Performance analysis of various leaf boundary edge detection algorithms

Abstract

Plants play an important role in both human life and other lives that exist on the earth. Due to environmental deterioration and lack of awareness, many rare plant species are at the margins of extinction. Despite the great advances made in botany, there are many plants yet to be discovered, classified, and utilized; unknown plants are treasures waiting to be found. Leaf classification and recognition for plant identification plays a vital role in all these endeavors. Detection of edges is considered to be one of the vital steps of preprocessing during leaf identification and recognition system. The current need is to have an edge detector which is both fast and efficient in identifying edges of a leaf image. This paper introduced a wavelet base edge detection technique, which used a dyadic wavelet transformation and p-tile thresholding method to find the edges. The proposed method was compared with conventional edge detectors like canny, sobel, prewitt, Roberts and Log. From the experimental results it was found that the proposed method is advantageous and can detect edges.