Bilateral Filtering with Clusters by Expectation Maximization
نویسنده
چکیده
Image restoration keeping sharp edges is achieved by bilateral filter. In this paper, an approach to improve edges for the filter is presented. The proposed algorithm relies on clustering by Expectation Maximization that produced clusters of intensive values. A stage is followed where standard deviation of Gaussian filters for scales of the spatial and intensity are adjusted by features of the clusters. This makes the filters have adaptive levels of smoothing for specific clusters and helps to preserve edges while remove noise. Experiments and evaluation by PSNR metrics indicated the restoration quality enhanced and the efficacy of the proposed adaptive bilateral filter algorithm. Khôi phục ảnh đồng thời giữ các cạnh sắc nét được thực hiện bởi bộ lọc Gaussian kép. Trong bài báo này, một cách tiếp cận được trình bày để nâng cao độ nét cạnh cho bộ lọc trên. Thuật toán đề xuất được dựa trên kết quả phân cụm của phương pháp phân cụm kỳ vọng cực đại (EM) các giá trị mức xám. Theo đó, độ lệch chuẩn không gian và độ lệch chuẩn cường độ của bộ lọc Gaussian kép được điều chỉnh bởi đặc điểm của các cụm. Điều này làm cho bộ lọc Gaussian kép thích nghi với mức độ mịn của mỗi cụm, và giúp duy trì độ sắc nét cạnh trong khi loại bỏ nhiễu. Kết quả thực nghiệm với chỉ số đánh giá mức độ nhiễu (PSNR) cho thấy bộ lọc Gaussian kép mới đề xuất có chất lượng và hiệu quả được tăng cường. Index terms Image Processing, Bilateral Filter, Expectation Maximization Segmentation, Denoising
منابع مشابه
Bilateral Filter with Clusters by Expectation Maximization
Image restoration keeping sharp edges is achieved by bilateral filter. In this paper, an approach to improve edges for the filter is presented. The proposed algorithm relies on clustering by Expectation Maximization that produced clusters of intensive values. A stage is followed where standard deviation of Gaussian filters for scales of the spatial and intensity are adjusted by features of the ...
متن کاملDesign of Movie Recommendation System by Means of Collaborative Filtering
The purpose of this study is to develop a ‘Movie Recommendation System’ with the help of Collaborative Filtering approach. For this purpose we have used the Netflix prize dataset which is available for download from www.netflix.com. This dataset contains huge number of files. There are total 480189 users who have rated the movies. For clustering the users in the dataset according to their respe...
متن کاملClustered regression with unknown clusters
We consider a collection of prediction experiments, which are clustered in the sense that groups of experiments exhibit similar relationship between the predictor and response variables. The experiment clusters as well as the regression relationships are unknown. The regression relationships define the experiment clusters, and in general, the predictor and response variables may not exhibit any...
متن کاملSigma-Point Filtering and Smoothing Based Parameter Estimation in Nonlinear Dynamic Systems
We consider approximate maximum likelihood parameter estimation in nonlinear state-space models. We discuss both direct optimization of the likelihood and expectation– maximization (EM). For EM, we also give closed-form expressions for the maximization step in a class of models that are linear in parameters and have additive noise. To obtain approximations to the filtering and smoothing distrib...
متن کاملEffect of Post-Reconstruction Gaussian Filtering on Image Quality and Myocardial Blood Flow Measurement with N-13 Ammonia PET
Objective(s): In order to evaluate the effect of post-reconstruction Gaussian filtering on image quality and myocardial blood flow (MBF) measurement by dynamic N-13 ammonia positron emission tomography (PET), we compared various reconstruction and filtering methods with image characteristics. Methods: Dynamic PET images of three patients with coronary artery disease (male-female ratio of 2:1; a...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2015