This paper compares the performance of various popular convolutional neural network (CNN) architectures for image classification on the CIFAR10 dataset. The comparison includes CNN architectures such as Inception V3, Inception-ResNet-v2, ResNetV1, and V2, ResNeXt, MobileNet, and DenseNet, with the addition of two attention mechanisms - Convolutional Block Attention Module (CBAM), and Squeeze and Excitation (SE). CBAM and SE are believed to improve CNNs' performance, especially for complex images with multiple objects and backgrounds. The models are evaluated using loss and accuracy. The main focus of this study is to identify the most effective CNN architecture for image classification on the CIFAR10 dataset with attention mechanisms. The study aims to compare the accuracy of various CNN architectures with and without attention mechanisms and to identify the critical differences between these architectures in terms of their ability to handle complex images. The findings of this study could have implications for developing advanced CNN architectures that can potentially improve the accuracy of computer vision systems in various applications.

Wheat leaf disease prevention and treatment requires a accurate and rapid classification of wheat leaf diseases and their extent. Using healthy wheat, leaf rust, crown and root rot, and wheat loose smut as research objects, this study proposes a deep learning-based technique for classifying the wheat leaf diseases. A collaborative generative adversarial network is used as an image imputation in the proposed methodology, allowing a generator and discriminator network to properly estimate the missing data in the dataset using the residual method. It is used to improve the feature extraction in wheat leaf images. The major contribution of this study is to use a pre-trained deep learning convolutional neural network architecture as a foundation to improve and construct an automated tool for wheat leaf disease image categorization. To classify wheat leaf diseases, a modification to ResNet50 is being suggested. The ′Conv′, ′Batch Normaliz′, and ′Activation Leaky Relu′ layers were added as part of this modification. These layers are inserted into the ResNet50 architecture for accurate feature extraction and discrimination. Extensive tests are carried out to evaluate the proposed model's performance on photos from a large wheat disease classification dataset. The suggested approach outperforms ResNet50, InceptionV3, and DenseNet, according to the experimental findings. The suggested method achieves the greatest identification accuracy of 98.44%. These discoveries might aid in the accurate detection and categorization of wheat leaf diseases.