Computer Aided Visual Inspection of Aircraft Surfaces

Non Destructive Inspections (NDI) plays a vital role in aircraft industry as it determines thestructural integrity of aircraft surface and material characterization. The existing NDI methods aretime consuming, we propose a new NDI approach using Digital Image Processing that has thepotential to substantially decrease the inspection time. Automatic Marking of cracks have beenachieved through application of Thresholding, Gabor Filter and Non Subsampled Contourlettransform. For a novel method of NDI, the aircraft imagery is analyzed by three methods i.eNeural Networks, Contourlet Transform (CT) and Discrete Cosine Transform (DCT). With the helpof Contourlet Transform the two dimensional (2-D) spectrum is divided into fine slices, usingiterated directional filterbanks. Next, directional energy components for each block of thedecomposed subband outputs are computed. These energy values are used to distinguishbetween the crack and scratch images using the Dot Product classifier. In next approach, the aircraftimagery is decomposed into high and low frequency components using DCT and the firstorder moment is determined to form feature vectors.A correlation based approach is then usedfor distinction between crack and scratch surfaces. A comparative examination between the twotechniques on a database of crack and scratch images revealed that texture analysis using thecombined transform based approach gave the best results by giving an accuracy of 96.6% for theidentification of crack surfaces and 98.3% for scratch surfaces.