Paper
17 May 2019 Laser medical image processing based on neighborhood concerning Gaussian mixture model
Author Affiliations +
Proceedings Volume 11170, 14th National Conference on Laser Technology and Optoelectronics (LTO 2019); 111703K (2019) https://doi.org/10.1117/12.2533924
Event: Fourteenth National Conference on Laser Technology and Optoelectronics, 2019, Shanghai, China
Abstract
The traditional Gaussian Mixture Model is sensitive to noise in laser medical image processing, and its segmentation accuracy is not high enough. In order to remedy these defects, the Neighborhood Concerning Gaussian Mixture Model is adopted. The gray value of the center pixel in every neighborhood block is updated by the information of the neighborhood pixels according to the close correlation between them. After remolded, the image is decomposed by double Gaussian Mixture Model, which further refines the decomposition of the mixture model and improves the accuracy and anti-noise performance of image segmentation. Finally, experiments are carried out to verify the feasibility of the method. The results of the experiments show that the method of image segmentation based on the Neighborhood Concerning Gaussian Mixture Model can significantly improve the noise suppression ability, and achieve a good segmentation effect with high efficiency, while retaining the image contour and details better.
© (2019) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yuting Lv "Laser medical image processing based on neighborhood concerning Gaussian mixture model", Proc. SPIE 11170, 14th National Conference on Laser Technology and Optoelectronics (LTO 2019), 111703K (17 May 2019); https://doi.org/10.1117/12.2533924
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Image processing

Image segmentation

Medical imaging

Medical laser equipment

Laser processing

Speckle

Performance modeling

RELATED CONTENT


Back to Top