Plant Leaves Extraction Method Under Complex Background Based on Closed-Form Matting Algorithm

摘要

The extraction of plant leaves is a technique to separate the captured leaves from the background. In natural scene, the captured picture of plant leaves always contain of complex background and it's hard to extract the leaves. Considering that the drawback of extraction technique based on image segmentation algorithm, this paper introduces Closed-form algorithm into this technique and proposes an improved extraction method which can extract plant leaves from complex background. Firstly, this algorithm requires users to add certain constraints for plant leaves image with complex background. Secondly, the algorithm identifies foreground and background via estimating diaphaneity α of unknown sample. Through these two steps, the plant leaves within complex background could be extracted efficiently. Experimental comparison results show that the proposed method has a better effect.