From Heuristic Optimization to Dictionary Learning: A Review and Comprehensive Comparison of Image Denoising Algorithms

By Ling Shao, Ruomei Yan, Xuelong Li, Yan Liu. IEEE Transactions on Cybernetics, 2014

389 citations recorded by the source corpus.

Image denoising is a well explored topic in the field of image processing. In the past several decades, the progress made in image denoising has benefited from the improved modeling of natural images. In this paper, we introduce a new taxonomy based on image representations for a better understanding of state-of-the-art image denoising techniques. Within each category, several representative algorithms are selected for evaluation and comparison. The experimental results are discussed and analyzed to determine the overall advantages and disadvantages of each category. In general, the nonlocal methods within each category produce better denoising results than local ones. In addition, methods based on overcomplete representations using learned dictionaries perform better than others. The comprehensive study in this paper would serve as a good reference and stimulate new research ideas in image denoising.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/a1bd227c468843a269f00f33708c6cc453b75136

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Junwei Han
2. Gong Cheng
3. Xiwen Yao
4. Xuelong Li
5. Dacheng Tao
6. Lei Guo
7. Shuhui Bu
8. Peicheng Zhou

In Media Technology

1. Junwei Han
2. Gong Cheng
3. Xiwen Yao
4. Lei Guo
5. Shuhui Bu
6. Zhenbao Liu
7. Xuelong Li
8. Peicheng Zhou

Citations in the corpus

• Single-cell RNA-seq denoising using a deep count autoencoder (2018, methodology): “The concept of denoising is commonly used to delineate signal from noise in imaging 16 .”
• Rain Removal in Traffic Surveillance: Does it Matter? (2018, neutral): “ing techniques is given in [15].”
• Robust Joint Graph Sparse Coding for Unsupervised Spectral Feature Selection (2017, methodology): “1) Basis Extraction: The bases are derived from the training data via an existing dictionary learning method, such as [24] and [25].”
• Dirty Pixels: Optimizing Image Classification Architectures for Raw Sensor Data (2017, methodology): “The architecture combines low-level and high-level image processing units in a pipeline that takes raw sensor data as input and outputs image labels.”
• A coarse-to-fine model for airport detection from remote sensing images using target-oriented visual saliency and CRF (2015, background): “Recent developments have shown the effectiveness of using learning sparse representations in the context of face recognition [17,18], image denoising [19], image classification [20–22], and hyperspectral imagery analysis [23].”