DWT-CompCNN：面向高吞吐量 JPEG 2000 压缩文档的深度图像分类网络
DWT-CompCNN: Deep Image Classification Network for High Throughput JPEG 2000 Compressed Documents
AI总结：
本文提出 DWT-CompCNN，通过五层卷积网络直接学习 HTJ2K 压缩文档的小波系数，在 Tobacco-3482 和 RVL-CDIP 上实现更高效且更准确的压缩域分类。
摘要：
对于任何涉及文档图像的数字应用（例如检索），文档图像分类都成为一个必不可少的阶段。传统上，为实现该目的，文档的完整版本，即未压缩的文档图像，构成输入数据集；由于容纳文档完整版本需要巨大的数据量，这带来了威胁。因此，如果能够利用文档的压缩表示直接完成同一分类任务（仅进行部分解压），从而使整个过程在计算上更加高效，将是一种新颖的方案。在本研究工作中，提出了一种新颖的深度学习模型 DWT CompCNN，用于对采用高吞吐量 JPEG 2000（HTJ2K）算法压缩的文档进行分类。所提出的 DWT-CompCNN 包含五个卷积层，随着层数递增，滤波器数量依次为 16、32、64、128 和 256，以增强从压缩图像中提取的小波系数的学习。实验在 Tobacco-3482 和 RVL-CDIP 两个基准数据集上进行，结果表明所提出的模型具有时间和空间效率，并且在压缩域中取得了更好的分类准确率。
英文摘要：
For any digital application with document images such as retrieval, the classification of document images becomes an essential stage. Conventionally for the purpose, the full versions of the documents, that is the uncompressed document images make the input dataset, which poses a threat due to the big volume required to accommodate the full versions of the documents. Therefore, it would be novel, if the same classification task could be accomplished directly (with some partial decompression) with the compressed representation of documents in order to make the whole process computationally more efficient. In this research work, a novel deep learning model, DWT CompCNN is proposed for classification of documents that are compressed using High Throughput JPEG 2000 (HTJ2K) algorithm. The proposed DWT-CompCNN comprises of five convolutional layers with filter sizes of 16, 32, 64, 128, and 256 consecutively for each increasing layer to improve learning from the wavelet coefficients extracted from the compressed images. Experiments are performed on two benchmark datasets- Tobacco-3482 and RVL-CDIP, which demonstrate that the proposed model is time and space efficient, and also achieves a better classification accuracy in compressed domain.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
