An automatic leaf analysis system

Abstract

In this paper, we proposed an automatic vision-based leaf analysis system. The proposed system is composed of two units: image acquisition unit and image analysis unit. To evaluate the proposed scheme, several images containing different species of leaves are captured by using low-cost webcam. Experimental results show that our proposed system can detect leaf regions well in images. In addition, the accuracy rate of the proposed system for measuring leaf area is more than 99%. In addition, the average execution time for each image is 143.8 msecs. These results demonstrate that our proposed system can not only efficiently locate and measure leaf areas but also count leaves in real-world noisy images.