Table 2 The quality metrics for leaf segmentation.
From: A CNN-based model to count the leaves of rosette plants (LC-Net)
Sl. | Parameters | Remarks |
|---|---|---|
1 | AC | AC is the ratio of the total of correctly identified pixels and the total number of Pixels. The higher accuracy value represents better results42. |
2 | IoU | The value of IoU ranges between 0 and 1. It indicates the amount of overlapping is present between predicted image and the ground truth. In the existing literature, if the IoU for the outputs of the model is more than 0.5, then it is considered that the model is predicting well43. |
3 | DI | Basically DI indicates the similarity of predicted image with the ground truth. It is calculated by matching the overlapped region of the predicted image by the technique and the ground truth image44. A higher DI value better performance of the participating technique. |
