شناسايي ترك در سازه هاي مصالح بنايي به كمك بينايي كامپيوتر بر اساس يادگيري عميق

Crack detection in masonry structures using computer vision based on deep learning

سازه‌هاي مصالح بنايي، شمار بالايي از سازه‌هاي ساخته‌ي دست بشر را تشكيل مي‌دهند. با توجه به اهميت تاريخي و همچنين شرايط محيطي، بسياري از سازه‌هاي مذكور نيازمند يك سيستم پايش سلامت كارا هستند. پيشرفت‌هاي اخير در حوزه‌ي هوش مصنوعي و بينايي رايانه، دستيابي به سيستم موردنظر را بيش از پيش نزديك كرده است. در پژوهش حاضر، فرايند شناسايي ترك در تصاوير سازه‌هاي مصالح بنايي با استفاده از يك مدل يادگيري عميق انجام پذيرفته است. در توسعه‌ي الگوريتم مذكور، از يك مدل رمزگذار ـ رمزگشاي كانولوشنال استفاده شده است كه خروجي آن، تصويري بخش‌بندي شده از تصوير ورودي بوده است. در پژوهش حاضر، مجموعه‌ي داده‌يي متشكل از 115 تصوير از سازه‌هاي مصالح بنايي كه وقوع ترك در آن‌ها رخ داده است، گردآوري و پس از برچسب‌گذاري دستي آن‌ها، اقدام به آموزش و ارزيابي شده است. درنهايت، دقت‌هاي صحت، فراخواني و معيار F1 به ترتيب برابر%81/444، %71/411 و %75/366 براي داده‌هاي ارزيابي و در مقياس پيكسل حاصل شده است.

Masonry structures comprise a large proportion of human-made building stocks around the world. In many cases, aged masonry structures have been found to be vulnerable to earthquakes and seismic loads. Due to the historical importance and vulnerable conditions of these structures, an efficient structural health monitoring system is required to detect every sign of degradation. Thus, a suitable restoration scheme could be taken into account. Manual visual inspection is one of the earliest monitoring schemes used to inspect these structures. Due to the limitations and dangers imposed by using human resources, new strategies are required to achieve this purpose. Recent developments in artificial Intelligence and computer vision have helped researchers develop a new generation of autonomous inspection systems. In the present study, we are going to use a deep learning model with an encoder-decoder architecture to automate crack detection in masonry structure images. In the current study, semantic segmentation is proposed as a detailed solution to accurately predict the location and condition of cracks in masonry images. In the development of the main model of the study, we used EfficientNet-B3 as the encoder while the decoder was defined according to U-Net’s expansion path in order to predict the accurate segmentation mask for the corresponding input images. For training and evaluation of the proposed model, a dataset composed of 115 images is generated and manually annotated. In the proposed method, transfer learning is used to train the model and the data augmentation techniques are implemented to achieve the optimal results on the present dataset. Furthermore, using the Dice-Coefficient loss function directly optimizes the model for F1-Score, which is the main evaluation parameter in semantic segmentation tasks. Finally, the evaluation demonstrated 81.444% Precision, 71.411% Recall, and 75.366% F1-Score for the never-seen test data. The study shows that the deep learning approach can be accurate and trustworthy for this task. Also, the limited number of training data and the complex background images in the dataset prove the robustness of the proposed model.