Explore the vast range of titles available.

A blind/no-reference (NR) image quality assessment (IQA) algorithm based on natural scenes is developed. The proposed algorithm does not need training on databases of human judgments of distorted images or even prior knowledge about expected distortions (as is the case in most general NR IQA algorithms). To measure the image quality, the introduced approach uses a set of novel features in a spatial domain. The devised features are formed using the Weibull statistics of log-derivatives. When testing the proposed algorithm on the LIVE database, experiments showed that it correlates well with subjective opinion scores. In addition, they indicated that the new method has a good performance when compared with the state-of-the-art methods.

A fast and efficient method [fast efficient blind (FEB)] for no-reference image quality assessment (IQA) is presented. Two new features, log-energy and variance, are proposed in the spatial domain, which make the IQA algorithm faster and more efficient. FEB obviates the training process of distortion images and subjective opinion scores due to the properties of the new features. The experiment shows that the proposed method outperforms conventional methods in terms of both accuracy and execution speed and is also consistent with the subjective assessment of human beings. Owing to the simplicity of the features proposed, FEB can realise real-time IQA completely.

Image quality assessment (IQA) research strongly depends upon subjective experiments to provide ground truth to train and evaluate the IQA algorithms. These subjective experiments are cumbersome and expensive. An objective method based on human visual characteristics is proposed to generate the ground truth for distortion images. The proposed metric called Normalised Objective Distortion Score (NODS), using the logarithm of distortion parameter as the image quality score, is easily realised so that much manpower and time cost can be saved. The effectiveness of NODS has been analysed through experiments on five state-of-the-art IQA algorithms, and the result shows that the NODS is stable and can work as well as the subjective score when evaluating the performance of the IQA algorithms.

To improve data transmission efficiency, image compression is a commonly used method with the disadvantage of accompanying image distortion. There are many image restoration (IR) algorithms, and one of the most advanced algorithms is the generative adversarial network (GAN)-based method with a high correlation to the human visual system (HVS). To evaluate the performance of GAN-based IR algorithms, we proposed an ensemble image quality assessment (IQA) called ATDIQA (Auxiliary Transformer with DISTS IQA) to give weights on multiscale features global self-attention transformers and local features of convolutional neural network (CNN) IQA of DISTS. The result not only performed better on the perceptual image processing algorithms (PIPAL) dataset with images by GAN IR algorithms but also has good model generalization over LIVE and TID2013 as traditional distorted image datasets. The ATDIQA ensemble successfully demonstrates its performance with a high correlation with the human judgment score of distorted images.

Contrast is one of the most popular forms of distortion. Recently, the existing image quality assessment algorithms (IQAs) works focusing on distorted images by compression, noise and blurring. Reduced-reference image quality metric for contrast-changed images (RIQMC) and no reference-image quality assessment (NR-IQA) for contrast-distorted images (NR-IQA-CDI) have been created for CDI. NR-IQA-CDI showed poor performance in two out of three image databases, where the pearson correlation coefficient (PLCC) were only 0.5739 and 0.7623 in TID2013 and CSIQ database, respectively. Spatial domain features are the basis of NR-IQA-CDI architecture. Therefore, in this paper, the spatial domain features are complementary with curvelet domain features, in order to take advantage of the potent properties of the curvelet in extracting information from images such as multiscale and multidirectional. The experimental outcome rely on K-fold cross validation (K ranged 2-10) and statistical test showed that the performance of NR-IQA-CDI rely on curvelet domain features (NR-IQA-CDI-CvT) significantly surpasses those which are rely on five spatial domain features.

Image quality assessment (IQA) plays a critical role in automatically detecting and correcting defects in images, thereby enhancing the overall performance of image processing and transmission systems. While research on reference-based IQA is well-established, studies on no-reference image IQA remain underdeveloped. In this article, we propose a novel no-reference IQA algorithm based on transfer learning (IQA-NRTL). This algorithm leverages a deep convolutional neural network (CNN) due to its ability to effectively capture multi-scale semantic information features, which are essential for representing the complex visual perception in images. These features are extracted through a visual perception module. Subsequently, an adaptive fusion network integrates these features, and a fully connected regression network correlates the fused semantic information with global semantic information to perform the final quality assessment. Experimental results on authentically distorted datasets (KonIQ-10k, BIQ2021), synthetically distorted datasets (LIVE, TID2013), and an artificial intelligence (AI)-generated content dataset (AGIQA-1K) show that the proposed IQA-NRTL algorithm significantly improves performance compared to mainstream no-reference IQA algorithms, depending on variations in image content and complexity.

Underwater Optical Wireless Communication systems face severe signal attenuation, scattering, and turbulence, which significantly degrade image transmission quality and limit the communication range. To address these challenges, this paper proposes a secure and high-capacity RGB image transmission framework based on Optical Code Division Multiple Access (OCDMA) using Identity Row Shift Matrix (IRSM) codes. The IRSM-OCDMA scheme enhances data confidentiality by assigning unique orthogonal codes to each user while supporting simultaneous multiuser transmission with an aggregate rate of up to 30 Gbps. System performance is analyzed across five water types: Pure Seawater (PS), Clear Ocean, Coastal Ocean, Harbour I, and Harbour II (HR II), covering a broad range of attenuation coefficients. Image quality is quantitatively evaluated using standard metrics including Root Mean Square Error, Signal-to-Noise Ratio, Peak Signal-to-Noise Ratio, Structural Similarity Index Measure, and Correlation Coefficient. Two distinct post-processing methods are applied: median filtering for impulsive noise reduction and a Particle Swarm Optimization-based correction algorithm that adaptively restores image features under underwater channel conditions. Simulation results show a maximum transmission distance of 27 m in PS and 4 m in turbid HR II water, demonstrating the effectiveness of the proposed framework. The combination of IRSM coding with adaptive post-processing offers a robust solution for secure, high-quality image transmission in Internet of Underwater Things applications.

In this paper, we present an unsupervised, completely blind, no-reference (NR) stereoscopic (S3D) image quality prediction model to assess the perceptual quality of natural S3D images. We study the joint dependencies between color and depth features of S3D images and empirically model these dependencies by using a bivariate generalized Gaussian distribution (BGGD). We compute the parameters of BGGD, and we also obtain the determinant and the coherence values from the covariance matrix of the proposed BGGD model. We extract the features of BGGD model and covariance matrix from the reference S3D image, followed by multivariate Gaussian (MVG) distribution modeling on the predicted features of the reference. We estimate the joint color and depth quality of the S3D images by computing the likelihood of the image features with respect to the reference MVG model. We apply the popular 2D unsupervised NIQE model on individual stereo views to estimate the overall spatial quality of the S3D images. Finally, we pool the likelihood scores and the spatial NIQE scores to achieve the estimation for the overall perceived quality of the S3D images. The performance of the proposed model is evaluated on the MICT, LIVE Phase I and II S3D image datasets. The results indicate consistent and robust performance for all datasets. Our proposed estimator is completely blind, as it requires neither training on subjective scores nor reference S3D images.

Many image quality assessment algorithms (IQAs) have been developed during the past decade. However, most of them are designed for images distorted by compression, noise and blurring. There are very few IQAs designed specifically for Contrast Distorted Images (CDI), e.g. Reduced-reference Image Quality Metric for Contrast-changed images (RIQMC) and NR-IQA for Contrast-Distorted Images (NR-IQA-CDI). The existing NR-IQA-CDI relies on features designed by human or handcrafted features because considerable level of skill, domain expertise and efforts are required to design good handcrafted features. Recently, there is great advancement in machine learning with the introduction of deep learning through Convolutional Neural Networks (CNN) which enable machine to learn good features from raw image automatically without any human intervention. Therefore, it is tempting to explore the ways to transform the existing NR-IQA-CDI from using handcrafted features to machine-crafted features using deep learning, specifically Convolutional Neural Networks (CNN).The results show that NR-IQA-CDI based on non-pre-trained CNN (NR-IQA-CDI-NonPreCNN) significantly outperforms those which are based on handcrafted features. In addition to showing best performance, NR-IQA-CDI-NonPreCNN also enjoys the advantage of zero human intervention in designing feature, making it the most attractive solution for NR-IQA-CDI.

Image quality assessment (IQA) of fundus images constitutes a foundational step in automated disease analysis. This process is pivotal in supporting the automation of screening, diagnosis, follow-up, and related academic research for diabetic retinopathy (DR). This study introduced a deep learning-based approach for IQA of ultra-widefield optical coherence tomography angiography (UW-OCTA) images of patients with DR. Given the novelty of ultra-widefield technology, its limited prevalence, the high costs associated with equipment and operational training, and concerns regarding ethics and patient privacy, UW-OCTA datasets are notably scarce. To address this, we initially pre-train a vision transformer (ViT) model on a dataset comprising 6 mm × 6 mm OCTA images, enabling the model to acquire a fundamental understanding of OCTA image characteristics and quality indicators. Subsequent fine-tuning on 12 mm × 12 mm UW-OCTA images aims to enhance accuracy in quality assessment. This transfer learning strategy leverages the generic features learned during pre-training and adjusts the model to evaluate UW-OCTA image quality effectively. Experimental results demonstrate that our proposed method achieves superior performance compared to ResNet18, ResNet34, and ResNet50, with an AUC of 0.9026 and a Kappa value of 0.7310. Additionally, ablation studies, including the omission of pre-training on 6 mm × 6 mm OCTA images and the substitution of the backbone network with the ViT base version, resulted in varying degrees of decline in AUC and Kappa values, confirming the efficacy of each module within our methodology.