Loading [a11y]/accessibility-menu.js
Endoscopic Image Classification using Block-based Color Feature Descriptors | IEEE Conference Publication | IEEE Xplore

Endoscopic Image Classification using Block-based Color Feature Descriptors


Abstract:

Compared to biopsy, endoscopy is a less invasive way to screen abnormalities in gastrointestinal (GI) track. In chromoendoscopy, one of the advancements to endoscopy, sus...Show More

Abstract:

Compared to biopsy, endoscopy is a less invasive way to screen abnormalities in gastrointestinal (GI) track. In chromoendoscopy, one of the advancements to endoscopy, suspicious regions are made more visible visually by spraying dyes over the mucosal surface. However, identifying and categorizing abnormal regions are challenging tasks. Nowadays, automated systems or computer-aided diagnosis (CAD) systems has been developing. Such systems are helpful for gastroenterologists by pointing out abnormal frame. Extraction of features in endoscopic images plays an important role in every CAD system. In this paper, new color feature descriptors for endoscopic images have been proposed. These descriptors are constituted from Block Difference of Inverse Probabilities (BDIP) and Block Variation of Local Correlation coefficients (BVLC) features extracted on color space. Experimental results show that the method is prospective to investigate more.
Date of Conference: 20-22 December 2022
Date Added to IEEE Xplore: 18 January 2023
ISBN Information:
Print on Demand(PoD) ISSN: 2162-786X
Conference Location: Ho Chi Minh City, Vietnam

References

References is not available for this document.