Zhang Honghao, Chen Bi, Gao Xianwei, Yao Xiang, Hou Linyu
Beijing Electronic Science and Technology Institute, Beijing 100070, China.
Sensors (Basel). 2024 Sep 14;24(18):5976. doi: 10.3390/s24185976.
Compressive sensing (CS) is a notable technique in signal processing, especially in multimedia, as it allows for simultaneous signal acquisition and dimensionality reduction. Recent advancements in deep learning (DL) have led to the creation of deep unfolding architectures, which overcome the inefficiency and subpar quality of traditional CS reconstruction methods. In this paper, we introduce a novel CS image reconstruction algorithm that leverages the strengths of the fast iterative shrinkage-thresholding algorithm (FISTA) and modern Transformer networks. To enhance computational efficiency, we employ a block-based sampling approach in the sampling module. By mapping FISTA's iterative process onto neural networks in the reconstruction module, we address the hyperparameter challenges of traditional algorithms, thereby improving reconstruction efficiency. Moreover, the robust feature extraction capabilities of Transformer networks significantly enhance image reconstruction quality. Experimental results show that the FusionOpt-Net model surpasses other advanced methods on various public benchmark datasets.
压缩感知（CS）是信号处理领域，尤其是多媒体领域的一项重要技术，因为它能够同时进行信号采集和降维。深度学习（DL）的最新进展催生了深度展开架构，克服了传统CS重建方法的低效率和质量欠佳问题。在本文中，我们介绍了一种新颖的CS图像重建算法，该算法利用了快速迭代收缩阈值算法（FISTA）和现代Transformer网络的优势。为提高计算效率，我们在采样模块中采用基于块的采样方法。通过在重建模块中将FISTA的迭代过程映射到神经网络上，我们解决了传统算法的超参数挑战，从而提高了重建效率。此外，Transformer网络强大的特征提取能力显著提升了图像重建质量。实验结果表明，FusionOpt-Net模型在各种公共基准数据集上优于其他先进方法。
