Identifying flaws, such as fractures, corrosion, and voids in industrial structures, presents substantial obstacles. These challenges arise from difficult-to-reach places, the presence of vast facilities, and the noise found in infrared (IR) images. In response to this issue, the present project proposes an enhanced defect detection system that amalgamates wavelet-based image processing with artificial intelligence. The main goal of this work is to perform precise and automatic identification, and localization, of infrared thermal anomalies in UAV-acquired infrared images and thereby promote the effectiveness and consistency of inspections. The system developed suggests the use of UAVs to capture high resolution thermal images, wavelet decomposition for the separation of images into multi-resolution components, and efficient extraction of subtle background noise from defect features. The features are then inspected with machine learning techniques in the MATLAB platform. Experimental assessments carried out with a thoroughly collected set of UAV-recorded infrared images show superior detection capabilities compared with existing techniques. The originality of the research works in combining UAV-based infrared imaging, wavelet transform for the decomposition of images into multi-resolution parts, and AI-powered defect recognition to construct a completely automated, scalable, and non-destructive inspection system.
S et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: