Structural Crack Detection Using High Boost Filtering Based Enhanced Average Thresholding

Structural Health Monitoring (SHM) is a procedure which detects and monitors the destructive part of various infrastructures like buildings, water tanks, bridges, etc. There are several Non Destructive Evaluation (NDE) methods available for SHM, however, there are limitations like non-safety of the inspection team, high labor cost and low accuracy of the subsurface defect detection. These limitations can be overcome by utilizing camera based Wireless Sensor Network (WSN) for continuous monitoring and detection of defects in the buildings. Hence, this work aims at formulating an efficient image processing methodology to identify the defects in infrastructures. The developed defect detection algorithm contains pre-processing, segmentation and post processing phases. An enhanced average thresholding (EAT) strategy along with high boost filtering is proposed to increase the accuracy of crack detection in structures. Simulation results depict that the proposed algorithm has better similarity measures in recognizing the cracks on infrastructures.