{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T15:14:12Z","timestamp":1777043652106,"version":"3.51.4"},"reference-count":50,"publisher":"MDPI AG","issue":"11","license":[{"start":{"date-parts":[[2016,11,11]],"date-time":"2016-11-11T00:00:00Z","timestamp":1478822400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100012774","name":"Innovation Fund Denmark","doi-asserted-by":"publisher","award":["16-2014-0"],"award-info":[{"award-number":["16-2014-0"]}],"id":[{"id":"10.13039\/100012774","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Convolutional neural network (CNN)-based systems are increasingly used in autonomous vehicles for detecting obstacles. CNN-based object detection and per-pixel classification (semantic segmentation) algorithms are trained for detecting and classifying a predefined set of object types. These algorithms have difficulties in detecting distant and heavily occluded objects and are, by definition, not capable of detecting unknown object types or unusual scenarios. The visual characteristics of an agriculture field is homogeneous, and obstacles, like people, animals and other obstacles, occur rarely and are of distinct appearance compared to the field. This paper introduces DeepAnomaly, an algorithm combining deep learning and anomaly detection to exploit the homogenous characteristics of a field to perform anomaly detection. We demonstrate DeepAnomaly as a fast state-of-the-art detector for obstacles that are distant, heavily occluded and unknown. DeepAnomaly is compared to state-of-the-art obstacle detectors including \u201cFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks\u201d (RCNN). In a human detector test case, we demonstrate that DeepAnomaly detects humans at longer ranges (45\u201390 m) than RCNN. RCNN has a similar performance at a short range (0\u201330 m). However, DeepAnomaly has much fewer model parameters and (182 ms\/25 ms =) a 7.28-times faster processing time per image. Unlike most CNN-based methods, the high accuracy, the low computation time and the low memory footprint make it suitable for a real-time system running on a embedded GPU (Graphics Processing Unit).<\/jats:p>","DOI":"10.3390\/s16111904","type":"journal-article","created":{"date-parts":[[2016,11,11]],"date-time":"2016-11-11T10:05:56Z","timestamp":1478858756000},"page":"1904","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":123,"title":["DeepAnomaly: Combining Background Subtraction and Deep Learning for Detecting Obstacles and Anomalies in an Agricultural Field"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3207-164X","authenticated-orcid":false,"given":"Peter","family":"Christiansen","sequence":"first","affiliation":[{"name":"Department of Engineering, Aarhus University, Aarhus 8200, Denmark"}]},{"given":"Lars","family":"Nielsen","sequence":"additional","affiliation":[{"name":"Danske Commodities, Aarhus 8000, Denmark"}]},{"given":"Kim","family":"Steen","sequence":"additional","affiliation":[{"name":"AgroIntelli, Aarhus 8200, Denmark"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1329-1674","authenticated-orcid":false,"given":"Rasmus","family":"J\u00f8rgensen","sequence":"additional","affiliation":[{"name":"Department of Engineering, Aarhus University, Aarhus 8200, Denmark"}]},{"given":"Henrik","family":"Karstoft","sequence":"additional","affiliation":[{"name":"Department of Engineering, Aarhus University, Aarhus 8200, Denmark"}]}],"member":"1968","published-online":{"date-parts":[[2016,11,11]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/1541880.1541882","article-title":"Anomaly detection: A survey","volume":"41","author":"Chandola","year":"2009","journal-title":"ACM Comput. Surv. (CSUR)"},{"key":"ref_2","unstructured":"McLachlan, G.J., and Basford, K.E. (1988). Statistics: Textbooks and Monographs, Dekker."},{"key":"ref_3","unstructured":"Stauffer, C., and Grimson, W.E.L. (1999, January 23\u201325). Adaptive background mixture models for real-time tracking. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Fort Collins, CO, USA."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"773","DOI":"10.1016\/j.patrec.2005.11.005","article-title":"Efficient adaptive density estimation per image pixel for the task of background subtraction","volume":"27","author":"Zivkovic","year":"2006","journal-title":"Pattern Recognit. Lett."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Bouwmans, T., Porikli, F., H\u00f6ferlin, B., and Vacavant, A. (2014). Background Modeling and Foreground Detection for Video Surveillance, CRC Press.","DOI":"10.1201\/b17223"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Kragh, M., J\u00f8rgensen, R.N., and Henrik, P. (2015, January 6\u20139). Object Detection and Terrain Classification in Agricultural Fields Using 3D Lidar Data. Proceedings of the International Conference on Computer Vision Systems, Copenhagen, Denmark.","DOI":"10.1007\/978-3-319-20904-3_18"},{"key":"ref_7","unstructured":"Kragh, M., Christiansen, P., Korthals, T., Jungeblut, T., Karstoft, H., and J\u00f8rgensen, R.N. (2016, January 26\u201329). Multi-Modal Obstacle Detection and Evaluation of Occupancy Grid Mapping in Agriculture. Proceedings of the International Conference on Agricultural Engineering, Aarhus, Denmark."},{"key":"ref_8","first-page":"167","article-title":"Laser scanner based collision prevention system for autonomous agricultural tractor","volume":"13","author":"Oksanen","year":"2015","journal-title":"Agron. Res."},{"key":"ref_9","unstructured":"Krizhevsky, A., Sutskever, I., and Hinton, G. (2012, January 3\u20136). Imagenet classification with deep convolutional neural networks. Proceedings of the Advances in Neural Information Processing Systems, Lake Tahoe, NV, USA."},{"key":"ref_10","unstructured":"Simonyan, K., and Zisserman, A. (2014). Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Szegedy, C., Reed, S., Sermanet, P., Vanhoucke, V., and Rabinovich, A. (2015, January 7\u201312). Going deeper with convolutions. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA.","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient Based Learning Applied to Document Recognition","volume":"86","author":"LeCun","year":"1998","journal-title":"Proc. IEEE"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2015). Deep Residual Learning for Image Recognition. arXiv.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2014, January 6\u201312). Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition. Proceedings of the European Conference on Computer Vision, Zurich, Switzerland.","DOI":"10.1007\/978-3-319-10578-9_23"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Girshick, R., Donahue, J., Darrell, T., Berkeley, U.C., and Malik, J. (2014, January 23\u201328). Rich feature hierarchies for accurate object detection and semantic segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Beijing, China.","DOI":"10.1109\/CVPR.2014.81"},{"key":"ref_16","unstructured":"Ren, S., He, K., Girshick, R., and Sun, J. (2015). Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. arXiv."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Girshick, R. (2015, January 7\u201313). Fast R-CNN. Proceedings of the IEEE International Conference on Computer Vision, Santiago, Chile.","DOI":"10.1109\/ICCV.2015.169"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2015). You Only Look Once: Unified, Real-Time Object Detection. arXiv.","DOI":"10.1109\/CVPR.2016.91"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Long, J., Shelhamer, E., and Darrell, T. (2015, January 7\u201312). Fully Convolutional Networks for Semantic Segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA.","DOI":"10.1109\/CVPR.2015.7298965"},{"key":"ref_20","unstructured":"Torr, P.H.S. (2014, January 24\u201327). Conditional Random Fields as Recurrent Neural Networks. Proceedings of the IEEE International Conference on Computer Vision, Columbus, OH, USA."},{"key":"ref_21","unstructured":"Chen, L.C., Papandreou, G., Kokkinos, I., Murphy, K., and Yuille, A.L. (2014). Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs. arXiv."},{"key":"ref_22","unstructured":"Chen, L.C., Papandreou, G., Kokkinos, I., Murphy, K., and Yuille, A.L. (2016). DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs. arXiv."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Lin, T.Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Doll\u00e1r, P., and Zitnick, C.L. (2014, January 6\u201312). Microsoft COCO: Common Objects in Context. Proceedings of the European Conference on Computer Vision, Zurich, Switzerland.","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"98","DOI":"10.1007\/s11263-014-0733-5","article-title":"The Pascal Visual Object Classes Challenge\u2014A Retrospective","volume":"111","author":"Everingham","year":"2015","journal-title":"Int. J. Comput. Vis."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1007\/s11263-015-0816-y","article-title":"Imagenet large scale visual recognition challenge","volume":"115","author":"Berg","year":"2015","journal-title":"Int. J. Comput. Vis."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Ross, P., English, A., Ball, D., Upcroft, B., Wyeth, G., and Corke, P. (June, January 31). Novelty-based visual obstacle detection in agriculture. Proceedings of the 2014 IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, China.","DOI":"10.1109\/ICRA.2014.6907080"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Ross, P., English, A., Ball, D., Upcroft, B., and Corke, P. (2015, January 26\u201330). Online novelty-based visual obstacle detection for field robotics. Proceedings of the 2014 IEEE International Conference on Robotics and Automation (ICRA), Seattle, WA, USA.","DOI":"10.1109\/ICRA.2015.7139748"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"5096","DOI":"10.3390\/s150305096","article-title":"Detection of bird nests during mechanical weeding by incremental background modeling and visual saliency","volume":"15","author":"Steen","year":"2015","journal-title":"Sensors"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"86","DOI":"10.1016\/j.asoc.2016.03.016","article-title":"Spatio-temporal analysis for obstacle detection in agricultural videos","volume":"45","author":"Campos","year":"2016","journal-title":"Appl. Soft Comput."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Xu, P., Ye, M., Li, X., Liu, Q., Yang, Y., and Ding, J. (2014, January 3\u20137). Dynamic Background Learning Through Deep Auto-encoder Networks. Proceedings of the 22nd ACM International Conference on Multimedia, Orlando, FL, USA.","DOI":"10.1145\/2647868.2654914"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Braham, M., and Van Droogenbroeck, M. (2016, January 23\u201325). Deep background subtraction with scene-specific convolutional neural networks. Proceedings of the 2016 International Conference on Systems, Signals and Image (IWSSIP), Bratislava, Slovakia.","DOI":"10.1109\/IWSSIP.2016.7502717"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Li, G., and Yu, Y. (2016). Deep Contrast Learning for Salient Object Detection. arXiv.","DOI":"10.1109\/CVPR.2016.58"},{"key":"ref_33","unstructured":"Christiansen, P., Kragh, M., Steen, K., Karstoft, H., and J\u00f8rgensen, R.N. (2015). Platform for Evaluating Sensors and Human Detection in Autonomous Mowing Operations. Precis. Agric., submitted."},{"key":"ref_34","first-page":"1330","article-title":"Advanced sensor platform for human detection and protection in autonomous farming","volume":"15","author":"Christiansen","year":"2015","journal-title":"Precis. Agric."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"6","DOI":"10.3390\/jimaging2010006","article-title":"Using Deep Learning to Challenge Safety Standard for Highly Autonomous Machines in Agriculture","volume":"2","author":"Steen","year":"2016","journal-title":"J. Imaging"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Jia, Y., Shelhamer, E., Donahue, J., Karayev, S., Long, J., Girshick, R., Guadarrama, S., and Darrell, T. (2014, January 3\u20137). Caffe: Convolutional Architecture for Fast Feature Embedding. Proceedings of the 22nd ACM International Conference on Multimedia, Orlando, FL, USA.","DOI":"10.1145\/2647868.2654889"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Zeiler, M.D., and Fergus, R. (2013). Visualizing and Understanding Convolutional Networks. arXiv.","DOI":"10.1007\/978-3-319-10590-1_53"},{"key":"ref_38","unstructured":"Donahue, J., Jia, Y., Vinyals, O., Hoffman, J., Zhang, N., Tzeng, E., and Darrell, T. (2014, January 21\u201326). DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition. Proceedings of the 31st International Conference on Machine Learning (ICML-14), Beijing, China."},{"key":"ref_39","unstructured":"Yosinski, J., Clune, J., Bengio, Y., and Lipson, H. (2014, January 8\u201313). How transferable are features in deep neural networks?. Proceedings of the Advances in Neural Information Processing Systems, Montreal, QC, Canada."},{"key":"ref_40","unstructured":"Yu, F., and Koltun, V. (2015). Multi-Scale Context Aggregation by Dilated Convolutions. arXiv."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"033003","DOI":"10.1117\/1.3456695","article-title":"Comparative study of background subtraction algorithms","volume":"19","author":"Benezeth","year":"2010","journal-title":"J. Electron. Imaging"},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1111\/j.2517-6161.1977.tb01600.x","article-title":"Maximum Likelihood from Incomplete Data via the EM Algorithm","volume":"39","author":"Dempster","year":"1977","journal-title":"J. R. Stat. Soc. Ser. B Stat. Methodol."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"861","DOI":"10.1016\/j.patrec.2005.10.010","article-title":"An introduction to ROC analysis","volume":"27","author":"Fawcett","year":"2006","journal-title":"Pattern Recognit. Lett."},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Davis, J., and Goadrich, M. (2006, January 25\u201329). The Relationship Between Precision-Recall and ROC Curves. Proceedings of the 23rd International Conference on Machine Learning, Pittsburgh, PA, USA.","DOI":"10.1145\/1143844.1143874"},{"key":"ref_45","unstructured":"Van Rijsbergen, C.J. (1979). Information Retrieval, Butterworths."},{"key":"ref_46","unstructured":"Dollar, P., Belongie, S., and Perona, P. (September, January 30). The Fastest Pedestrian Detector in the West. Procedings of the British Machine Vision Conference, Aberystwyth, UK."},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"1532","DOI":"10.1109\/TPAMI.2014.2300479","article-title":"Fast Feature Pyramids for Object Detection","volume":"36","author":"Appel","year":"2014","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_48","unstructured":"Nam, W., Doll\u00e1r, P., and Han, J.H. (2014). Local Decorrelation For Improved Detection. arXiv."},{"key":"ref_49","unstructured":"Christiansen, P., S\u00f8rensen, R., Skovsen, S., J\u00e6ger, C.D., J\u00f8rgensen, R.N., Karstoft, H., and Arild Steen, K. (2016, January 26\u201329). Towards Autonomous Plant Production using Fully Convolutional Neural Networks. Procedings of the International Conference on Agricultural Engineering, Aarhus, Denmark."},{"key":"ref_50","doi-asserted-by":"crossref","unstructured":"Mottaghi, R., Chen, X., Liu, X., Cho, N.G., Lee, S.W., Fidler, S., Urtasun, R., and Yuille, A. (2014, January 23\u201328). The Role of Context for Object Detection and Semantic Segmentation in the Wild. Procedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition, Beijing, China.","DOI":"10.1109\/CVPR.2014.119"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/16\/11\/1904\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T19:35:20Z","timestamp":1760211320000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/16\/11\/1904"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2016,11,11]]},"references-count":50,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2016,11]]}},"alternative-id":["s16111904"],"URL":"https:\/\/doi.org\/10.3390\/s16111904","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2016,11,11]]}}}