Noise removal using statistical operators for efficient leaf identification

Abstract

Plant identification based on leaf shape is becoming a popular trend, since each leaf carries substantial information that can be used to identify plant. This is difficult because the features of a leaf shape can be influenced by other leaves that have similar features but different categories. This paper presents the most popular statistical operators: mean filtering technique (MFT), median filtering technique (MDFT), Wiener filtering technique (WFT), rank order filtering technique (ROFT) and adaptive two-pass rank order filtering technique (ATRFT) for enhancing preprocessing stage. The performance of these techniques was evaluated using mean square error (MSE) and peak signal to noise ratio (PSNR). Ten features were extracted from the pre-processed leaf images and identification performance was also evaluated using precision and recall. It is found that WFT is the best filtering technique and gives the best identification accuracy of 95.1%.