Skip to main content
Fig. 8 | Plant Methods

Fig. 8

From: Panicle-SEG: a robust image segmentation method for rice panicles in the field based on deep learning and superpixel optimization

Fig. 8

Patch generation approaches based on SLIC. The black point represents the weighted center of the current SLIC superpixel region (irregularly shaped red regions). The region centered on the black point and enclosed by a white box (32 × 32 pixels) is the training patch on the corresponding SLIC superpixel region. (1), (2), and (3) are the zoomed-in versions of the white box training patches, and the percentage is the ratio between the foreground panicle regions to the corresponding SLIC superpixel region in each patch

Back to article page

Plant Methods

ISSN: 1746-4811