{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,8]],"date-time":"2026-01-08T09:17:45Z","timestamp":1767863865474,"version":"3.49.0"},"reference-count":22,"publisher":"MDPI AG","issue":"15","license":[{"start":{"date-parts":[[2022,8,4]],"date-time":"2022-08-04T00:00:00Z","timestamp":1659571200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000780","name":"ERA-NET Cofund ICT-AGRI-FOOD","doi-asserted-by":"publisher","award":["41946"],"award-info":[{"award-number":["41946"]}],"id":[{"id":"10.13039\/501100000780","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Ground vehicles equipped with vision-based perception systems can provide a rich source of information for precision agriculture tasks in orchards, including fruit detection and counting, phenotyping, plant growth and health monitoring. This paper presents a semi-supervised deep learning framework for automatic pomegranate detection using a farmer robot equipped with a consumer-grade camera. In contrast to standard deep-learning methods that require time-consuming and labor-intensive image labeling, the proposed system relies on a novel multi-stage transfer learning approach, whereby a pre-trained network is fine-tuned for the target task using images of fruits in controlled conditions, and then it is progressively extended to more complex scenarios towards accurate and efficient segmentation of field images. Results of experimental tests, performed in a commercial pomegranate orchard in southern Italy, are presented using the DeepLabv3+ (Resnet18) architecture, and they are compared with those that were obtained based on conventional manual image annotation. The proposed framework allows for accurate segmentation results, achieving an F1-score of 86.42% and IoU of 97.94%, while relieving the burden of manual labeling.<\/jats:p>","DOI":"10.3390\/s22155821","type":"journal-article","created":{"date-parts":[[2022,8,5]],"date-time":"2022-08-05T02:12:39Z","timestamp":1659665559000},"page":"5821","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":23,"title":["In-Field Automatic Identification of Pomegranates Using a Farmer Robot"],"prefix":"10.3390","volume":"22","author":[{"given":"Rosa Pia","family":"Devanna","sequence":"first","affiliation":[{"name":"Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing, National Research Council, Via G. Amendola 122D\/O, 70126 Bari, Italy"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9456-7590","authenticated-orcid":false,"given":"Annalisa","family":"Milella","sequence":"additional","affiliation":[{"name":"Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing, National Research Council, Via G. Amendola 122D\/O, 70126 Bari, Italy"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5599-903X","authenticated-orcid":false,"given":"Roberto","family":"Marani","sequence":"additional","affiliation":[{"name":"Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing, National Research Council, Via G. Amendola 122D\/O, 70126 Bari, Italy"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-2825-2090","authenticated-orcid":false,"given":"Simone Pietro","family":"Garofalo","sequence":"additional","affiliation":[{"name":"Department of Agricultural and Environmental Science (DiSAAT), University of Bari Aldo Moro, Via G. Amendola 165\/A, 70126 Bari, Italy"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5798-2573","authenticated-orcid":false,"given":"Gaetano Alessandro","family":"Vivaldi","sequence":"additional","affiliation":[{"name":"Department of Agricultural and Environmental Science (DiSAAT), University of Bari Aldo Moro, Via G. Amendola 165\/A, 70126 Bari, Italy"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6699-3485","authenticated-orcid":false,"given":"Simone","family":"Pascuzzi","sequence":"additional","affiliation":[{"name":"Department of Agricultural and Environmental Science (DiSAAT), University of Bari Aldo Moro, Via G. Amendola 165\/A, 70126 Bari, Italy"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7065-9968","authenticated-orcid":false,"given":"Rocco","family":"Galati","sequence":"additional","affiliation":[{"name":"Department of Mechanics, Mathematics & Management, Polytechnic of Bari, Via Orabona 4, 70125 Bari, Italy"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1793-4419","authenticated-orcid":false,"given":"Giulio","family":"Reina","sequence":"additional","affiliation":[{"name":"Department of Mechanics, Mathematics & Management, Polytechnic of Bari, Via Orabona 4, 70125 Bari, Italy"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2022,8,4]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"114","DOI":"10.1016\/j.biosystemseng.2015.12.010","article-title":"Ambient awareness for agricultural robotic vehicles","volume":"146","author":"Reina","year":"2016","journal-title":"Biosyst. Eng."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1039","DOI":"10.1002\/rob.21699","article-title":"Image Segmentation for Fruit Detection and Yield Estimation in Apple Orchards","volume":"34","author":"Bargoti","year":"2017","journal-title":"J. Field Robot."},{"key":"ref_3","first-page":"90","article-title":"Fruit Recognition Using Color and Texture Features","volume":"1","author":"Arivazhagan","year":"2010","journal-title":"J. Emerg. Trends Comput. Inf. Sci."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Zhao, J., Tow, J., and Katupitiya, J. (2005, January 2\u20136). On-Tree Fruit Recognition Using Texture Properties and Color Data. Proceedings of the 2005 IEEE\/RSJ International Conference on Intelligent Robots and Systems, Edmonton, AB, Canada. IROS 263\u2013268.","DOI":"10.1109\/IROS.2005.1545592"},{"key":"ref_5","first-page":"184","article-title":"Shape-Based Fruit Recognition and Classification","volume":"776","author":"Jana","year":"2017","journal-title":"Commun. Comput. Inf. Sci."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"387","DOI":"10.1007\/s11119-020-09736-0","article-title":"Deep neural networks for grape bunch segmentation in natural images from a consumer-grade camera","volume":"22","author":"Marani","year":"2021","journal-title":"Precis. Agric."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"293","DOI":"10.1016\/j.compag.2018.11.026","article-title":"In-Field High Throughput Grapevine Phenotyping with a Consumer-Grade Depth Camera","volume":"156","author":"Milella","year":"2019","journal-title":"Comput. Electron. Agric."},{"key":"ref_8","unstructured":"Bargoti, S., and Underwood, J.P. (June, January 29). Deep Fruit Detection in Orchards. Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), Singapore."},{"key":"ref_9","unstructured":"Yosinski, J., Clune, J., Bengio, Y., and Lipson, H. (2014). How transferable are features in deep neural networks?. arXiv."},{"key":"ref_10","unstructured":"Lee, D.-H. (2013, January 16). Pseudo-Label: The Simple and Efficient Semi-Supervised Learning Method for Deep Neural Networks. Proceedings of the ICML 2013 Workshop on Challenges in Representation Learning, Atlanta, GA, USA."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Heras, J., Marani, R., and Milella, A. (2021, January 18\u201322). Semi-supervised semantic segmentation for grape bunch identification in natural images. Proceedings of the European Conference on Precision Agriculture, Budapest, Hungary.","DOI":"10.3920\/978-90-8686-916-9_39"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Akin, C., Kirci, M., Gunes, E.O., and Cakir, Y. (2012, January 2\u20134). Detection of the Pomegranate Fruits on Tree Using Image Processing. Proceedings of the 2012 1st International Conference on Agro-Geoinformatics, Agro-Geoinformatics, Shanghai, China.","DOI":"10.1109\/Agro-Geoinformatics.2012.6311724"},{"key":"ref_13","first-page":"92","article-title":"A Review of Image Processing for Pomegranate Disease Detection","volume":"6","author":"Bhange","year":"2014","journal-title":"Int. J. Comput. Sci. Inf. Technol."},{"key":"ref_14","first-page":"798","article-title":"Yield Estimation of Pomegranate Using Image Processing Techniques","volume":"8","author":"Behera","year":"2019","journal-title":"Int. J. Innov. Technol. Explor. Eng."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Sharath, D.M., Rohan, M.G., Kumar, S.A., and Prathap, C. (2020, January 15\u201317). Disease Detection in Pomegranate using Image Processing. Proceedings of the 2020 4th International Conference on Trends in Electronics and Informatics (ICOEI) (48184), Tirunelveli, India.","DOI":"10.1109\/ICOEI48184.2020.9142972"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"19043","DOI":"10.1007\/s11042-021-10704-7","article-title":"Fruits yield estimation using Faster R-CNN with MIoU","volume":"80","author":"Behera","year":"2021","journal-title":"Multimed. Tools Appl."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"110","DOI":"10.1016\/j.biosystemseng.2016.11.004","article-title":"Agricultural robots for field operations. Part 2: Operations and systems","volume":"153","author":"Bechar","year":"2017","journal-title":"Biosyst. Eng."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"423","DOI":"10.1007\/s11119-018-9605-2","article-title":"A multi-sensor robotic platform for ground mapping and estimation beyond the visible spectrum","volume":"20","author":"Milella","year":"2019","journal-title":"Precis. Agric."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Fountas, S., Mylonas, N., Malounas, I., Rodias, E., Hellmann Santos, C., and Pekkeriet, E. (2020). Agricultural robotics for field operations. Sensors, 20.","DOI":"10.3390\/s20092672"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Chen, L.C., Zhu, Y., Papandreou, G., Schroff, F., and Adam, H. (2018). Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation. Lecture Notes in Computer Science, Springer (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Springer. LNCS 833\u2013851.","DOI":"10.1007\/978-3-030-01234-2_49"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Casado-Garc\u00eda, A., Heras, J., Milella, A., and Marani, R. (2022). Semi-supervised deep learning and low-cost cameras for the semantic segmentation of natural images in viticulture. Precis. Agric., 1\u201326.","DOI":"10.1007\/s11119-022-09929-9"},{"key":"ref_22","unstructured":"Oltean, M. (2018). Fruits 360 dataset. Mendeley Data, 1."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/22\/15\/5821\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T00:04:01Z","timestamp":1760141041000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/22\/15\/5821"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,8,4]]},"references-count":22,"journal-issue":{"issue":"15","published-online":{"date-parts":[[2022,8]]}},"alternative-id":["s22155821"],"URL":"https:\/\/doi.org\/10.3390\/s22155821","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,8,4]]}}}