{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T07:04:39Z","timestamp":1791443079467,"version":"4.3.3"},"reference-count":40,"publisher":"MDPI AG","issue":"11","license":[{"start":{"date-parts":[[2025,6,5]],"date-time":"2025-06-05T00:00:00Z","timestamp":1749081600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Lhasa Science and Technology Plan Project","award":["LSKJ202405"],"award-info":[{"award-number":["LSKJ202405"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Thangka image detection poses unique challenges due to complex iconography, densely packed small-scale elements, and stylized color\u2013texture compositions. Existing detectors often struggle to capture both global structures and local details and rarely leverage domain-specific visual priors. To address this, we propose a frequency- and prior-enhanced detection framework based on YOLOv11, specifically tailored for Thangka images. We introduce a Learnable Lifting Wavelet Block (LLWB) to decompose features into low- and high-frequency components, while LLWB_Down and LLWB_Up enable frequency-guided multi-scale fusion. To incorporate chromatic and directional cues, we design a Color-Gabor Block (CGBlock), a dual-branch attention module based on HSV histograms and Gabor responses, and embed it via the Color-Gabor Cross Gate (C2CG) residual fusion module. Furthermore, we redesign all detection heads with decoupled branches and introduce center-ness prediction, alongside an additional shallow detection head to improve recall for ultra-small targets. Extensive experiments on a curated Thangka dataset demonstrate that our model achieves 89.5% mAP@0.5, 59.4% mAP@[0.5:0.95], and 84.7% recall, surpassing all baseline detectors while maintaining a compact size of 20.9 M parameters. Ablation studies validate the individual and synergistic contributions of each proposed component. Our method provides a robust and interpretable solution for fine-grained object detection in complex heritage images.<\/jats:p>","DOI":"10.3390\/s25113565","type":"journal-article","created":{"date-parts":[[2025,6,5]],"date-time":"2025-06-05T09:43:18Z","timestamp":1749116598000},"page":"3565","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":13,"title":["Enhanced Object Detection in Thangka Images Using Gabor, Wavelet, and Color Feature Fusion"],"prefix":"10.3390","volume":"25","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-9898-918X","authenticated-orcid":false,"given":"Yukai","family":"Xian","sequence":"first","affiliation":[{"name":"School of Information Science and Technology, Tibet University, Lhasa 850000, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-1575-4065","authenticated-orcid":false,"given":"Yurui","family":"Lee","sequence":"additional","affiliation":[{"name":"School of Humanities, Tibet University, Lhasa 850000, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-3829-1677","authenticated-orcid":false,"given":"Te","family":"Shen","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, Tibet University, Lhasa 850000, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ping","family":"Lan","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, Tibet University, Lhasa 850000, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qijun","family":"Zhao","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, Tibet University, Lhasa 850000, China"},{"name":"College of Computer Science, Sichuan University, Chengdu 610065, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liang","family":"Yan","sequence":"additional","affiliation":[{"name":"School of Humanities, Tibet University, Lhasa 850000, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2025,6,5]]},"reference":[{"key":"ref_1","first-page":"91","article-title":"Faster r-cnn: Towards real-time object detection with region proposal networks","volume":"28","author":"Ren","year":"2015","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y., and Berg, A.C. (2016, January 11\u201314). Ssd: Single shot multibox detector. Proceedings of the Computer Vision\u2013ECCV 2016: 14th European Conference, Amsterdam, The Netherlands. Proceedings, Part I.","DOI":"10.1007\/978-3-319-46448-0_2"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2016, January 27\u201330). You only look once: Unified, real-time object detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.91"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Lin, T.Y., Goyal, P., Girshick, R., He, K., and Doll\u00e1r, P. (2017, January 22\u201329). Focal loss for dense object detection. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.","DOI":"10.1109\/ICCV.2017.324"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Cai, Z., and Vasconcelos, N. (2018, January 18\u201323). Cascade r-cnn: Delving into high quality object detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.","DOI":"10.1109\/CVPR.2018.00644"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Tan, M., Pang, R., and Le, Q.V. (2020, January 13\u201319). Efficientdet: Scalable and efficient object detection. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.01079"},{"key":"ref_7","unstructured":"Tian, Z., Shen, C., Chen, H., and He, T. (November, January 27). Fcos: Fully convolutional one-stage object detection. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Seoul, Republic of Korea."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., and Zagoruyko, S. (2020, January 23\u201328). End-to-end object detection with transformers. Proceedings of the European Conference on Computer Vision, Glasgow, UK.","DOI":"10.1007\/978-3-030-58452-8_13"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"3919","DOI":"10.1007\/s00371-024-03397-0","article-title":"A Thangka cultural element classification model based on self-supervised contrastive learning and MS Triplet Attention","volume":"40","author":"Tang","year":"2024","journal-title":"Vis. Comput."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"137","DOI":"10.1186\/s40494-023-00983-8","article-title":"Ancient thangka Buddha face recognition based on the Dlib machine learning library and comparison with secular aesthetics","volume":"11","author":"Yang","year":"2023","journal-title":"Herit. Sci."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"1081","DOI":"10.32604\/iasc.2022.028763","article-title":"Research on the Identification of Hand-Painted and Machine-Printed Thangka Using CBIR","volume":"34","author":"Pan","year":"2022","journal-title":"Intell. Autom. Soft Comput."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"89687","DOI":"10.1109\/ACCESS.2022.3201086","article-title":"Thangka image segmentation method based on enhanced receptive field","volume":"10","author":"Wang","year":"2022","journal-title":"IEEE Access"},{"key":"ref_13","first-page":"1","article-title":"Weakly Supervised Semantic Segmentation for Headdress of Thangka Images","volume":"39","author":"Hu","year":"2023","journal-title":"J. Inf. Sci. Eng."},{"key":"ref_14","first-page":"1129","article-title":"Thangka image inpainting algorithm based on wavelet transform and structural constraints","volume":"16","author":"Yao","year":"2020","journal-title":"J. Inf. Process. Syst."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Hsieh, T., Zhao, Q., Pan, F., Danzeng, P., Gao, D., and Dorji, G. (2024, January 15\u201319). Text and Edge Guided Thangka Image Inpainting with Diffusion Model. Proceedings of the 2024 IEEE International Conference on Multimedia and Expo (ICME), Niagara Falls, ON, Canada.","DOI":"10.1109\/ICME57554.2024.10687698"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Wang, K., Wang, T., Guo, X., Xu, K., and Wu, J. (2024). Thangka Image\u2014Text Matching Based on Adaptive Pooling Layer and Improved Transformer. Appl. Sci., 14.","DOI":"10.3390\/app14020807"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Hu, W., Qiao, L., Kang, W., and Shi, X. (2023). Thangka image captioning based on semantic concept prompt and multimodal feature optimization. J. Imaging, 9.","DOI":"10.3390\/jimaging9080162"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Li, H., Fang, J., Jia, Y., Ji, L., Chen, X., and Wang, N. (2023). Thangka sketch colorization based on multi-level adaptive-instance-normalized color fusion and skip connection attention. Electronics, 12.","DOI":"10.3390\/electronics12071745"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"732","DOI":"10.1080\/09540091.2022.2041554","article-title":"Few shot object detection for headdresses and seats in Thangka Yidam based on ResNet and deformable convolution","volume":"34","author":"Wenjin","year":"2022","journal-title":"Connect. Sci."},{"key":"ref_20","first-page":"245","article-title":"Application of YOLOv5 neural network based on improved attention mechanism in recognition of Thangka image defects","volume":"16","author":"Fan","year":"2022","journal-title":"KSII Trans. Internet Inf. Syst. (TIIS)"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Xian, Y., Zhao, Q., Yang, X., and Gao, D. (2024, January 25\u201327). Thangka Image Object Detection Method Based on Improved YOLOv8. Proceedings of the 2024 13th International Conference on Computing and Pattern Recognition, Tianjin, China.","DOI":"10.1145\/3704323.3704365"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Wang, S., and Fan, F. (2023). Thangka hyperspectral image super-resolution based on a spatial\u2013spectral integration network. Remote Sens., 15.","DOI":"10.3390\/rs15143603"},{"key":"ref_23","unstructured":"Jocher, G., and Qiu, J. (2025, January 25). Ultralytics YOLOv11. Version 11.0.0. Available online: https:\/\/github.com\/ultralytics\/ultralytics."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"350","DOI":"10.14429\/dsj.64.4503","article-title":"A quaternionic wavelet transform-based approach for object recognition","volume":"64","author":"Priyadharshini","year":"2014","journal-title":"Def. Sci. J."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Wang, H., and Xin, Y. (2020). Wavelet-based contourlet transform and kurtosis map for infrared small target detection in complex background. Sensors, 20.","DOI":"10.3390\/s20030755"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Alaba, S.Y., and Ball, J.E. (2022). Wcnn3d: Wavelet convolutional neural network-based 3d object detection for autonomous driving. Sensors, 22.","DOI":"10.20944\/preprints202209.0060.v1"},{"key":"ref_27","unstructured":"Finder, S.E., Amoyal, R., Treister, E., and Freifeld, O. (October, January 29). Wavelet convolutions for large receptive fields. Proceedings of the European Conference on Computer Vision, Milan, Italy."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"457","DOI":"10.1109\/JAS.2024.124368","article-title":"DI-YOLOv5: An Improved Dual-Wavelet-Based YOLOv5 for Dense Small Object Detection","volume":"12","author":"Li","year":"2025","journal-title":"IEEE\/CAA J. Autom. Sin."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"1443","DOI":"10.1109\/TIP.2006.871081","article-title":"Color object detection using spatial-color joint probability functions","volume":"15","author":"Luo","year":"2006","journal-title":"IEEE Trans. Image Process."},{"key":"ref_30","unstructured":"Khan, F.S., Anwer, R.M., Van De Weijer, J., Bagdanov, A.D., Vanrell, M., and Lopez, A.M. (2012, January 16\u201321). Color attributes for object detection. Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition, Providence, RI, USA."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"10873","DOI":"10.1007\/s11042-019-07970-x","article-title":"Exploiting color name space for salient object detection","volume":"79","author":"Lou","year":"2020","journal-title":"Multimed. Tools Appl."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Gu, P., Lan, X., and Li, S. (2021). Object detection combining cnn and adaptive color prior features. Sensors, 21.","DOI":"10.3390\/s21082796"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"3457","DOI":"10.1016\/j.procs.2024.09.211","article-title":"Colour Channel Separation and Recombination of Images to Improve Object Detection of Diffuse Characters","volume":"246","author":"Roy","year":"2024","journal-title":"Procedia Comput. Sci."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"4357","DOI":"10.1109\/TIP.2018.2835143","article-title":"Gabor convolutional networks","volume":"27","author":"Luan","year":"2018","journal-title":"IEEE Trans. Image Process."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"330","DOI":"10.1016\/j.jvcir.2019.06.005","article-title":"Salient object detection using background subtraction, Gabor filters, objectness and minimum directional backgroundness","volume":"62","author":"Srivastava","year":"2019","journal-title":"J. Vis. Commun. Image Represent."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"1116","DOI":"10.1016\/j.dt.2019.12.002","article-title":"Gabor-CNN for object detection based on small samples","volume":"16","author":"Hu","year":"2020","journal-title":"Def. Technol."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"108495","DOI":"10.1016\/j.patcog.2021.108495","article-title":"Adaptive Gabor convolutional networks","volume":"124","author":"Yuan","year":"2022","journal-title":"Pattern Recognit."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"100197","DOI":"10.1016\/j.jnlest.2023.100197","article-title":"SAR target detection based on the optimal fractional Gabor spectrum feature","volume":"21","author":"Peng","year":"2023","journal-title":"J. Electron. Sci. Technol."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Zhu, J.Y., Park, T., Isola, P., and Efros, A.A. (2017, January 22\u201329). Unpaired image-to-image translation using cycle-consistent adversarial networks. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.","DOI":"10.1109\/ICCV.2017.244"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., and Batra, D. (2017, January 22\u201329). Grad-cam: Visual explanations from deep networks via gradient-based localization. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.","DOI":"10.1109\/ICCV.2017.74"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/25\/11\/3565\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T17:47:10Z","timestamp":1760032030000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/25\/11\/3565"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,5]]},"references-count":40,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2025,6]]}},"alternative-id":["s25113565"],"URL":"https:\/\/doi.org\/10.3390\/s25113565","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,5]]}}}