Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2021 Feb 8;40(5):1450–1460. doi: 10.1109/TMI.2021.3057884

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/

PMC Copyright notice

Fig. 2. — Overview of the semi-synthetic generation method. We illustrate both the concept of source image and the process to generate semi-synthetic images. For an image to be considered as source image, it must fulfill two conditions. First, its labeling must be available or easier than a manual annotation of the tool in a real endoscopic image. Second, it must be a close approximation of a real clinical image so that when complementary source images are blended, the result is close to a real endoscopic image. The automatic segmentation of source images is hue-based for (detailed in section IV-C) and trivial for . We assume that the correct way to blend complementary source images to form a real clinical image is unknown. In this figure, three possible ways to do it are illustrated. A green arrow is shown to highlight a part of the semi-synthetically generated image, the border of the tool, and the tooltip itself, whose appearance differs depending on the choice of blending function (best viewed in the electronic version of the manuscript).

Inline graphic — Overview of the semi-synthetic generation method. We illustrate both the concept of source image and the process to generate semi-synthetic images. For an image to be considered as source image, it must fulfill two conditions. First, its labeling must be available or easier than a manual annotation of the tool in a real endoscopic image. Second, it must be a close approximation of a real clinical image so that when complementary source images are blended, the result is close to a real endoscopic image. The automatic segmentation of source images is hue-based for (detailed in section IV-C) and trivial for . We assume that the correct way to blend complementary source images to form a real clinical image is unknown. In this figure, three possible ways to do it are illustrated. A green arrow is shown to highlight a part of the semi-synthetically generated image, the border of the tool, and the tooltip itself, whose appearance differs depending on the choice of blending function (best viewed in the electronic version of the manuscript).