{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T06:05:29Z","timestamp":1790661929202,"version":"4.1.0"},"reference-count":65,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T00:00:00Z","timestamp":1790640000000},"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. Plant Sci."],"abstract":"<jats:p>\n                    Accurate recognition and instance segmentation of chili flowers are fundamental to the development of intelligent breeding robots and the improvement of operational efficiency in protected agriculture. However, in unstructured field environments, chili flowers are typically small-scale targets and are frequently affected by occlusion from leaves and branches as well as illumination variations, leading to degraded feature representation and suboptimal segmentation performance in existing models. To address these challenges, this study proposes a high-precision detection and segmentation framework, termed Chili-Seg, built upon the YOLO11n-seg baseline architecture (YOLO: You Only Look Once). To enhance detection and segmentation performance, an Efficient Head is introduced to replace the original decoupled head. By optimizing the parameter-sharing mechanism between classification and regression branches, the proposed head improves both localization accuracy and segmentation quality while maintaining computational efficiency. To further strengthen feature extraction under complex field conditions, a C2f with Dynamic Position Bias (C2DPB) module is designed to replace the original C2PSA module based on Position-Sensitive Attention (PSA) in the backbone network. The dynamic position-bias mechanism enhances the representation of fine-grained morphological characteristics of chili flowers while effectively suppressing background interference, thereby reducing mis-segmentation caused by environmental noise. Moreover, a Generalized Dynamic Feature Pyramid Network (GDFPN) is introduced to reconstruct the neck architecture. By enabling dynamic cross-scale feature interaction and adaptive weighted fusion, the proposed design mitigates feature loss caused by large variations in object scale and significantly improves multiscale feature representation capability. Experimental results on a self-constructed chili flower dataset demonstrate that Chili-Seg achieves improved overall instance-segmentation performance compared with representative YOLO-based baselines, including YOLOv5n-seg and YOLO11nseg; semantic-segmentation models are evaluated separately using their corresponding pixel-level metrics. For detection tasks, the proposed method achieves a precision of 96.8%, mean Average Precision at an Intersection over Union (IoU) threshold of 0.50 (\n                    <jats:italic>mAP<\/jats:italic>\n                    <jats:sub>50<\/jats:sub>\n                    ) of 90.6%, and\n                    <jats:italic>mAP<\/jats:italic>\n                    <jats:sub>50\u201395<\/jats:sub>\n                    , averaged over IoU thresholds from 0.50 to 0.95, of 77.9%, improving the baseline by 1.7, 1.2, and 1.8 percentage points, respectively. For segmentation tasks, it achieves a precision of 96.6%,\n                    <jats:italic>mAP<\/jats:italic>\n                    <jats:sub>50<\/jats:sub>\n                    of 89.8%, and\n                    <jats:italic>mAP<\/jats:italic>\n                    <jats:sub>50\u201395<\/jats:sub>\n                    of 64.7%, with improvements of 2.1, 1.1, and 1.3 percentage points over the original model. Furthermore, the proposed model is deployed on the NVIDIA Jetson AGX Orin platform for real-time evaluation, achieving a processing speed of 102.4 frames per second (FPS). The results indicate that Chili-Seg achieves a favorable trade-off between accuracy, segmentation quality, and real-time performance, providing a robust foundation for subsequent extraction of chili flower phenotypic parameters.\n                  <\/jats:p>","DOI":"10.3389\/fpls.2026.1955273","type":"journal-article","created":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T05:49:59Z","timestamp":1790660999000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Chili-Seg: a high-precision detection and instance segmentation model for chili flowers in complex environments"],"prefix":"10.3389","volume":"17","author":[{"given":"Minqiu","family":"Kuang","sequence":"first","affiliation":[{"name":"College of Electrical and Mechanical Engineering, Hunan Agricultural University","place":["Changsha, China"]},{"name":"Foshan-Zhongke Innovation Research Institute of Intelligent Agriculture and Robotics","place":["Foshan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaojian","family":"Li","sequence":"additional","affiliation":[{"name":"College of Electrical and Mechanical Engineering, Hunan Agricultural University","place":["Changsha, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhizhen","family":"Hu","sequence":"additional","affiliation":[{"name":"College of Electrical and Mechanical Engineering, Hunan Agricultural University","place":["Changsha, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dawei","family":"Liu","sequence":"additional","affiliation":[{"name":"College of Electrical and Mechanical Engineering, Hunan Agricultural University","place":["Changsha, China"]},{"name":"Hunan Provincial Key Laboratory of Intelligent Agricultural Machinery Equipment","place":["Changsha, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yang","family":"Xiang","sequence":"additional","affiliation":[{"name":"College of Electrical and Mechanical Engineering, Hunan Agricultural University","place":["Changsha, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fangping","family":"Xie","sequence":"additional","affiliation":[{"name":"College of Electrical and Mechanical Engineering, Hunan Agricultural University","place":["Changsha, China"]},{"name":"Hunan Provincial Key Laboratory of Intelligent Agricultural Machinery Equipment","place":["Changsha, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuxuan","family":"Zhang","sequence":"additional","affiliation":[{"name":"College of Intelligent Science and Engineering, Beijing University of Agriculture","place":["Beijing, China"]},{"name":"Department of Computer and Electrical Engineering, Mid Sweden University","place":["Sundsvall, Sweden"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xu","family":"Li","sequence":"additional","affiliation":[{"name":"College of Electrical and Mechanical Engineering, Hunan Agricultural University","place":["Changsha, China"]},{"name":"Hunan Provincial Key Laboratory of Intelligent Agricultural Machinery Equipment","place":["Changsha, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,29]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"52","DOI":"10.1016\/j.tplants.2013.09.008","article-title":"Field high-throughput phenotyping: The new crop breeding frontier","volume":"19","author":"Araus","year":"2014","journal-title":"Trends Plant Sci."},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2004.10934","article-title":"Yolov4: Optimal speed and accuracy of object detection","author":"Bochkovskiy","year":"2020","journal-title":"arXiv preprint arXiv:2004.10934"},{"key":"B3","doi-asserted-by":"publisher","first-page":"833","DOI":"10.1007\/978-3-030-01234-249","article-title":"Encoder-decoder with atrous separable convolution for semantic image segmentation","volume":"11211","author":"Chen","year":"2018","journal-title":"Comput. Vision \u2013 ECCV 2018"},{"key":"B4","first-page":"1290","article-title":"Masked-attention mask transformer for universal image segmentation","author":"Cheng","year":"2022"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1016\/j.atech.2025.101435","article-title":"Fruit detection and yield estimation in camellia oleifera based on improved yolov8 and bytetrack algorithm","volume":"12","author":"Dong","year":"2025","journal-title":"Smart Agric. Technol."},{"key":"B6","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2021"},{"key":"B7","doi-asserted-by":"publisher","first-page":"267","DOI":"10.1146\/annurev-arplant-050312-120137","article-title":"Future scenarios for plant phenotyping","volume":"64","author":"Fiorani","year":"2013","journal-title":"Annu. Rev. Plant Biol."},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2107.08430","article-title":"Yolox: Exceeding yolo series in 2021","author":"Ge","year":"2021","journal-title":"arXiv preprint arXiv:2107.08430"},{"key":"B9","first-page":"2961","article-title":"Mask r-cnn","author":"He","year":"2017"},{"key":"B10","first-page":"8523","article-title":"Real-time panoptic segmentation from dense detections","author":"Hou","year":"2020"},{"key":"B11","first-page":"2989","article-title":"Oneformer: One transformer to rule universal image segmentation","author":"Jain","year":"2023"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.34133\/2020\/4152816","article-title":"Convolutional neural networks for image-based high-throughput plant phenotyping: A review","volume":"2020","author":"Jiang","year":"2020","journal-title":"Plant Phenomics"},{"key":"B13","doi-asserted-by":"publisher","first-page":"1221","DOI":"10.1016\/j.cj.2022.09.001","article-title":"Crop phenotyping studies with application to crop monitoring","volume":"10","author":"Jin","year":"2022","journal-title":"Crop J."},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.3908559","article-title":"Ultralytics yolov5. Software, version 7.0","author":"Jocher","year":"2020"},{"key":"B15","article-title":"Ultralytics yolo11. Software, version 11.0.0","author":"Jocher","year":"2024"},{"key":"B16","article-title":"Ultralytics yolov8. Software, version 8.0.0","author":"Jocher","year":"2023"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2606.03748","article-title":"Ultralytics yolo26: Unified real-time end-to-end vision models","author":"Jocher","year":"2026","journal-title":"arXiv preprint arXiv:2606.03748"},{"key":"B18","doi-asserted-by":"publisher","first-page":"70","DOI":"10.1016\/j.compag.2018.02.016","article-title":"Deep learning in agriculture: A survey","volume":"147","author":"Kamilaris","year":"2018","journal-title":"Comput. Electron. Agric."},{"key":"B19","first-page":"4015","article-title":"Segment anything","author":"Kirillov","year":"2023"},{"key":"B20","doi-asserted-by":"publisher","first-page":"262","DOI":"10.6041\/j.issn.1000-1298.2026.10.026","article-title":"Pepper disease and pest identification algorithm based on pdpi-yolo","volume":"57","author":"Kuang","year":"2026","journal-title":"Trans. Chin. Soc. For. Agric. Machinery"},{"key":"B21","doi-asserted-by":"publisher","first-page":"198","DOI":"10.11975\/j.issn.1002-6819.202502114","article-title":"Lightweight chili flower target detection method based on improved yolov8n","volume":"41","author":"Kuang","year":"2025","journal-title":"Trans. Chin. Soc. Agric. Eng."},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2026.1824412","article-title":"Cf-detr: A robust transformer-based framework for small-scale chili flower detection in industrial chili production systems","volume":"17","author":"Kuang","year":"2026","journal-title":"Front. Plant Sci."},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.3390\/agriculture16131422","article-title":"Dcfp-yolo: A dual-backbone feature fusion network for multi-pose chili flower recognition and edge deployment","volume":"16","author":"Kuang","year":"2026","journal-title":"Agriculture"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2026.1778541","article-title":"Advancements and prospects in key technologies for robotic pollination in greenhouse pepper breeding: A review","volume":"17","author":"Kuang","year":"2026","journal-title":"Front. Plant Sci."},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2026.1861155","article-title":"Cfpr-yolo: Chili flower pose estimation for robotic pollination in unstructured environments","volume":"17","author":"Kuang","year":"2026","journal-title":"Front. Plant Sci."},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.1016\/j.atech.2026.102439","article-title":"Cs-obb: An improved oriented bounding box framework for fruit-stem detection and picking point localization of fresh chili peppers","volume":"15","author":"Kuang","year":"2026","journal-title":"Smart Agric. Technol."},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2026.1879612","article-title":"Gcp-yolo: High-precision detection of tiny chili flowers in complex greenhouse scenes","volume":"17","author":"Kuang","year":"2026","journal-title":"Front. Plant Sci."},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2025.114400","article-title":"A refined yolov5n-based method for detecting pepper flower objects integrating transfer learning","volume":"188","author":"Kuang","year":"2026","journal-title":"Appl. Soft Comput."},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2026.115541","article-title":"Ds-yolo: An improved oriented object detection model for joint localization of mature daylily fruits and stems in unstructured environments","volume":"201","author":"Kuang","year":"2026","journal-title":"Appl. Soft Comput."},{"key":"B30","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/jiot.2026.3717932","article-title":"Cmd-yolo: An iot-oriented lightweight vision framework for real-time chili maturity detection and robotic localization on edge devices","volume":"13","author":"Kuang","year":"2026","journal-title":"IEEE Internet Things J."},{"key":"B31","doi-asserted-by":"publisher","first-page":"91","DOI":"10.1007\/s44443-026-00559-z","article-title":"Ddm-yolo: A lightweight oriented detection model for mature daylily fruits in complex environments","volume":"38","author":"Kuang","year":"2026","journal-title":"J. King Saud Univ. Comput. Inf. Sci."},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2209.02976","article-title":"Yolov6: A single-stage object detection framework for industrial applications","author":"Li","year":"2022","journal-title":"arXiv preprint arXiv:2209.02976"},{"key":"B33","doi-asserted-by":"publisher","first-page":"20078","DOI":"10.3390\/s141120078","article-title":"A review of imaging techniques for plant phenotyping","volume":"14","author":"Li","year":"2014","journal-title":"Sensors"},{"key":"B34","doi-asserted-by":"publisher","first-page":"6027","DOI":"10.1109\/ICCV51070.2023.00554","article-title":"Learning to upsample by learning to sample","author":"Liu","year":"2023","journal-title":"Proc. IEEE\/CVF Int. Conf. Comput. Vision"},{"key":"B35","doi-asserted-by":"publisher","DOI":"10.3390\/s24103025","article-title":"Filtering empty video frames for efficient real-time object detection","volume":"24","author":"Liu","year":"2024","journal-title":"Sensors"},{"key":"B36","first-page":"10012","article-title":"Swin transformer: Hierarchical vision transformer using shifted windows","author":"Liu","year":"2021"},{"key":"B37","first-page":"3431","article-title":"Fully convolutional networks for semantic segmentation","author":"Long","year":"2015"},{"key":"B38","doi-asserted-by":"publisher","first-page":"113029","DOI":"10.1016\/j.ress.2026.113029","article-title":"Pipeline posterior scoring module for out-ofdistribution detection via attachable uncertainty quantification","volume":"277","author":"Lu","year":"2026","journal-title":"Reliability Eng. System Saf."},{"key":"B39","doi-asserted-by":"publisher","first-page":"126","DOI":"10.1109\/MSP.2015.2405111","article-title":"Image analysis: The new bottleneck in plant phenotyping","volume":"32","author":"Minervini","year":"2015","journal-title":"IEEE Signal Process Mag."},{"key":"B40","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2021.106533","article-title":"Fruit detection and load estimation of an orange orchard using the yolo models through simple approaches in different imaging and illumination conditions","volume":"191","author":"Mirhaji","year":"2021","journal-title":"Comput. Electron. Agric."},{"key":"B41","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2016.01419","article-title":"Using deep learning for image-based plant disease detection","volume":"7","author":"Mohanty","year":"2016","journal-title":"Front. Plant Sci."},{"key":"B42","doi-asserted-by":"publisher","DOI":"10.1016\/j.atech.2024.100690","article-title":"Grapevine inflorescence segmentation and flower estimation based on computer vision techniques for early yield assessment","volume":"10","author":"Moreira","year":"2025","journal-title":"Smart Agric. Technol."},{"key":"B43","doi-asserted-by":"publisher","first-page":"103","DOI":"10.1186\/s13007-019-0490-0","article-title":"Measuring crops in 3d: Using geometry for plant phenotyping","volume":"15","author":"Paulus","year":"2019","journal-title":"Plant Methods"},{"key":"B44","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1804.02767","article-title":"Yolov3: An incremental improvement","author":"Redmon","year":"2018","journal-title":"arXiv preprint arXiv:1804.02767"},{"key":"B45","first-page":"779","article-title":"You only look once: Unified, real-time object detection","author":"Redmon","year":"2016"},{"key":"B46","first-page":"234","article-title":"U-net: Convolutional networks for biomedical image segmentation","author":"Ronneberger","year":"2015"},{"key":"B47","first-page":"618","article-title":"Grad-cam: Visual explanations from deep networks via gradient-based localization","author":"Selvaraju","year":"2017"},{"key":"B48","doi-asserted-by":"publisher","first-page":"264","DOI":"10.1016\/j.biosystemseng.2020.03.008","article-title":"Instance segmentation of apple flowers using the improved mask r-cnn model","volume":"193","author":"Tian","year":"2020","journal-title":"Biosyst. Eng."},{"key":"B49","doi-asserted-by":"publisher","first-page":"78433","DOI":"10.48550\/arXiv.2502.12524","article-title":"Yolov12: Attention-centric real-time object detectors","volume":"38","author":"Tian","year":"2025","journal-title":"arXiv"},{"key":"B50","doi-asserted-by":"publisher","first-page":"107984","DOI":"10.52202\/079017-3429","article-title":"Yolov10: Real-time end-to-end object detection","volume":"37","author":"Wang","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"B51","first-page":"7464","article-title":"Yolov7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors","author":"Wang","year":"2023"},{"key":"B52","first-page":"1","article-title":"Yolov9: learning what you want to learn using programmable gradient information","author":"Wang","year":"2024"},{"key":"B53","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2022.106864","article-title":"Fusion of mask rcnn and attention mechanism for instance segmentation of apples under complex background","volume":"196","author":"Wang","year":"2022","journal-title":"Comput. Electron. Agric."},{"key":"B54","doi-asserted-by":"publisher","first-page":"3349","DOI":"10.1109\/TPAMI.2020.2983686","article-title":"Deep high-resolution representation learning for visual recognition","volume":"43","author":"Wang","year":"2021","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"B55","article-title":"Crossformer: a versatile vision transformer hinging on cross-scale attention","author":"Wang","year":"2022"},{"key":"B56","doi-asserted-by":"publisher","first-page":"8686","DOI":"10.1038\/s41598-022-12732-1","article-title":"Online recognition and yield estimation of tomato in plant factory based on yolov3","volume":"12","author":"Wang","year":"2022","journal-title":"Sci. Rep."},{"key":"B57","first-page":"12077","article-title":"Segformer: simple and efficient design for semantic segmentation with transformers","author":"Xie","year":"2021"},{"key":"B58","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2211.15444","article-title":"Damo-yolo: a report on real-time object detection design","author":"Xu","year":"2022","journal-title":"arXiv preprint arXiv:2211.15444"},{"key":"B59","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1016\/j.molp.2020.01.008","article-title":"Crop phenomics and high-throughput phenotyping: past decades, current challenges, and future perspectives","volume":"13","author":"Yang","year":"2020","journal-title":"Mol. Plant"},{"key":"B60","doi-asserted-by":"publisher","first-page":"104846","DOI":"10.1016\/j.compag.2019.06.001","article-title":"Fruit detection for strawberry harvesting robot in non-structural environment based on mask-rcnn","volume":"163","author":"Yu","year":"2019","journal-title":"Comput. Electron. Agric."},{"key":"B61","first-page":"112795","article-title":"Dynamic causal graph network for reliable pipeline leak detection","volume-title":"Reliability Engineering & System Safety","author":"Zhang","year":"2026"},{"key":"B62","doi-asserted-by":"publisher","first-page":"27453","DOI":"10.1109\/jiot.2026.3679508","article-title":"Tinyml-enabled iot edge framework with knowledge distillation for weed classification","volume":"12","author":"Zhang","year":"2026","journal-title":"IEEE Internet Things J"},{"key":"B63","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2026.1747863","article-title":"Real-time on-device weed identification using a hardware-efficient lightweight cnn","volume":"17","author":"Zhang","year":"2026","journal-title":"Front. Plant Sci."},{"key":"B64","doi-asserted-by":"publisher","first-page":"8964","DOI":"10.1038\/s41598-026-43534-4","article-title":"Tinyml pipeline for efficient crack classification in uav-based structural health inspections","volume":"16","author":"Zhang","year":"2026","journal-title":"Sci. Rep."},{"key":"B65","first-page":"2881","article-title":"Pyramid scene parsing network","author":"Zhao","year":"2017"}],"container-title":["Frontiers in Plant Science"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1955273\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T05:50:07Z","timestamp":1790661007000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1955273\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,29]]},"references-count":65,"alternative-id":["10.3389\/fpls.2026.1955273"],"URL":"https:\/\/doi.org\/10.3389\/fpls.2026.1955273","relation":{},"ISSN":["1664-462X"],"issn-type":[{"value":"1664-462X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,29]]},"article-number":"1955273"}}