{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,12]],"date-time":"2026-08-12T04:31:36Z","timestamp":1786509096811,"version":"build-2736575974"},"reference-count":45,"publisher":"MDPI AG","issue":"16","license":[{"start":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T00:00:00Z","timestamp":1786320000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Youth Innovation Team Research Program Project of the Shaanxi Provincial Department of Education","award":["24JP076"],"award-info":[{"award-number":["24JP076"]}]},{"name":"Shaanxi University Youth Innovation Team Construction Project"},{"DOI":"10.13039\/501100017596","name":"Natural Science Basic Research Program of Shaanxi Province","doi-asserted-by":"crossref","award":["2025JC-YBQN-779"],"award-info":[{"award-number":["2025JC-YBQN-779"]}],"id":[{"id":"10.13039\/501100017596","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Road crack images captured by smartphones suffer from low resolution, uneven illumination, and complex background interference. Mobile devices also have limited resources for real-time high-accuracy segmentation. A two-stage framework combines a high-accuracy dual-path teacher model with a knowledge-distilled lightweight student model. The teacher model integrates a CNN-KAN path for local texture extraction and a VMamba path for global context modeling at linear complexity. A dedicated KAN-based fusion module learns adaptive nonlinear mappings between the two feature streams. On public crack datasets, the teacher model achieves an mIoU of 0.8087 and an mDice of 0.9028. It is then transferred to a self-constructed smartphone dataset built from continuous 30 fps video, where it reaches an mIoU of 0.7084 with strong robustness to illumination and blur. A GAN-based super-resolution strategy further improves the mIoU by 4.01%. A ConvLSTM-based knowledge distillation framework compresses the teacher into a lightweight MobileViT student model. This reduces the parameter count from 57.80 M to 1.57 M and cuts the GPU inference time from 99.56 ms to 1.46 ms, while retaining an mIoU of 0.7078. The deployed student model runs at 15 to 20 frames per second on an Android smartphone. An ablation study confirms that the ConvLSTM-based temporal distillation contributes beyond standard distillation. This framework provides a practical solution for real-time road crack monitoring on smartphones.<\/jats:p>","DOI":"10.3390\/s26165071","type":"journal-article","created":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T15:10:24Z","timestamp":1786374624000},"page":"5071","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Real-Time Road Crack Detection on Smartphones Through ConvLSTM-Based Temporal Knowledge Distillation from a CNN-KAN and VMamba Dual-Path Network"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-5398-9698","authenticated-orcid":false,"given":"Mengzhao","family":"Nie","sequence":"first","affiliation":[{"name":"School of Civil Engineering and Smart Construction, Chang\u2019an University, Xi\u2019an 710061, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9791-3591","authenticated-orcid":false,"given":"Hua","family":"Huang","sequence":"additional","affiliation":[{"name":"School of Civil & Architecture Engineering, Xi\u2019an Technological University, Xi\u2019an 710021, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mengxue","family":"Guo","sequence":"additional","affiliation":[{"name":"School of Civil & Architecture Engineering, Xi\u2019an Technological University, Xi\u2019an 710021, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mingxia","family":"Dang","sequence":"additional","affiliation":[{"name":"School of Civil Engineering and Smart Construction, Chang\u2019an University, Xi\u2019an 710061, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ming","family":"Tang","sequence":"additional","affiliation":[{"name":"School of Civil & Architecture Engineering, Xi\u2019an Technological University, Xi\u2019an 710021, China"},{"name":"Chengdu Construction Sixth Construction Engineering Co., Ltd., Chengdu 610032, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,8,10]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"134220","DOI":"10.1016\/j.conbuildmat.2023.134220","article-title":"Deep residual learning for acoustic emission source localization in a steel-concrete composite slab","volume":"411","author":"Zhou","year":"2023","journal-title":"Constr. Build. Mater."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"111937","DOI":"10.1016\/j.ymssp.2024.111937","article-title":"Ambient vibration measurement-aided multi-1D CNNs ensemble for damage localization framework: Demonstration on a large-scale RC pedestrian bridge","volume":"224","author":"Zhou","year":"2025","journal-title":"Mech. Syst. Signal Process."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"123658","DOI":"10.1016\/j.eswa.2024.123658","article-title":"MultiScaleCrackNet: A parallel multiscale deep CNN architecture for concrete crack classification","volume":"249","author":"Russel","year":"2024","journal-title":"Expert Syst. Appl."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"129238","DOI":"10.1016\/j.conbuildmat.2022.129238","article-title":"Review on computer vision-based crack detection and quantification methodologies for civil structures","volume":"356","author":"Deng","year":"2022","journal-title":"Constr. Build. Mater."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"103989","DOI":"10.1016\/j.autcon.2021.103989","article-title":"Structural crack detection using deep convolutional neural networks","volume":"133","author":"Ali","year":"2022","journal-title":"Autom. Constr."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"105297","DOI":"10.1016\/j.autcon.2024.105297","article-title":"Evaluation and optimisation of pre-trained CNN models for asphalt pavement crack detection and classification","volume":"160","author":"Matarneh","year":"2024","journal-title":"Autom. Constr."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"126265","DOI":"10.1016\/j.conbuildmat.2021.126265","article-title":"UNet-based model for crack detection integrating visual explanations","volume":"322","author":"Liu","year":"2022","journal-title":"Constr. Build. Mater."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"131852","DOI":"10.1016\/j.conbuildmat.2023.131852","article-title":"A novel transformer-based network with attention mechanism for automatic pavement crack detection","volume":"391","author":"Guo","year":"2023","journal-title":"Constr. Build. Mater."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"134134","DOI":"10.1016\/j.conbuildmat.2023.134134","article-title":"A weakly-supervised transformer-based hybrid network with multi-attention for pavement crack detection","volume":"411","author":"Wang","year":"2024","journal-title":"Constr. Build. Mater."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"104297","DOI":"10.1016\/j.dsp.2023.104297","article-title":"SwinCrack: Pavement crack detection using convolutional swin-transformer network","volume":"145","author":"Wang","year":"2024","journal-title":"Digit. Signal Process."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"1743","DOI":"10.1111\/mice.13103","article-title":"A lightweight encoder\u2013decoder network for automatic pavement crack detection","volume":"39","author":"Zhu","year":"2024","journal-title":"Comput. Aided Civ. Infrastruct. Eng."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Liu, Z., Wang, Y., Vaidya, S., Ruehle, F., Halverson, J., Solja\u010di\u0107, M., Hou, T.Y., and Tegmark, M. (2024). Kan: Kolmogorov-arnold networks. arXiv.","DOI":"10.1103\/4t7t-v19l"},{"key":"ref_13","unstructured":"Li, C., Liu, X., Li, W., Wang, C., Liu, H., Liu, Y., Chen, Z., and Yuan, Y. (March, January 25). U-kan makes strong backbone for medical image segmentation and generation. Proceedings of the AAAI Conference on Artificial Intelligence, Philadelphia, PA, USA."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"105","DOI":"10.20517\/ir.2025.07","article-title":"SKPNet: Snake KAN perceive bridge cracks through semantic segmentation","volume":"5","author":"Yang","year":"2025","journal-title":"Intell. Robot."},{"key":"ref_15","unstructured":"Gu, A., and Dao, T. (2024, January 7\u20139). Mamba: Linear-time sequence modeling with selective state spaces. Proceedings of the First Conference on Language Modeling, Philadelphia, PA, USA."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"103031","DOI":"10.52202\/079017-3273","article-title":"Vmamba: Visual state space model","volume":"37","author":"Liu","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"105770","DOI":"10.1016\/j.autcon.2024.105770","article-title":"Enhancing pixel-level crack segmentation with visual mamba and convolutional networks","volume":"168","author":"Han","year":"2024","journal-title":"Autom. Constr."},{"key":"ref_18","unstructured":"Ji, T., Hou, Y., and Zhang, D. (2024). A comprehensive survey on Kolmogorov Arnold networks (KAN). arXiv."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"105482","DOI":"10.1016\/j.autcon.2024.105482","article-title":"CNN-based network with multi-scale context feature and attention mechanism for automatic pavement crack segmentation","volume":"164","author":"Liang","year":"2024","journal-title":"Autom. Constr."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Cheon, M., and Mun, C. (2024). Combining KAN with CNN: KonvNeXt\u2019s performance in remote sensing and patent insights. Remote Sens., 16.","DOI":"10.20944\/preprints202407.0663.v1"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"140793","DOI":"10.1016\/j.conbuildmat.2025.140793","article-title":"Deep learning for automated detection and classification of crack severity level in concrete structures","volume":"472","author":"Shi","year":"2025","journal-title":"Constr. Build. Mater."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"125941","DOI":"10.1016\/j.eswa.2024.125941","article-title":"An ultra-lightweight network combining Mamba and frequency-domain feature extraction for pavement tiny-crack segmentation","volume":"264","author":"Zhang","year":"2025","journal-title":"Expert Syst. Appl."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"105845","DOI":"10.1016\/j.autcon.2024.105845","article-title":"Topology-aware mamba for crack segmentation in structures","volume":"168","author":"Zuo","year":"2024","journal-title":"Autom. Constr."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"1914","DOI":"10.1111\/mice.12881","article-title":"Tiny-Crack-Net: A multiscale feature fusion network with attention mechanisms for segmentation of tiny cracks","volume":"37","author":"Chu","year":"2022","journal-title":"Comput. Aided Civ. Infrastruct. Eng."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"11710","DOI":"10.1109\/TITS.2021.3106647","article-title":"A crack detection algorithm for concrete pavement based on attention mechanism and multi-features fusion","volume":"23","author":"Qu","year":"2021","journal-title":"IEEE Trans. Intell. Transp. Syst."},{"key":"ref_26","first-page":"4467","article-title":"Dual path networks","volume":"30","author":"Chen","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Zhang, L., Yang, F., Zhang, Y.D., and Zhu, Y.J. (2016, January 25\u201328). Road crack detection using deep convolutional neural network. Proceedings of the 2016 IEEE International Conference on Image Processing (ICIP), Phoenix, AZ, USA.","DOI":"10.1109\/ICIP.2016.7533052"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"139","DOI":"10.1016\/j.neucom.2019.01.036","article-title":"DeepCrack: A deep hierarchical feature learning architecture for crack segmentation","volume":"338","author":"Liu","year":"2019","journal-title":"Neurocomputing"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Cui, L., Qi, Z., Chen, Z., Meng, F., and Shi, Y. (2015). Pavement distress detection using random decision forests. Proceedings of the International Conference on Data Science, Sydney, Australia, 8\u20139 August 2015, Springer.","DOI":"10.1007\/978-3-319-24474-7_14"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"227","DOI":"10.1016\/j.patrec.2011.11.004","article-title":"CrackTree: Automatic crack detection from pavement images","volume":"33","author":"Zou","year":"2012","journal-title":"Pattern Recognit. Lett."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"101575","DOI":"10.1016\/j.aei.2022.101575","article-title":"A generative adversarial learning strategy for enhanced lightweight crack delineation networks","volume":"52","author":"Ni","year":"2022","journal-title":"Adv. Eng. Inform."},{"key":"ref_32","unstructured":"Chu, C., Zhmoginov, A., and Sandler, M. (2017). Cyclegan, a master of steganography. arXiv."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3390462","article-title":"A deep journey into super-resolution: A survey","volume":"53","author":"Anwar","year":"2020","journal-title":"ACM Comput. Surv. (CSUR)"},{"key":"ref_34","first-page":"301","article-title":"Crack growth prediction on a concrete structure using deep ConvLSTM","volume":"33","author":"An","year":"2024","journal-title":"Smart Struct. Syst."},{"key":"ref_35","first-page":"8024","article-title":"Pytorch: An imperative style, high-performance deep learning library","volume":"32","author":"Paszke","year":"2019","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Sudre, C.H., Li, W., Vercauteren, T., Ourselin, S., and Jorge Cardoso, M. (2017). Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations. Proceedings of the International Workshop on Deep Learning in Medical Image Analysis, Qu\u00e9bec City, QC, Canada, 14 September 2017, Springer.","DOI":"10.1007\/978-3-319-67558-9_28"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention, Munich, Germany, 5\u20139 October 2015, Springer.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"94","DOI":"10.1016\/j.isprsjprs.2020.01.013","article-title":"ResUNet-a: A deep learning framework for semantic segmentation of remotely sensed data","volume":"162","author":"Diakogiannis","year":"2020","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Zhao, H., Shi, J., Qi, X., Wang, X., and Jia, J. (2017, January 21\u201326). Pyramid scene parsing network. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.","DOI":"10.1109\/CVPR.2017.660"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Lin, T.Y., Doll\u00e1r, P., Girshick, R., He, K., Hariharan, B., and Belongie, S. (2017, January 21\u201326). Feature pyramid networks for object detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.","DOI":"10.1109\/CVPR.2017.106"},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"834","DOI":"10.1109\/TPAMI.2017.2699184","article-title":"Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs","volume":"40","author":"Chen","year":"2017","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Chen, L.C., Zhu, Y., Papandreou, G., Schroff, F., and Adam, H. (2018, January 8\u201314). Encoder-decoder with atrous separable convolution for semantic image segmentation. Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany.","DOI":"10.1007\/978-3-030-01234-2_49"},{"key":"ref_43","unstructured":"Chen, J., Lu, Y., Yu, Q., Luo, X., Adeli, E., Wang, Y., Lu, L., Yuille, A.L., and Zhou, Y. (2021). Transunet: Transformers make strong encoders for medical image segmentation. arXiv."},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Liu, H., Miao, X., Mertz, C., Xu, C., and Kong, H. (2021, January 11\u201317). Crackformer: Transformer network for fine-grained crack detection. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Virtual.","DOI":"10.1109\/ICCV48922.2021.00376"},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"138206","DOI":"10.1016\/j.conbuildmat.2024.138206","article-title":"DefNet: A multi-scale dual-encoding fusion network aggregating Transformer and CNN for crack segmentation","volume":"448","author":"Xiong","year":"2024","journal-title":"Constr. Build. Mater."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/16\/5071\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,12]],"date-time":"2026-08-12T04:18:39Z","timestamp":1786508319000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/16\/5071"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,10]]},"references-count":45,"journal-issue":{"issue":"16","published-online":{"date-parts":[[2026,8]]}},"alternative-id":["s26165071"],"URL":"https:\/\/doi.org\/10.3390\/s26165071","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,10]]}}}