為了提高含有水漬和落葉的路面裂縫圖像的檢測，提出了一種結合路面裂縫圖像和非對稱Unet（Asy-unet）圖像分割來檢測路面裂縫的訓練模型。裂縫分割是公共基礎設施檢測和維護的重要研究課題。由於路面材質背景不同，以及真實道路上的水漬、落葉等，導致利用公共裂縫數據集和現有的路面裂縫模型識別效果並不理想。在本研究中，設計了一種路面裂縫修復系統來處理實際的路面裂縫問題。在本研究中，使用不同的訓練模型來進行裂紋識別的比較。使用所提出的Asy-unet圖像分割訓練模型對圖像進行下採樣、拼接時，圖像不會丟失太多裂紋細節。系統就可以在適當的範圍內準確地完成分割任務。與裂紋分割中常用的Unet訓練模型及其變體相比，該訓練模型有效提高了裂紋圖像的IoU（Intersection over Union）評估。 此外，整個裂縫檢測過程希望被設計為真實路面裂縫修復的系統，以提供便捷高效的路面裂縫修復評估機制。利用深度學習以像素為單位標記路面裂縫併計算裂縫的大致面積，可以估算裂縫路面修復的成本。 可為後續道路修復施工提供參考。

In order to improve the detection of pavement crack images containing water stains and fallen leaves, a training model that combines pavement crack images and an Asymmetric Unet (Asy-unet) image segmentation to detect pavement crack is proposed. Crack segmentation is an important research topic for detection and maintenance of public infrastructures. Due to different pavement material backgrounds, as well as the water stains and fallen leaves on the real road, it leads to unsatisfactory recognition by using the public crack datasets and existing pavement crack models. In this study, a system for pavement crack repairs is designed to deal with real pavement crack problems. In this study, different training models are used for the comparison of crack identification. The image will not lose too much crack details when the image is down-sampled, and copy and crop by using the proposed Asy-unet image segmentation training model. Then, the system can accurately complete the segmentation task within an appropriate range. Comparing with the Unet training model commonly used in crack segmentation and its variants, the proposed training model effectively improves the IoU (Intersection over Union) evaluation of crack images. In addition, the entire crack detection process is expected to be designed as a system for real pavement crack repairs, which provides a convenient and efficient pavement crack repair evaluation mechanism. Deep learning is used to mark pavement cracks in units of pixels and calculate the approximate area of cracks and it is possible to estimate the cost of cracked pavement repairs. It can be employed as a reference for subsequent road repair construction.

References
[1] D. Feng et al., "Deep multi-modal object detection and semantic segmentation for autonomous driving: datasets, methods, and challenges," IEEE Transactions on Intelligent Transportation Systems, vol. 22, no. 3, pp. 1341-1360, March 2021, doi: 10.1109/TITS.2020.2972974.
[2] H. Wang et al., "SFNet-N: An improved SFNet algorithm for semantic segmentation of low-light autonomous driving road scenes," IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 11, pp. 21405-21417, Nov. 2022, doi: 10.1109/TITS.2022.3177615.
[3] D. Yi, H. Fang, Y. Hua, J. Su, M. Quddus, and J. Han, "Improving synthetic to realistic semantic segmentation with parallel generative ensembles for autonomous urban driving," IEEE Transactions on Cognitive and Developmental Systems, vol. 14, no. 4, pp. 1496-1506, Dec. 2022, doi: 10.1109/TCDS.2021.3117925.
[4] H. Huan, Y. Liu, Y. Xie, C. Wang, D. Xu, and Y. Zhang, "MAENet: Multiple attention encoder–decoder network for farmland segmentation of remote sensing images," IEEE Geoscience and Remote Sensing Letters, vol. 19, pp. 1-5, 2022, doi: 10.1109/LGRS.2021.3137522.
[5] T. Anand, S. Sinha, M. Mandal, V. Chamola, and F. R. Yu, "AgriSegNet: Deep aerial semantic segmentation framework for IoT-assisted precision agriculture," IEEE Sensors Journal, vol. 21, no. 16, pp. 17581-17590, 15 Aug.15, 2021, doi: 10.1109/JSEN.2021.3071290.
[6] M. A. Khan et al., "An optimized method for segmentation and classification of apple diseases based on strong correlation and genetic algorithm based feature selection," IEEE Access, vol. 7, pp. 46261-46277, 2019, doi: 10.1109/ACCESS.2019.2908040.
[7] S. Gonzalez, C. Arellano, and J. E. Tapia, "Deepblueberry: Quantification of blueberries in the wild using instance segmentation," IEEE Access, vol. 7, pp. 105776-105788, 2019, doi: 10.1109/ACCESS.2019.2933062.
[8] C. Yu, Y. Liu, X. Xia, D. Lan, X. Liu, and S. Wu, "Precise and fast segmentation of offshore farms in high-resolution SAR Images based on model fusion and half-precision parallel inference," IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 15, pp. 4861-4872, 2022, doi: 10.1109/JSTARS.2022.3181355.
[9] G. Qin, S. Wang, F. Wang, Y. Zhou, Z. Wang, and W. Zou, "U_EFF_NET: High-precision segmentation of offshore farms from high-resolution SAR remote sensing images," IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 15, pp. 8519-8528, 2022, doi: 10.1109/JSTARS.2022.3208185.
[10] H. -Y. Lin, S. -L. Tseng, and J. -Y. Li, "SUR-Net: A deep network for fish detection and segmentation with limited training data," IEEE Sensors Journal, vol. 22, no. 18, pp. 18035-18044, 15 Sept.15, 2022, doi: 10.1109/JSEN.2022.3195365.
[11] L. Sun, B. Luo, T. Liu, Y. Liu, and Y. Wei, "Algorithm of adaptive fast clustering for fish swarm color image segmentation," IEEE Access, vol. 7, pp. 178753-178762, 2019, doi: 10.1109/ACCESS.2019.2956988.
[12] P. Yin, R. Yuan, Y. Cheng, and Q. Wu, "Deep guidance network for biomedical image segmentation," IEEE Access, vol. 8, pp. 116106-116116, 2020, doi: 10.1109/ACCESS.2020.3002835.
[13] M. Arsalan, T. M. Khan, S. S. Naqvi, M. Nawaz, and I. Razzak, "Prompt deep light-weight vessel segmentation network (PLVS-Net)," IEEE/ACM Transactions on Computational Biology and Bioinformatics, vol. 20, no. 2, pp. 1363-1371, 1 March-April 2023, doi: 10.1109/TCBB.2022.3211936.
[14] Z. Zhou, M. R. Siddiquee, N. Tajbakhsh, and J. Liang, "UNet++: a nested U-Net architecture for medical image segmentation," 4th Deep Learning in Medical Image Analysis (DLMIA) Workshop, vol. 11045, pp. 3-11, 2018.
[15] D. Yu, S. Ji, X. Li, Z. Yuan, and C. Shen, "Earthquake crack detection from aerial images using a deformable convolutional neural network," IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1-12, 2022, Art no. 4412012, doi: 10.1109/TGRS.2022.3183157.
[16] M. Panta, M. T. Hoque, M. Abdelguerfi, and M. C. Flanagin, "IterLUNet: Deep learning architecture for pixel-wise crack detection in levee systems," IEEE Access, vol. 11, pp. 12249-12262, 2023, doi: 10.1109/ACCESS.2023.3241877.
[17] D. Stromer, A. Vetter, H. C. Oezkan, C. Probst, and A. Maier, "Enhanced crack segmentation (eCS): A reference algorithm for segmenting cracks in multicrystalline silicon solar cells," IEEE Journal of Photovoltaics, vol. 9, no. 3, pp. 752-758, May 2019, doi: 10.1109/JPHOTOV.2019.2895808.
[18] X. Chen et al., "Automatic crack segmentation and feature extraction in electroluminescence images of solar modules," IEEE Journal of Photovoltaics, vol. 13, no. 3, pp. 334-342, May 2023, doi: 10.1109/JPHOTOV.2023.3249970.
[19] C. Zhang, M. Jamshidi, C. -C. Chang, X. Liang, Z. Chen, and W. Gui, "Concrete crack quantification using voxel-based reconstruction and bayesian data fusion," IEEE Transactions on Industrial Informatics, vol. 18, no. 11, pp. 7512-7524, Nov. 2022, doi: 10.1109/TII.2022.3147814.
[20] Z. Qu, S. -Q. Chen, Y. -Q. Liu, and L. Liu, "Linear seam elimination of tunnel crack images based on statistical specific pixels ratio and adaptive fragmented segmentation," IEEE Transactions on Intelligent Transportation Systems, vol. 21, no. 9, pp. 3599-3607, Sept. 2020, doi: 10.1109/TITS.2019.2929483.
[21] Q. Zhou, Z. Qu, Y. -X. Li, and F. -R. Ju, "Tunnel crack detection with linear seam based on mixed attention and multiscale feature fusion," IEEE Transactions on Instrumentation and Measurement, vol. 71, pp. 1-11, 2022, doi: 10.1109/TIM.2022.3184351.
[22] S. Meng, S. Kuang, Z. Ma, and Y. Wu, "MtlrNet: An effective deep multitask learning architecture for rail crack detection," IEEE Transactions on Instrumentation and Measurement, vol. 71, pp. 1-10, 2022, doi: 10.1109/TIM.2022.3181940.
[23] S. Wang, Y. Pan, M. Chen, Y. Zhang, and X. Wu, "FCN-SFW: Steel structure crack segmentation using a fully convolutional network and structured forests," IEEE Access, vol. 8, pp. 214358-214373, 2020, doi: 10.1109/ACCESS.2020.3040939.
[24] D. Qiu, M. Xiao, S. Wan, C. Qin, and Z. Zhu, "Pavement crack detection in infrared Images using a DCNN and CCL algorithm," IEEE Sensors Journal, vol. 23, no. 5, pp. 4548-4555, 1 March1, 2023, doi: 10.1109/JSEN.2022.3161104.
[25] Z. Hong et al., "Highway crack segmentation from unmanned aerial vehicle images using deep learning," IEEE Geoscience and Remote Sensing Letters, vol. 19, pp. 1-5, 2022, doi: 10.1109/LGRS.2021.3129607.
[26] J. Cheng, W. Xiong, W. Chen, Y. Gu, and Y. Li, "Pixel-level crack detection using U-Net," TENCON 2018 - 2018 IEEE Region 10 Conference, Jeju, Korea (South), 2018, pp. 0462-0466, doi: 10.1109/TENCON.2018.8650059.
[27] R. Augustaukas and A. Lipnickas, "Pixel-wise road pavement defects detection using U-Net deep neural network," 2019 10th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS), 2019, pp. 468-471, doi: 10.1109/IDAACS.2019.8924337.
[28] V. Polovnikov, D. Alekseev, I. Vinogradov, and G. V. Lashkia, "DAUNet: Deep augmented neural network for pavement crack segmentation," IEEE Access, vol. 9, pp. 125714-125723, 2021, doi: 10.1109/ACCESS.2021.3111223.
[29] W. Qiao, Q. Liu, X. Wu, B. Ma, and G. Li, "Automatic pixel-level pavement crack recognition using a deep feature aggregation segmentation network with a scSE attention mechanism module," Sensors, 2021, doi: 10.3390/s21092902.
[30] D. Ma, H. Fang, N. Wang, C. Zhang, J. Dong and H. Hu, "Automatic detection and counting system for pavement cracks based on PCGAN and YOLO-MF," IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 11, pp. 22166-22178, Nov. 2022, doi: 10.1109/TITS.2022.3161960.
[31] Y. Hou et al., "A deep learning method for pavement crack identification based on limited field images," IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 11, pp. 22156-22165, Nov. 2022, doi: 10.1109/TITS.2022.3160524.
[32] J. Chen, N. Zhao, R. Zhang, L. Chen, K. Huang and Z. Qiu, "Refined crack detection via LECSFormer for autonomous road inspection vehicles," IEEE Transactions on Intelligent Vehicles, vol. 8, no. 3, pp. 2049-2061, March 2023, doi: 10.1109/TIV.2022.3204583.
[33] X. Sun, Y. Xie, L. Jiang, Y. Cao and B. Liu, "DMA-Net: DeepLab with multi-scale attention for pavement crack segmentation," IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 10, pp. 18392-18403, Oct. 2022, doi: 10.1109/TITS.2022.3158670.
[34] C. Han, T. Ma, J. Huyan, X. Huang and Y. Zhang, "CrackW-Net: A novel pavement crack image segmentation convolutional neural network," IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 11, pp. 22135-22144, Nov. 2022, doi: 10.1109/TITS.2021.3095507.
[35] S. L. H. Lau, E. K. P. Chong, X. Yang and X. Wang, "Automated pavement crack segmentation using U-Net-based convolutional neural network," IEEE Access, vol. 8, pp. 114892-114899, 2020, doi: 10.1109/ACCESS.2020.3003638.
[36] J. König, M. David Jenkins, P. Barrie, M. Mannion, and G. Morison, "A convolutional neural network for pavement surface crack segmentation using residual connections and attention gating," 2019 IEEE International Conference on Image Processing (ICIP), Taipei, Taiwan, 2019, pp. 1460-1464, doi: 10.1109/ICIP.2019.8803060.
[37] S. Wu, J. Fang, X. Zheng, and X. Li, "Sample and structure-guided network for road crack detection," IEEE Access, vol. 7, pp. 130032-130043, 2019, doi: 10.1109/ACCESS.2019.2940767.
[38] T. Zhang, D. Wang, A. Mullins, and Y. Lu, "Integrated APC-GAN and AttuNet framework for automated pavement crack pixel-level segmentation: A New Solution to Small Training Datasets," IEEE Transactions on Intelligent Transportation Systems, vol. 24, no. 4, pp. 4474-4481, April 2023, doi: 10.1109/TITS.2023.3236247.
[39] 經濟部工業局所屬工業區服務中心代辦挖掘道路修護費一覽表，https://www.moeaidb.gov.tw/iphw/downLoad?index=file&uid=MDUxMjAwMXw1MDR8NjQ4
[40] Crack segmentation Dataset, https://www.kaggle.com/datasets/lakshaymiddha/ crack-segmentation-dataset