Aircraft Authentication for Image Processing

Aircraft recognition remains a challenging problem despite a great deal of effort to automate the recognition process. Most of the aircraft recognition methods involve extraction of overall shapes of aircraft at first and then represent the aircraft based on the extracted shape with different features for recognition. The major disadvantage of these methods is that they require a huge amount of analysis on shape extraction, which is too unrealistic for targets in satellite images. we propose a new hybrid recognition method that combines contour wavelet features and correlation on shape analysis. Initially, direction estimation is used to align the aircraft in same direction. Then, reconstruction based similarity measure is introduced to transform the type recognition problem into a reconstruction problem. Filtration is done for Gaussian noise removal purpose and wavelet transform is used as a feature extraction method. Fuzzy clustering is done in final to carry out segmentation on aircraft image. The performance of these used algorithms will be differentiated though precision and recall rate metrics. The experimental results illuminate the accuracy and efficiency of the proposed system to recognize aircraft in satellite remote sensing images.