Skip to main content
Fig. 9 | Plant Methods

Fig. 9

From: SpikeSegNet-a deep learning approach utilizing encoder-decoder network with hourglass for spike segmentation and counting in wheat plant from visual imaging

Fig. 9Fig. 9

Segmentation performance with different levels of illuminated images: First column shows the illuminated images at different gamma values; Second column represents the output images after applying SpikeSegNet approach; For visually analyzing the pixel count error, output of SpikeSegNet at different level of illumination is superimposed on the ground truth segmented mask image (manually prepared).The colored pixel (pink and green) represents the wrongly classified pixels where pink indicates that actual spike pixels are not identified and the green pixels indicate that non-spike pixels are misclassified as spike pixels. The circles on the figure represent the spike pixels which are not detected

Back to article page

Plant Methods

ISSN: 1746-4811