Dilated Residual Learning With Skip Connections for Real-Time Denoising of Laser Speckle Imaging of Blood Flow in a Log-Transformed Domain
Dilated Residual Learning With Skip Connections for Real-Time Denoising of Laser Speckle Imaging of Blood Flow in a Log-Transformed Domain
复制标题
使用跳跃连接进行扩张残差学习，用于对数变换域中血流激光散斑成像的实时去噪
DOI：
10.1109/tmi.2019.2953626
复制
发表时间：
2020-05-01
影响因子：
10.6
通讯作者：
Li, Pengcheng
中科院分区：
文献类型：
--
作者：
Chen, Weimin;Lu, Jinling;Li, Pengcheng
激光散斑对比成像（LSCI）是一种宽视场、非接触式的血流成像技术。虽然提出了基于块匹配和三维变换域协同滤波（BM3D）的去噪方法，显著提高了其信噪比，但由于处理时间长，难以实现实时去噪。此外，考虑到存在明显的噪声和伪影，使用少量原始散斑图像仍然难以获得可接受的信噪比水平。一种前馈去噪卷积神经网络（DnCNN）在自然图像去噪方面达到了最先进的性能，并得到了GPU的有效加速。然而，由于噪声分布的不均匀性，该算法在学习LSCI原始散斑对比度图像时表现不佳。因此，我们提出在对数变换域中训练DnCNN来提高LSCI的训练精度，其峰值信噪比（PSNR）提高了5.13 dB。为了减少推理时间和提高去噪性能，我们进一步提出了一种带跳跃连接的扩展深度残差学习网络（DRSNet）。基于5张原始斑点图像的DRSNet图像质量评价优于基于20张原始斑点图像的空间平均去噪。DRSNet在NVIDIA 1070 GPU上以$486\times648$像素对血流图像进行降噪需要35 ms（即每秒28帧），大约比DnCNN快2.5倍。在测试集中，DRSNet的PSNR也比DnCNN提高了0.15 dB。该网络在实时血流监测方面具有良好的应用前景。
Laser speckle contrast imaging (LSCI) is a wide-field and noncontact imaging technology for mapping blood flow. Although the denoising method based on block-matching and three-dimensional transform-domain collaborative filtering (BM3D) was proposed to improve its signal-to-noise ratio (SNR) significantly, the processing time makes it difficult to realize real-time denoising. Furthermore, it is still difficult to obtain an acceptable level of SNR with a few raw speckle images given the presence of significant noise and artifacts. A feed-forward denoising convolutional neural network (DnCNN) achieves state-of-the-art performance in denoising nature images and is efficiently accelerated by GPU. However, it performs poorly in learning with original speckle contrast images of LSCI owing to the inhomogeneous noise distribution. Therefore, we propose training DnCNN for LSCI in a log-transformed domain to improve training accuracy and it achieves an improvement of 5.13 dB in the peak signal-to-noise ratio (PSNR). To decrease the inference time and improve denoising performance, we further propose a dilated deep residual learning network with skip connections (DRSNet). The image-quality evaluations of DRSNet with five raw speckle images outperform that of spatially average denoising with 20 raw speckle images. DRSNet takes 35 ms (i.e., 28 frames per second) for denoising a blood flow image with $486\times648$ pixels on an NVIDIA 1070 GPU, which is approximately 2.5 times faster than DnCNN. In the test sets, DRSNet also improves 0.15 dB in the PSNR than that of DnCNN. The proposed network shows good potential in real-time monitoring of blood flow for biomedical applications.