Explore the vast range of titles available.

Image deblurring for dynamic scenes is a serious challenge in computer vision. Motion blur is caused by camera shaking or object movement during the exposure time. Many photos cannot be reproduced at the moment they were taken, its contents cannot be restored if motion blur occurs. In this article, we proposed a deblurring system that uses a two-stage convolutional neural network (CNN) to achieve image deblurring through a joint learning strategy. The first-stage network predicts the deblur kernel of each pixel and pre-deblurs the input image, and then the second-stage network directly predicts clear images based on U-Net architecture. In the first-stage network, the deblur kernel uses the surrounding information to restore the centre pixel, which can effectively remove the tiny motion blur. To additionally deal with large motion blur, we extend the second-stage network is used to compensate for the limited receptive field of the first-stage deblurring kernel. We evaluate the proposed method on benchmark blur datasets. Experimental results show that the proposed method can produce better results than state-of-the-art methods, both quantitatively and qualitatively. The proposed method can achieve the best PSNR at 32.59db, 27.21db and 31.96db for the GOPRO, Köhler, and Su datasets, respectively.

Deep learning has been developed by computer scientists. Here, we discuss techniques for improving the image quality of diagnostic computed tomography and magnetic resonance imaging with the aid of deep learning. We categorize the techniques for improving the image quality as “noise and artifact reduction”, “super resolution” and “image acquisition and reconstruction”. For each category, we present and outline the features of some studies.

Background Magnetic resonance imaging (MRI) is an essential tool for medical diagnosis. However, artifacts may degrade images obtained through MRI, especially owing to patient movement. Existing methods that mitigate the artifact problem are subject to limitations including extended scan times. Deep learning architectures, such as U-Net, may be able to address these limitations. Optimizing deep learning networks with batch normalization (BN) and dropout layers enhances their convergence and accuracy. However, the influence of this strategy on U-Net has not been explored for artifact removal. Methods This study developed a U-Net-based regression network for the removal of motion artifacts and investigated the impact of combining BN and dropout layers as a strategy for this purpose. A Transformer-based network from a previous study was also adopted for comparison. In total, 1200 images (with and without motion artifacts) were used to train and test three variations of U-Net. Results The evaluation results demonstrated a significant improvement in network accuracy when BN and dropout layers were implemented. The peak signal-to-noise ratio of the reconstructed images was approximately doubled and the structural similarity index was improved by approximately 10% compared with those of the artifact images. Conclusions Although this study was limited to phantom images, the same strategy may be applied to more complex tasks, such as those directed at improving the quality of MR and CT images. We conclude that the accuracy of motion artifact removal can be improved by integrating BN and dropout layers into a U-Net-based network, with due consideration of the correct location and dropout rate.

In recent years the application of digital images has increased in a rapid manner, but due to noise the limited applications are currently openly adopting these applications. The noise can degrade the image quality and the application’s quality of service too. However, in literature, there are a number of different kinds of noise available and to rectify different types of noise different image filtering techniques are also available. But most of the techniques are computationally expensive or less effective, less efficient, and not able to preserve the image features for higher levels of noise. Therefore, in this paper, we introduced an optimization technique for impulse noise removal and measured its effect on the different levels of noise in the image. The proposed filter detects and removes impulse noise from digital grey-scale images. Thus the algorithm first classifies the image pixels in terms of noisy and non-noisy pixels. Here the classification of pixels has been carried out using the regression analysis of the image vector. After locating the corrupted pixel the mean of self and neighbor pixels which are non-noisy (except pixel values 0 and 255) has been used to replace the noisy pixel. However, this technique is not completely removing the noise in a single step thus we eliminate the noise in an iterative manner. Additionally to deal with the blurring effect and to preserve the image edges we employ L0 smoothing. Finally, in the last step, we utilize the median filter for constructing the final output image. The simulation of the proposed algorithm has been carried out with MATLAB and with the help of a publically available dataset. The experiments have been carried out and performance is measured in terms of the visual quality histogram and PSNR (pick signal to noise ratio). The comparison with the relevant techniques demonstrates the effective denoising consequences of the proposed technique.

The object recognition under the sea is a difficult task due to the presence of different environmental conditions around oceans. Various techniques have been used to enhance the properties of underwater images. This paper proposed the pre-processing and feature extraction techniques for object recognition under the seawater. The proposed methodology consists of several pre-processing steps for underwater images to make a compatible image for feature extraction purposes, which helps in object recognition. The proposed pre-processing works on the intensity, contrast, and sharpness of the object to improve their visualization quality. We compared the proposed ALE algorithm (Atmospheric-Light-Enhancement Algorithm) with existing image enhancement techniques such as Intensity-based, Histogram based, Contrast-based image enhancement techniques. From the analysis, the ALE algorithm is most effective for underwater images. Also, a comparative analysis has been presented for feature extraction from underwater images using PCA (Principal Component Analysis), SIFT (Scale-Invariant Feature Transform), and SURF (Speeded up Robust Features) to extract the unique feature. The proposed technique SURF shows better results attainment in contrary to other feature extraction techniques. At last, in simulation analysis, we observed that the error rate and feature extraction time taken by SURF is better than PCA as well as SIFT, due to the speed up methodology of SURF during the feature points filtration.

The average error rate in liver cirrhosis classification on B-mode ultrasound images using the traditional pattern recognition approach is still too high. In order to improve the liver cirrhosis classification performance, image correction methods and a convolution neural network (CNN) approach are focused on. The impact of image correction methods on region of interest (ROI) images that are input into the CNN for the purpose of classifying liver cirrhosis based on data from B-mode ultrasound images is investigated. In this paper, image correction methods based on tone curves are developed. The experimental results show positive benefits from the image correction methods by improving the image quality of ROI images. By enhancing the image contrast of ROI images, the image quality improves and thus the generalization ability of the CNN also improves.

This paper presents a statistically grounded algorithm for surface imaging with linear frequency-modulated continuous wave synthetic aperture radar. The approach is based on the maximum likelihood principle, where solving the optimization problem naturally leads to the introduction of a spectral decorrelation filter. The proposed method increases the effective number of statistically independent samples, reduces speckle, and improves the accuracy of radar cross section estimation. Simulation experiments demonstrate consistent advantages over classical SAR processing: the proposed method achieves up to a 21% improvement in feature similarity metrics and an average 4% improvement across standard quantitative image quality measures.

Overcoming visual degradation in challenging imaging scenarios is essential for accurate scene understanding. Although deep learning methods have integrated various perceptual capabilities and achieved remarkable progress, their high computational cost limits practical deployment under resource-constrained conditions. Moreover, when confronted with diverse degradation types, existing methods often fail to effectively model the inconsistent attenuation across color channels and spatial regions. To tackle these challenges, we propose DWMamba , a degradation-aware and weight-efficient Mamba network for image quality enhancement. Specifically, DWMamba introduces an Adaptive State Space Module (ASSM) that employs a dual-stream channel monitoring mechanism and a soft fusion strategy to capture global dependencies. With linear computational complexity, ASSM strengthens the models ability to address non-uniform degradations. In addition, by leveraging explicit edge priors and region partitioning as guidance, we design a Structure-guided Residual Fusion (SGRF) module to selectively fuse shallow and deep features, thereby restoring degraded details and enhancing low-light textures. Extensive experiments demonstrate that the proposed network delivers superior qualitative and quantitative performance, with strong generalization to diverse extreme lighting conditions. The code is available at https://github.com/WindySprint/DWMamba .

[...]studies in this field can be divided into two aspects: first, new end-to-end neural network models for merging constituent parts during the image fusion process; Second, the embodiment of artificial neural networks for image processing systems. [...]a feature pyramid module (SC-FP) based on spatial and channel attention can perform the multi-scale fusion of features accompanied by feature selection. Electricity transmission line monitoring in hazy weather will face some problems, such as reduced contrast and chromatic aberration. [...]Zhang M et al. proposed an image defogging algorithm for the electricity transmission line monitoring system. In this research, an optimized quadtree segmentation method for calculating global atmospheric light was proposed. [...]the detail sharpening post-processing based on visibility and air light level was introduced to enhance the detail level of electricity transmission lines in the defogging image.