Figure 3
A flowchart illustrating the transfer learning process of DenseNet121 for tomato disease identification. The flowchart begins with a source dataset labeled ImageNet, which leads to the creation of a source model. This source model undergoes a transfer process to form a target model. The target model is then used for modeling with a target dataset labeled Primary and Plantvillage. The modeled target dataset is subsequently tested and deployed. The lower part of the diagram shows the structure of DenseNet121, starting with an input layer, followed by a series of convolution blocks, dense blocks, pooling layers, and ending with a classification layer that includes global average pooling, a fully connected layer, and a softmax layer.DenseNet121 for tomato disease classification in detail