Edge Detection for Hardwood Seedlings Leaves Based on Intuitionistic Fuzzy Set

Abstract

The task of reliably segmenting leaves is significant for plants recognition and reconstruction. Leaves image segmentation, registration and identification are based on edge detection. In this paper, a novel method to detect hardwood leaves edges is proposed, which clusters, thresholds, and then detects edges of hardwood seedlings leaves using intuitionistic fuzzy set (IFS) theory. Clustering segments image into several clusters and histogram threshold eliminates unwanted clusters that are not related to leaves region. Finally, image edge is detected, where a clear boundary is obtained. Proposed method performs better than classical edge detection methods. Experiments for kinds of hardwood seedlings are carried out and the results indicate that the proposed method is effectiveness to detect the leaves edges.