Title :
Texture-Energy-Analysis-Based Image Recognition Algorithm for Identifying the Early Stage Tumor
Author :
He, Yuanlie ; Chen, Ping ; Chen, Yangchun
Author_Institution :
Fac. of Comput., Guangdong Univ. of Technol., Guangzhou
Abstract :
A new algorithm (image registration and texture energy analysis - ITEAR), which can accurately identify the early stage tumor in CT images, is developed in this paper. The ITEAR algorithm aligns the corresponding PET and CT images by the mutual information registration method and applies the adaptive threshold segmentation algorithm to segmenting the region of interest (ROI) in PET images. Then, the texture energy measurement method and edge detection processes are implemented to locate the accurate position of the early stage tumor and extract its edge. It has been proved by experiments that the ITEAR algorithm can be used to accurately locate the early stage tumor in any part of human body.
Keywords :
cancer; computerised tomography; edge detection; image recognition; image registration; image segmentation; image texture; medical image processing; positron emission tomography; tumours; CT images; ITEAR; PET images; adaptive threshold segmentation algorithm; early stage tumor; edge detection; image recognition algorithm; image registration; mutual information registration; region of interest; texture-energy-analysis; Algorithm design and analysis; Computed tomography; Image analysis; Image edge detection; Image recognition; Image registration; Image segmentation; Image texture analysis; Neoplasms; Positron emission tomography; Image Recognition; Image Registration; Texture Analysis;
Conference_Titel :
Computing, Communication, Control, and Management, 2008. CCCM '08. ISECS International Colloquium on
Conference_Location :
Guangzhou
Print_ISBN :
978-0-7695-3290-5
DOI :
10.1109/CCCM.2008.233