Performance Comparison of FOD based Edge Detector and Traditional Edge Detectors on Fish Image Edge Detection

Detection of edge in image is a fundamental requirement involved in computer vision and image processing applications. In this paper, the performance of traditional edge detectors is compared with Grunwald-Letnikov(G-L) based Fractional Order Derivative (FOD) based edge detector. The performance is measured for both types of detectors under noise free and noisy conditions on fish images. Image quality assessment (IQA) parameters Mean Square Error (MSE), Peak Signal-to-Noise-Ratio (PSNR), Structural Similarity Index (SSIM) and Feature Similarity Index (FSIM) are used for quantitative comparison of the edge detection. From the experimental results, it is observed that FOD based edge detector shows better results than the traditional edge detectors under noisy conditions either in terms of quality or quantity.