A Novel Method for Concrete Crack Detection Using Image Processing Technique

Currently, the number of obsolete high-rise buildings around the world is increasing, and majority of them are built using concrete. Concrete can also lose strength due to constant pressure and environmental influences. Damage to the mold's exterior surface may ensue as a result (cracks and chips). Mold integrity may be jeopardised if these defects are left unidentified and untreated. The template must therefore be updated every day. The drone has been employed in some earlier studies, most of which were based on visions, to capture and chronicle how the country is currently shaped. To ascertain whether the application of classification, localisation, and segmentation techniques was damaging, the recorded videos and images were evaluated. In other research, a server receives the statistics that the drones have gathered overWi-Fi. However, sophisticated, time-consuming, and bandwidth-intensive advanced structures are a drawback. A Python-based real-time multiple picture identification engine is used to resolve this issue. The software platform for this machine is Jupyter, and it is tested against a dataset made up of 20000 (480 480 pixel) pictures of various damage patterns coming from various sources. Based on Python categorization, manually labelled photos are changed for learning and experience. In the future, this version can be used in combination with various modules to completely avoid life-related risks.