During image collection, images are often polluted by noise because of imaging conditions and equipment limitations. Images are also disturbed by external noise during compression and transmission, which adversely affects consequent processing, like image segmentation, target recognition, and text detection. A two-dimensional amplitude image is one of the most common image categories, which is widely used in people’s daily life and work. Research on this kind of image-denoising algorithm is a hotspot in the field of image denoising. Conventional denoising methods mainly use the nonlocal self-similarity of images and sparser representatives in the converted domain for image denoising. In particular, the three-dimensional block matching filtering (BM3D) algorithm not only effectively removes the image noise but also better retains the detailed information in the image. As artificial intelligence develops, the deep learning-based image-denoising method has become an important research direction. This review provides a general overview and comparison of traditional image-denoising methods and deep neural network-based image-denoising methods. First, the essential framework of classic traditional denoising and deep neural network denoising approaches is presented, and the denoising approaches are classified and summarized. Then, existing denoising methods are compared with quantitative and qualitative analyses on a public denoising dataset. Finally, we point out some potential challenges and directions for future research in the field of image denoising. This review can help researchers clearly understand the differences between various image-denoising algorithms, which not only helps them to choose suitable algorithms or improve and innovate on this basis but also provides research ideas and directions for subsequent research in this field.

Convolutional neural networks have achieved excellent results in image denoising; however, there are still some problems: (1) The majority of single-branch models cannot fully exploit the image features and often suffer from the loss of information. (2) Most of the deep CNNs have inadequate edge feature extraction and saturated performance problems. To solve these problems, this paper proposes a two-branch convolutional image denoising network based on nonparametric attention and multiscale feature fusion, aiming to improve the denoising performance while better recovering the image edge and texture information. Firstly, ordinary convolutional layers were used to extract shallow features of noise in the image. Then, a combination of two-branch networks with different and complementary structures was used to extract deep features from the noise information in the image to solve the problem of insufficient feature extraction by the single-branch network model. The upper branch network used densely connected blocks to extract local features of the noise in the image. The lower branch network used multiple dilation convolution residual blocks with different dilation rates to increase the receptive field and extend more contextual information to obtain the global features of the noise in the image. It not only solved the problem of insufficient edge feature extraction but also solved the problem of the saturation of deep CNN performance. In this paper, a nonparametric attention mechanism is introduced in the two-branch feature extraction module, which enabled the network to pay attention to and learn the key information in the feature map, and improved the learning performance of the network. The enhanced features were then processed through the multiscale feature fusion module to obtain multiscale image feature information at different depths to obtain more robust fused features. Finally, the shallow features and deep features were summed using a long jump join and were processed through an ordinary convolutional layer and output to obtain a residual image. In this paper, Set12, BSD68, Set5, CBSD68, and SIDD are used as a test dataset to which different intensities of Gaussian white noise were added for testing and compared with several mainstream denoising methods currently available. The experimental results showed that this paper’s algorithm had better objective indexes on all test sets and outperformed the comparison algorithms. The method in this paper not only achieved a good denoising effect but also effectively retained the edge and texture information of the original image. The proposed method provided a new idea for the study of deep neural networks in the field of image denoising.

Traditional median filtering with a fixed window easily leads to edge blurring and adaptive median filtering requires manual presetting of the maximum window parameter and has insufficient retention of details when dealing with high-density salt-and-pepper noise. Aiming at these problems, this paper proposes a hybrid decision-making adaptive median filtering algorithm with dual-window detection in collaboration with particle swarm optimization (PSO). The algorithm quickly locates suspected noise points through a 3 × 3 small window and enhances noise identification accuracy by using a PSO dynamically optimized 5–35-pixel large window. Meanwhile, a hybrid decision-making mechanism based on local statistical properties was introduced to dynamically select median filtering, weighted average based on spatial distance, or pixel preservation strategy to balance noise suppression and detail preservation, and the PSO algorithm was used to automatically find the optimal parameters of the large window’s size to avoid the manual parameter-tuning process. Experiments were conducted on standard grayscale and color images and compared with four traditional methods and two more advanced methods. The experiments showed that the algorithm improved the peak signal-to-noise ratio (PSNR) value by 2–4 dB and the structural similarity index measure (SSIM) metric by 0.05–0.2 under high salt-and-pepper noise density compared with the traditional methods, which effectively improved the contradiction between noise suppression and detail retention in traditional filtering algorithms and provided a highly efficient and intelligent solution for image denoising in high-noise scenarios.

To address the limitations of traditional non-local mean (NLM) denoising algorithms in terms of neighborhood similarity metrics, weight calculation, and computational efficiency, this paper proposed a structural-tensor-driven and dynamic window-based fast non-local mean denoising algorithm with dual kernel weighting. First, a Gaussian–Tukey dual-kernel weighting function was designed to optimize similarity metrics. Then, spatial neighborhood features were adopted. By measuring both grayscale similarity and spatial correlation, the weight distribution rationality was further enhanced. Second, structural tensor eigenvalues were used to quantify regional structural properties. A dynamic window allocation function was designed to adaptively match search window sizes to different image regions. Finally, an integral image acceleration mechanism was proposed, significantly improving algorithm execution efficiency. Experimental results demonstrated that the proposed algorithm achieved both excellent denoising performance and edge/texture preservation capabilities. In high-noise environments, its Peak Signal-to-Noise Ratio (PSNR) outperformed the Gauss kernel non-local mean algorithm by an average of 1.96 dB, while Structural Similarity (SSIM) improved by an average of 5.7%. Moreover, the algorithm’s execution efficiency increased by approximately 7–11 times, indicating strong potential for real-time application in digital image processing.