Automtic leaf surface region segmentation from crop image
1. College of Mathematics and Informatics, South China Agricultural University, Guangzhou, Guangdong 510642, China; 2. Department of Biosystems and Agricultural Engineering, Oklahoma State University, Stillwater, Oklahoma 74075, USA
Abstract:Leaf surface region segmentation from crop images is a crucial step for obtaining crop information. The leaf surface region segmentation (LSRS)from crop images was proposed. The color segmentation was used to obtain green region in crop images, and the smoothness segmentation was applied to identify smooth parts in green regions. Using the erosion method and the pixel number of each region, the noises were eliminated by LSRS to separate big noise and leaf regions into different parts by connectivity scheme. The dilation method was finally used to obtain whole leaf surface region. The experiments were conducted with several crop leaf images to verify the proposed method.The results show that compared to existing methods, the proposed system can automatically segment leaf surface region from complicated crop images.