Skip to main content
. 2024 Aug 20;6:0236. doi: 10.34133/plantphenomics.0236

Fig. 3.

Fig. 3.

The overall architecture of CSNet, including Backbone, convolutional block attention module (CBAM), multiscale perception module (MPM), and counting module (CM). (A) The detailed structure of CBAM is used to improve the attention to the wheat region. (B) The flow of the MPM utilizes the wheat ear features at multiple scales to improve the generalization performance of the network. (C) The detailed operation of the mixer layer in the MPM achieves sensing wheat ears without location information by fusing global features.