Certain investigations on machine Learning techniques for crack Identification in composite Material images

Abstract

Image visualization finds applications in diverse areas of avionics, newlinearchaeology, medicine, video communication, and electronic games. The non newlinedestructive testing and structural health monitoring are essential for safety and newlinereliability of the systems in aviation. Techniques of screening for anomalies newlinesuch as, crack analysis, screening for damages in metal plates, aircraft newlinetrajectory analysis, aerodynamic heating and aircraft multi-skin inspection newlinebenefit greatly from recent advances in image processing. Occurrence of newlinecrack on the aircraft structure is one among the major problems being faced newlineby aviation. The expansion of cracks in aeroplane constructions is newlinesignificantly influenced by fatigue. Additionally, wear issues and corrosion newlinedamage can cause skin problems and rupture on the structures. Appropriate newlineupkeep and regular predetermined tests can prevent unexpected failure. newlineIn recent decades, composite material, a new alloying ingredient is newlinecommonly used in fracture resistance design of aircraft structures. It is crucial newlineto create real-time and visible monitoring systems for the structural safety. newlineTechniques including Lamb waves, acoustic emission, infrared thermography, newlineinfrared piezoelectric transducers, infrared thermography, infrared acoustic newlineemission, and Digital Image Correlation (DIC) have all been widely used to newlineassess aircraft structural safety. However, their use in large-scale and complex newlinestructural components has been constrained due to the limited sensor size, newlineexpensive and sophisticated measuring equipment, as well as difficult newlineprocessing methods. The evaluation of fatigue damage urgently requires full newlinefield, real-time, and visible Structural Health Monitoring (SHM) approaches newline newline