Qualitative Comparison of Image Stitching Algorithms for Multi-Camera Systems in Laparoscopy

Abstract

Multi-camera systems were recently introduced into laparoscopy to increase the narrow field of view of the surgeon. The video streams are stitched together to create a panorama that is easier for the surgeon to comprehend. Multi-camera prototypes for laparoscopy use quite basic algorithms and have only been evaluated on simple laparoscopic scenarios. The more recent state-of-the-art algorithms, mainly designed for the smartphone industry, have not yet been evaluated in laparoscopic conditions. We developed a simulated environment to generate a dataset of multi-view images displaying a wide range of laparoscopic situations, which is adaptable to any multi-camera system. We evaluated classical and state-of-the-art image stitching techniques used in non-medical applications on this dataset, including one unsupervised deep learning approach. We show that classical techniques that use global homography fail to provide a clinically satisfactory rendering and that even the most recent techniques, despite providing high quality panorama images in non-medical situations, may suffer from poor alignment or severe distortions in simulated laparoscopic scenarios. We highlight the main advantages and flaws of each algorithm within a laparoscopic context, identify the main remaining challenges that are specific to laparoscopy, and propose methods to improve these approaches. We provide public access to the simulated environment and dataset.

Rapid digital pathology of H&E-stained fresh human brain specimens as an alternative to frozen biopsy

“…In the domain of digital surgical pathology, our mNLOG platform, for the first time to the best of our knowledge, provides a truly WSI-competing whole specimen superficial imaging (WSSI) digital ITA solution enabling multicolor imaging of a 1 cm 2 area in <120 s with a total of 86 G bits or 3.6 Gigapixels (24-bit) preserving a submicron digital resolution with no requirement of post-acquisition data/ image processing. It is noted that despite substantial contributions being made towards image/video or panoramic stitching techniques [38][39][40][41][42][43] , feature-based sophisticated algorithms are often not suitable for a large-field high-pixel-rate dynamic microscopy in the context of parallel implementation, distortion compensation, immunity to high-frequency noise, and especially to assist with half-a-second computational complexity for real-time stitching of ultra-high resolution (such as >800 M bit) imaging tiles. To fit our specific need, a Compute Unified Device Architecture (CUDA)-accelerated rapid artifact-compensated 2D largefield mosaic-stitching (rac2D-LMS) approach is streamlined to our large-FOV (≥1 mm 2 ) high-NFOM (>1) multi-channel nonlinear optical laser-scanning and acquisition system.…”

Designing a Virtual Campus Tour using Image Stitching Techniques to Provide Information on College Entrance Test

“…The image stitching method can be classified into two general methods: direct and feature-based. The direct method compares all the pixel intensities of the image with each other, while the feature-based method aims to determine the relationship between images through various features extracted from the processed image [20]- [23].…”

Optimized Seam-Driven Image Stitching Method Based on Scene Depth Information

“…Image stitching [1] refers to the use of a set of images of the same scene taken from different perspectives to create a single fused image with a wider field of view. It is widely used in multimedia content generation, image analysis/understanding, industrial inspection, and other fields (such as panoramic imaging [2], aerial image generation [3], medical synthetic image generation [4,5], virtual reality [6], remote visual inspection [7], and so on).…”