The paper proposes an A-ResNet model to improve ResNet. The residual attention module with shortcut connection is introduced to enhance the focus on the target object; the dropout layer is introduced to prevent the overfitting phenomenon and improve the recognition accuracy; the network architecture is adjusted to accelerate the training convergence speed and improve the recognition accuracy. The experimental results show that the A-ResNet model achieves a top-1 accuracy improvement of about 2% compared with the traditional ResNet network. Image recognition is one of the core technologies of computer vision, but its application in the field of tea is relatively small, and tea recognition still relies on sensory review methods. A total of 1,713 images of eight common green teas were collected, and the modeling effects of different network depths and different optimization algorithms were explored from the perspectives of predictive ability, convergence speed, model size, and recognition equilibrium of recognition models.