A Novel Approach to Edge Detection and Performance Measurebased on the Theory of “Range” and “Bowley’s Measure ofSkewness” in a Noisy Environment

Abstract

We proposed a new edge detection algorithm in a salt and pepper noisy environment without any pre-processing. The algorithm is a simple, unique mathematical expression. It is the product of three terms: median, range and some edge image enhancement constant. The algorithm assumes that the edge exists in an image at non-zero skewness. We also proposed a performance measure technique by using “Bowley's measure of Skewness”. The coefficient of Skewness due to Bowleylies between–1 and +1. We measured the absolute value of the coefficient of Skewness. The effectiveness of the above edge detection algorithm as well as performance measure technique were experimented and compared with other existing techniques. In all the experiments, our method demonstrates comparatively better results in the Salt and Pepper noisy environment as well as in the noiseless image.