Figure 4

Channel attention guidance module CAGM. Obtaining \({\mathrm{M}}_{\mathrm{s}}\) and \({\mathrm{A}}_{\mathrm{S}}\) by two types of pooling and one convolution after classification.

Channel attention guidance module CAGM. Obtaining \({\mathrm{M}}_{\mathrm{s}}\) and \({\mathrm{A}}_{\mathrm{S}}\) by two types of pooling and one convolution after classification.