Multiscale deep desmoking for laparoscopic surgery

摘要

In minimally invasive surgery, smoke generated by such as electrocautery and laser ablation deteriorates imagequality severely. This creates discomfortable view for the surgeon which may increase surgical risk and degradethe performance of computer assisted surgery algorithms such as segmentation, reconstruction, tracking, etc.Therefore, real-time smoke removal is required to keep a clear field of view. In this paper, we propose a real-timesmoke removal approach based on Convolutional Neural Network (CNN). An encoder-decoder architecture withLaplacian image pyramid decomposition input strategy is proposed. This is an end-to-end network which takesthe smoke image and its Laplacian image pyramid decomposition as inputs, and outputs a smoke free imagedirectly without relying on any physical models or estimation of intermediate parameters. This design can befurther embedded to deep learning based follow-up image guided surgery processes such as segmentation andtracking tasks easily. A dataset with synthetic smoke images generated from Blender and Adobe Photoshop isemployed for training the network. The result is evaluated quantitatively on synthetic images and qualitativelyon a laparoscopic dataset degraded with real smoke. Our proposed method can eliminate smoke effectively whilepreserving the original colors and reaches 26 fps for a video of size 512 on 512 on our training machine. Theobtained results not only demonstrate the efficiency and effectiveness of the proposed CNN structure, but alsoprove the potency of training the network on synthetic dataset.