Civil aircraft surface defects detection based on histogram of oriented gradient

Visual inspection is the primary means of non-destructive testing of civil aircraft skin surfaces but it may lead to the inability to detect due to human factors and bad weather. We propose an automatic recognition method for aircraft surface defects based on histogram of oriented gradient (HOG) and analyze three typical types of failures: scratches, corrosion, and paint detachment. The experimental results show that this method has good detection accuracy and has strong application potential.