International Journal of Computer Science Issues

More than a traditional journal...

Edge detection is still a challenging problem and researchers are focusing to investigate this problem using different techniques. Edge detection is an important preprocessing step in most of image processing applications. The application ranges from realtime video surveillance, traffic surveillance to medical imaging applications. Current state-of-the-art methods for edge detection are filter based and do not incorporate spatial-temporal information among the consecutive frames. We propose a robust approach for edge detection by exploiting spatial temporal information that possess an important cue to robust edge detection. This is achieved by extracting hybrid features in terms of pairwise local binary pattern (P-LBP) and scale invariant feature transform (SIFT). These features are used to train an MLP neural network during the training stage, and the edges are inferred from the test videos during the testing stage. The experimental evaluation is conducted on a benchmark dataset commonly used for edge detection.