense object counting in remote sensing images","author":"Gao","year":"2020"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"4822","DOI":"10.1109\/tcyb.2020.3034316","article-title":"Feature-aware adaptation and density alignment for crowd counting in video surveillance","volume":"51","author":"Gao","year":"2021","journal-title":"IEEE Trans. Cyber."},{"key":"ref11","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tgrs.2024.3356492","article-title":"NWPU-MOC: a benchmark for fine-grained multicategory object counting in aerial images","volume":"62","author":"Gao","year":"2024","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref12","first-page":"5222","article-title":"Precise detection in densely packed scenes","author":"Goldman","year":"2019"},{"key":"ref13","first-page":"6133","article-title":"Crowd counting and density estimation by trellis encoder-decoder networks","author":"Jiang","year":"2019"},{"key":"ref14","first-page":"1324","article-title":"Learning to count objects in images","author":"Lempitsky","year":"2010"},{"key":"ref15","first-page":"1091","article-title":"CSRNet: dilated convolutional neural networks for understanding the highly congested scenes","author":"Li","year":"2018"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"1693","DOI":"10.1007\/s00371-019-01769-5","article-title":"Bilateral counting network for single-image object counting","volume":"36","author":"Li","year":"2020","journal-title":"Vis. Comput."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"2671","DOI":"10.1007\/s00371-022-02485-3","article-title":"CCST: crowd counting with Swin transformer","volume":"39","author":"Li","year":"2023","journal-title":"Vis. Comput."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"318","DOI":"10.1109\/TPAMI.2018.2858826","article-title":"Focal loss for dense object detection","volume":"42","author":"Lin","year":"2020","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref19","doi-asserted-by":"publisher","first-page":"8580","DOI":"10.1109\/tmm.2023.3238521","article-title":"FECANet: boosting few-shot semantic segmentation with feature-enhanced context-aware network","volume":"25","author":"Liu","year":"2023","journal-title":"IEEE Trans. Multimed."},{"key":"ref20","first-page":"849","article-title":"Crowd counting using deep recurrent spatial-aware network","author":"Liu","year":"2018"},{"key":"ref21","first-page":"3689","article-title":"Small instance detection by integer programming on object density maps","author":"Ma","year":"2015"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"15794","DOI":"10.1109\/tits.2022.3145467","article-title":"Traffic sign recognition based on semantic scene understanding and structural traffic sign location","volume":"23","author":"Min","year":"2022","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref23","first-page":"2423","article-title":"Density-aware person detection proceedings of and tracking in crowds","author":"Rodriguez","year":"2011"},{"key":"ref24","first-page":"4031","article-title":"Switching convolutional neural network for crowd counting","author":"Sam","year":"2017"},{"volume-title":"Very deep convolutional networks for large-scale image recognition","year":"2014","author":"Simonyan","key":"ref25"},{"key":"ref26","first-page":"67","article-title":"Real-time, embedded scene invariant crowd counting using scale-normalized histogram of moving gradients (HoMG)","author":"Siva","year":"2016"},{"key":"ref27","first-page":"2017","article-title":"Inception\u2013v4, inception res net and the impact of residual connections on learning","author":"Szegedy","year":"2017"},{"key":"ref28","first-page":"660","article-title":"Learning to count with CNN boosting","author":"Walach","year":"2016"},{"key":"ref29","first-page":"167","article-title":"Dynamic mixture of counter network for location-agnostic crowd counting","author":"Wang","year":"2023"},{"key":"ref30","first-page":"3653","article-title":"Fast visual object counting via example-based density estimation","author":"Wang","year":"2016"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"1570","DOI":"10.1109\/LSP.2021.3096119","article-title":"Dilated-scale-aware category-attention Convnet for multi-class object counting","volume":"28","author":"Xu","year":"2021","journal-title":"IEEE Signal Process. Letters"},{"key":"ref32","doi-asserted-by":"publisher","first-page":"3082","DOI":"10.1109\/jiot.2023.3294727","article-title":"An effective lightweight crowd counting method based on an encoder\u2013decoder network for internet of video things","volume":"11","author":"Yi","year":"2024","journal-title":"IEEE Internet Things J."},{"key":"ref33","doi-asserted-by":"publisher","first-page":"1032","DOI":"10.12305\/j.issn.1001-506X.2023.04.12","article-title":"Ship target detection in SAR image based on feature-enhanced network","volume":"45","author":"Zhang","year":"2023","journal-title":"Syst. Eng. Electron."},{"key":"ref34","first-page":"589","article-title":"Single-image crowd counting via multi-column convolutional neural network","author":"Zhang","year":"2016"},{"key":"ref35","doi-asserted-by":"publisher","first-page":"3512","DOI":"10.1049\/ipr2.12175","article-title":"MSR-FAN: multi-scale residual feature-aware network for crowd counting","volume":"15","author":"Zhao","year":"2021","journal-title":"IET Image Process."},{"key":"ref36","first-page":"535","article-title":"CSCNet: a shallow single column network for crowd counting","author":"Zhou","year":"2020"},{"key":"ref37","first-page":"437","article-title":"Visdrone-DET 2018: the vision meets drone object detection in image challenge results","author":"Zhu","year":"2019"}],"container-title":["Frontiers in Neurorobotics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2024.1383943\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,5,16]],"date-time":"2024-05-16T04:54:09Z","timestamp":1715835249000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2024.1383943\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,5,16]]},"references-count":37,"alternative-id":["10.3389\/fnbot.2024.1383943"],"URL":"https:\/\/doi.org\/10.3389\/fnbot.2024.1383943","relation":{},"ISSN":["1662-5218"],"issn-type":[{"type":"electronic","value":"1662-5218"}],"subject":[],"published":{"date-parts":[[2024,5,16]]},"article-number":"1383943"}}