Endoscopy image enhancement method by generalized imaging defect models based adversarial training
Endoscopy image enhancement method by generalized imaging defect models based adversarial training
复制标题
基于广义成像缺陷模型的对抗训练内窥镜图像增强方法
DOI：
10.1088/1361-6560/ac6724
复制
发表时间：
2022-04
影响因子：
3.5
通讯作者：
Jian Yang
中科院分区：
文献类型：
--
作者：
Wenjie Li;Jingfan Fan;Yating Li;Pengcheng Hao;Yucong Lin;Tianyu Fu;Danni Ai;Hong Song;Jian Yang
目标。烟雾、光线不均匀、颜色偏差是内镜手术中常见的问题，这些问题增加了手术的风险，甚至导致手术失败。的方法。在这项研究中，我们提出了一种新的物理模型驱动的半监督学习框架，用于高质量的逐像素内窥镜图像增强，该框架可用于除烟，光线调整和颜色校正。为了提高生成图像的真实性，从而提高网络性能，我们将特定的物理成像缺陷模型与CycleGAN框架集成在一起。不需要成对的真实数据。此外，我们提出了一个迁移学习框架，以解决内窥镜增强任务中的数据稀缺性，提高网络性能。主要的结果。定性和定量研究表明，所提出的网络优于最先进的图像增强方法。特别是，该方法在除烟任务中的结构相似度从0.7925提高到0.8648，彩色图像的特征相似度从0.8917提高到0.9283，四元数结构相似度从0.8097提高到0.8800。实验结果表明，该迁移学习方法在目标任务的小数据集上具有较好的训练效果。的意义。在内镜图像上的实验结果证明了该网络在除烟、调光、色彩校正等方面的有效性，具有良好的临床应用价值。
Objective. Smoke, uneven lighting, and color deviation are common issues in endoscopic surgery, which have increased the risk of surgery and even lead to failure. Approach. In this study, we present a new physics model driven semi-supervised learning framework for high-quality pixel-wise endoscopic image enhancement, which is generalizable for smoke removal, light adjustment, and color correction. To improve the authenticity of the generated images, and thereby improve the network performance, we integrated specific physical imaging defect models with the CycleGAN framework. No ground-truth data in pairs are required. In addition, we propose a transfer learning framework to address the data scarcity in several endoscope enhancement tasks and improve the network performance. Main results. Qualitative and quantitative studies reveal that the proposed network outperforms the state-of-the-art image enhancement methods. In particular, the proposed method performs much better than the original CycleGAN, for example, the structural similarity improved from 0.7925 to 0.8648, feature similarity for color images from 0.8917 to 0.9283, and quaternion structural similarity from 0.8097 to 0.8800 in the smoke removal task. Experimental results of the proposed transfer learning method also reveal its superior performance when trained with small datasets of target tasks. Significance. Experimental results on endoscopic images prove the effectiveness of the proposed network in smoke removal, light adjustment, and color correction, showing excellent clinical usefulness.
登录
查看更多内容
影响因子：
2.4
作者：
Ye Xin;Zhenhong Jia;Jie Yang;N. Kasabov
通讯作者：
Ye Xin;Zhenhong Jia;Jie Yang;N. Kasabov
一种基于颜色衰减先验的快速单图像去雾算法
影响因子：
10.6
作者：
Zhu, Qingsong;Mai, Jiaming;Shao, Ling
通讯作者：
Shao, Ling
影响因子：
1.3
作者：
Christopher C. Yang;S. Kwok
通讯作者：
Christopher C. Yang;S. Kwok
在内窥镜视觉中实现更好的手术器械分割：多角度特征聚合和轮廓监督
影响因子：
5.2
作者：
Fangbo Qin;Shan Lin;Yangming Li;R;all A. Bly;Kris S. Moe;Blake Hannaford
通讯作者：
Blake Hannaford
内窥镜图像增强，并抑制噪声。
影响因子：
2.1
作者：
Xia W;Chen ECS;Peters T
通讯作者：
Peters T