Figure 1From: Field phenotyping of grapevine growth using dense stereo reconstructionWorkflow of the proposed image-based phenotyping method. Stereo image pairs (A, B) are captured from a moving platform in a vineyard. From these images the method computes a dense depth map (C), a color classification based on the the green and blue color channels (D), and an edge detector (E). These features are used to segment the image domain to ‘leaf’, ‘stem’ and ‘background’ (F). The image segmentation allows for objective computation of phenotypic indicators like the visible leaf area and fruit-to-leaf ratio.Back to article page