Edge detection of plant roots image via Gabor wavelet theory

Abstract

The pictures extracted from the plant roots often have quite big noises and often affected greatly by the light intensity, neither could we achieve satisfactory results by using the traditional image edge detection method to detect the edge information. To this issue, a method for edge detection and its application in edge image detection for plant roots via gabor wavelet theory is proposed, which is able to filter from different directions and scales to determine the edge of the texture under perfect frequency, and proposes methods that using Gabor wavelet theory to detect the image edges of the plant root. Experimental results show that compared with other traditional edge detection methods, the proposed approach is effective in reduction of color space dimensionality and robust to noise.