Welcome to the InfoSci Platform
Restoration of CT Images Corrupted With Fixed Valued Impulse Noise Using an Optimum Decision-Based Filter

Restoration of CT Images Corrupted With Fixed Valued Impulse Noise Using an Optimum Decision-Based Filter

Priyank Saxena, R. Sukesh Kumar
Copyright: © 2018 |Pages: 20
ISBN13: 9781522552468|ISBN10: 1522552464|EISBN13: 9781522552475
DOI: 10.4018/978-1-5225-5246-8.ch008
Cite Chapter Cite Chapter

MLA

Saxena, Priyank, and R. Sukesh Kumar. "Restoration of CT Images Corrupted With Fixed Valued Impulse Noise Using an Optimum Decision-Based Filter." Intelligent Multidimensional Data and Image Processing, edited by Sourav De, et al., IGI Global, 2018, pp. 220-239. https://doi.org/10.4018/978-1-5225-5246-8.ch008

APA

Saxena, P. & Kumar, R. S. (2018). Restoration of CT Images Corrupted With Fixed Valued Impulse Noise Using an Optimum Decision-Based Filter. In S. De, S. Bhattacharyya, & P. Dutta (Eds.), Intelligent Multidimensional Data and Image Processing (pp. 220-239). IGI Global. https://doi.org/10.4018/978-1-5225-5246-8.ch008

Chicago

Saxena, Priyank, and R. Sukesh Kumar. "Restoration of CT Images Corrupted With Fixed Valued Impulse Noise Using an Optimum Decision-Based Filter." In Intelligent Multidimensional Data and Image Processing, edited by Sourav De, Siddhartha Bhattacharyya, and Paramartha Dutta, 220-239. Hershey, PA: IGI Global, 2018. https://doi.org/10.4018/978-1-5225-5246-8.ch008

Export Reference

Mendeley
Favorite

Abstract

The main aim of this chapter is to perform the restoration of computed tomography (CT) images acquired at the reduced level of radiation dose. Reduction in radiation dose affects the image quality as it increases noise and decreases low contrast resolution. In this chapter, an optimum decision-based filter (ODBF) is proposed as an image-space denoising technique, to detect and restore the low dose CT (LDCT) images corrupted with fixed valued impulse noise (salt and pepper) of unequal density. The detection stage employs k-means clustering to discriminate the noise-free pixels from the noisy-pixels by splitting the image data into three clusters of different intensities. The restoration stage employs mask else trimmed median (METM) estimation followed by an optional adaptive mask sizing for restoration of noisy pixels. The proposed method demonstrates noticeable improvement over other existing methods in restoration of LDCT images while maintaining the image contrast and edge details.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.