DocumentCode :
3684056
Title :
Preprocessing with image denoising and histogram equalization for endoscopy image analysis using texture analysis
Author :
Tomoyuki Hiroyasu;Katsutoshi Hayashinuma;Hiroshi Ichikawa;Nobuaki Yagi
Author_Institution :
Faculty of Life and Medical Science, Doshisha University, Kyoto, Japan
fYear :
2015
Firstpage :
789
Lastpage :
792
Abstract :
A preprocessing method for endoscopy image analysis using texture analysis is proposed. In a previous study, we proposed a feature value that combines a co-occurrence matrix and a run-length matrix to analyze the extent of early gastric cancer from images taken with narrow-band imaging endoscopy. However, the obtained feature value does not identify lesion zones correctly due to the influence of noise and halation. Therefore, we propose a new preprocessing method with a non-local means filter for de-noising and contrast limited adaptive histogram equalization. We have confirmed that the pattern of gastric mucosa in images can be improved by the proposed method. Furthermore, the lesion zone is shown more correctly by the obtained color map.
Keywords :
"Endoscopes","Image color analysis","Lesions","Noise reduction","Histograms","Cancer","Image analysis"
Publisher :
ieee
Conference_Titel :
Engineering in Medicine and Biology Society (EMBC), 2015 37th Annual International Conference of the IEEE
ISSN :
1094-687X
Electronic_ISBN :
1558-4615
Type :
conf
DOI :
10.1109/EMBC.2015.7318480
Filename :
7318480
Link To Document :
بازگشت