Fig. 2From: SpikeSegNet-a deep learning approach utilizing encoder-decoder network with hourglass for spike segmentation and counting in wheat plant from visual imagingPatch generation: a visual images (of size 1656 * 1356) as well as d ground truth mask images (of size 1656 * 1356) are divided into 100 pixels overlapping patches [b, c] of size 256 * 256Back to article page