2022/01/23 by Zhao Xiangyu, Xiangyu, Zhao · 4 citations
Computer Science · Mathematics · Medicine · Neuroscience · #Advanced Neural Network Applications #Artificial intelligence #Benchmark (surveying) #Block (permutation group theory) #Brain Tumor Detection and Classification #COVID-19 diagnosis using AI #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Computer vision #Contextual image classification #Convolutional neural network #Discrete wavelet transform #Domain (mathematical analysis) #FOS: Computer and information sciences #Feature (linguistics) #Feature extraction #Frequency domain #Image (mathematics) #Mathematics #Noise (video) #Pattern recognition (psychology) #Wavelet #Wavelet transform #cs.CV
paper · pdf · doi:10.48550/arxiv.2201.09271
published in arXiv (Cornell University) (Cornell University) · 17 pages, 5 figures
arxiv created 2022/01/23 · openalex publication_date 2022/01/23 · arxiv updated 2022/01/25 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01
The feature learning methods based on convolutional neural network (CNN) have successfully produced tremendous achievements in image classification tasks. However, the inherent noise and some other factors may weaken the effectiveness of the convolutional feature statistics. In this paper, we investigate Discrete Wavelet Transform (DWT) in the frequency domain and design a new Wavelet-Attention (WA) block to only implement attention in the high-frequency domain. Based on this, we propose a Wavelet-Attention convolutional neural network (WA-CNN) for image classification. Specifically, WA-CNN decomposes the feature maps into low-frequency and high-frequency components for storing the structures of the basic objects, as well as the detailed information and noise, respectively. Then, the WA block is leveraged to capture the detailed information in the high-frequency domain with different attention factors but reserves the basic object structures in the low-frequency domain. Experimental results on CIFAR-10 and CIFAR-100 datasets show that our proposed WA-CNN achieves significant improvements in classification accuracy compared to other related networks. Specifically, based on MobileNetV2 backbones, WA-CNN achieves 1.26% Top-1 accuracy improvement on the CIFAR-10 benchmark and 1.54% Top-1 accuracy improvement on the CIFAR-100 benchmark.