Novel Evaluation Index for Image Quality

Robust structural similarity index measure for images with non-Gaussian distortions

Aim: Image is the most powerful tool to analyze the information. Sometimes the captured image gets affected with blur and noise in the environment, which degrades the quality of the image. Image restoration is a technique in image processing where the degraded image can be restored or recovered to its nearest original image. Materials and Methods: In this research Lucy-Richardson algorithm is used for restoring blurred and noisy images using MATLAB software. And the proposed work is compared with Wiener filter, and the sample size for each group is 30. Results: The performance was compared based on three parameters, Power Signal to Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), Normalized Correlation (NC). High values of PSNR, SSIM and NC indicate the better performance of restoration algorithms. Lucy-Richardson provides a mean PSNR of 10.4086db, mean SSIM of 0.4173%, and NC of 0.7433% and Wiener filter provides a mean PSNR of 6.3979db, SSIM of 0.3016%, NC of 0.3276%. Conclusion: Based on the experimental results and statistical analysis using independent sample T test, image restoration using Lucy-Richardson algorithm significantly performs better than Wiener filter on restoring the degraded image with PSNR (P<0.001) and SSIM (P<0.001).

The size of pixels in a digital recording device, such as a CCD array, limits the spatial resolution in images obtained by an optical imaging system, thereby degrading the image quality. Digital super-resolution (DSR) techniques are used to reconstruct a high-resolution (HR) image from multiple sub-pixel-shifted low-resolution images in order to improve the image quality. In this article, we formulate a mathematical framework for DSR using the generalized sampling theorem (GST). The GST-based DSR method’s performance is evaluated by comparing it to existing resolution enhancement methods on the basis of evaluation metrics like percentage mean square error (%MSE), structural similarity index measure (SSIM), and peak signal-to-noise ratio (PSNR). The GST DSR method exhibits an overall superior quality image reconstruction based on quantitative analysis (with a near zero %MSE, SSIM around one, and improved PSNR in dB) and qualitative comparison. The robustness of the GST DSR method is further demonstrated in the presence of frame-to-frame shift estimation error using %MSE and SSIM by comparing it with multi-frame interpolation approaches.

Image quality assessment based on the space similarity decomposition model

In-memory computing facilitates efficient parallel computing based on the programmable memristor crossbar array. Proficient hardware image processing can be implemented by utilizing the analog vector-matrix operation with multiple memory states of the nonvolatile memristor in the crossbar array. Among various materials, 2D materials are great candidates for a switching layer of nonvolatile memristors, demonstrating low-power operation and electrical tunability through their remarkable physical and electrical properties. However, the intrinsic device-to-device (D2D) variation of memristors within the crossbar array can degrade the accuracy and performance of in-memory computing. Here, we demonstrate hardware image processing using the fabricated 2D hexagonal boron nitride-based memristor to investigate the effects of D2D variation on the hardware convolution process. The image quality is evaluated by peak-signal-to-noise ratio, structural similarity index measure, and Pratt’s figure of merit and analyzed according to D2D variations. Then, we propose a novel two-step gradual reset programming scheme to enhance the conductance uniformity of multiple states of devices. This approach can enhance the D2D variation and demonstrate the improved quality of the image processing result. We believe that this result suggests the precise tuning method to realize high-performance in-memory computing.

Introduction[18F]fluorodeoxyglucose ([18F]FDG) brain PET is used clinically to detect small areas of decreased uptake associated with epileptogenic lesions, e.g., Focal Cortical Dysplasias (FCD) but its performance is limited due to spatial resolution and low contrast. We aimed to develop a deep learning-based PET image enhancement method using simulated PET to improve lesion visualization.MethodsWe created 210 numerical brain phantoms (MRI segmented into 9 regions) and assigned 10 different plausible activity values (e.g., GM/WM ratios) resulting in 2100 ground truth high quality (GT-HQ) PET phantoms. With a validated Monte-Carlo PET simulator, we then created 2100 simulated standard quality (S-SQ) [18F]FDG scans. We trained a ResNet on 80% of this dataset (10% used for validation) to learn the mapping between S-SQ and GT-HQ PET, outputting a predicted HQ (P-HQ) PET. For the remaining 10%, we assessed Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), and Root Mean Squared Error (RMSE) against GT-HQ PET. For GM and WM, we computed recovery coefficients (RC) and coefficient of variation (COV). We also created lesioned GT-HQ phantoms, S-SQ PET and P-HQ PET with simulated small hypometabolic lesions characteristic of FCDs. We evaluated lesion detectability on S-SQ and P-HQ PET both visually and measuring the Relative Lesion Activity (RLA, measured activity in the reduced-activity ROI over the standard-activity ROI). Lastly, we applied our previously trained ResNet on 10 clinical epilepsy PETs to predict the corresponding HQ-PET and assessed image quality and confidence metrics.ResultsCompared to S-SQ PET, P-HQ PET improved PNSR, SSIM and RMSE; significatively improved GM RCs (from 0.29 ± 0.03 to 0.79 ± 0.04) and WM RCs (from 0.49 ± 0.03 to 1 ± 0.05); mean COVs were not statistically different. Visual lesion detection improved from 38 to 75%, with average RLA decreasing from 0.83 ± 0.08 to 0.67 ± 0.14. Visual quality of P-HQ clinical PET improved as well as reader confidence.ConclusionP-HQ PET showed improved image quality compared to S-SQ PET across several objective quantitative metrics and increased detectability of simulated lesions. In addition, the model generalized to clinical data. Further evaluation is required to study generalization of our method and to assess clinical performance in larger cohorts.

Speckle noise in ultrasound images compromises image quality and hinders diagnostic accuracy. Traditional ultrasound denoising methods often struggle to preserve anatomical details while effectively reducing noise, especially under high-noise conditions. In this study, we propose an innovative approach that integrates a lightweight channel attention mechanism (LCAM) within a convolutional variational autoencoder (CVAE) framework to enhance ultrasound image denoising. The proposed approach efficiently reduces speckle noise while maintaining essential anatomical features. Comprehensive evaluations across six diverse ultrasound datasets demonstrate that the LCAM-CVAE outperforms conventional denoising techniques in both subjective image quality and objective performance metrics, including peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), standard deviation in PSNR (SD-PSNR), standard deviation in SSIM (SD-SSIM), PSNR statistical relevance tests, and computational efficiency (CE). The LCAM-CVAE approach demonstrates exceptional performance, particularly under high-noise conditions, ensuring the preservation of key anatomical structures for accurate diagnosis. These results highlight the LCAM-CVAE approach as a robust and promising solution for ultrasound image denoising, with significant clinical potential to improve diagnostic quality in noisy environments.

Quiescent frame, contrast-enhanced coronary magnetic resonance angiography reconstructed using limited number of physiologic frames from 5D free-running acquisitions

Steganography as a practice is highly useful as it allows information to be concealed within digital images while maintaining visual quality. However, it can be a challenge to balance imperceptibility, capacity, and robustness while ensuring that embedded data remains secure against statistical and Deep Learning (DL)-based steganalysis. Therefore, we propose a new framework that utilizes the BOSSbase dataset for cover images, the USC-SIPI dataset for secreted images, and introduces adaptive image steganography, which combines fused map-based embedding with evolutionary optimization to solve existing problems. The system starts by encrypting the secret image through Blowfish cipher encryption to protect its confidentiality. The model generates a fused map through a combination of entropy and Laplacian-based noise maps from the cover image to determine the most suitable embedding areas. Next, we optimize the process of embedding location selection and ordering using Particle Swarm Optimization (PSO) to maximize both security and image quality. Next, we use priority-guided Least Significant Bits (LSB) substitution in the embedding process to generate an optimized priority map for inserting the encrypted bitstream into the image. The system can function on different cover image sizes, so at [Formula: see text] the resolution is 51-61 dB Peak Signal-to-Noise Ratio (PSNR) and 0.9972-1.0 Structural Similarity Index Measure (SSIM) for stego images at 0.1-1.0 Bits Per Pixel (BPP) with 262,144 bits while maintaining perfect secret reconstruction (SSIM = 1.000) and processing times under 0.25 seconds for embedding and extraction. Security tests demonstrate that the proposed system remains undetectable since the Area Under Curve (AUC) for [Formula: see text] reaches 0.49-0.57 across all payload levels and Regular Singular (RS) statistics remain at 0.59-0.66 between 0.1-1.0 BPP. The proposed model provides effective capacity while maintaining high visual quality and security against DL-based steganalysis, such as that using Convolutional Neural Networks (CNN).

Sinogram domain metal artifact correction of CT via deep learning

In this paper an efficient approach “Hybrid Assessment and Enhancement technique for Foreign Object Detection (HAE-FOD)” is proposed to detect debris in the runway. Foreign objects such as engine fasteners or aircraft parts are left in the runway when an aircraft takes off. These objects may cause damage to life and property when they are not detected. In practice, they are identified manually which is tiresome and not always accurate. Computer vision using image processing can be of aid to detect the foreign objects. So, an automated system for detection of foreign objects in runway is proposed in this paper. It facilitates accurate foreign object detection under varying lighting and environment conditions. Images from runway are assessed for their quality and enhanced for detection of foreign objects. Images are assessed to determine the quality using the combined Image Quality Assessment Techniques of Histogram, SSIM (Structural Similarity Index Measure) and PSNR Ratio. Then, the image quality is improved using Enhancement Techniques with the combination of Gamma Correction, CLAHE (Contrast Limited Adaptive Histogram Equalization) and Wiener Filter. To identify Foreign Objects from the enhanced images, Image Segmentation Techniques such as Edge Detection, Background Subtraction and Temporal Differencing are used. The proposed hybrid system with the combination of enhancement techniques based on type of degradation provides a better way for object detection in runway.

Recent work has shown that deep convolutional neural network is capable of solving inverse problems in computational imaging, and recovering the stress field of the loaded object from the photoelastic fringe pattern can also be regarded as an inverse problem solving process. However, the formation of the fringe pattern is affected by the geometry of the specimen and experimental configuration. When the loaded object produces complex fringe distribution, the traditional stress analysis methods still face difficulty in unwrapping. In this study, a deep convolutional neural network based on the encoder–decoder structure is proposed, which can accurately decode stress distribution information from complex photoelastic fringe images generated under different experimental configurations. The proposed method is validated on a synthetic dataset, and the quality of stress distribution images generated by the network model is evaluated using mean squared error (MSE), structural similarity index measure (SSIM), peak signal-to-noise ratio (PSNR), and other evaluation indexes. The results show that the proposed stress recovery network can achieve an average performance of more than 0.99 on the SSIM.

In this work we propose a novel SSIM (Structural Similarity Index Measure)-guided brain tissue classification approach, implementing Kernel Fisher Discriminant Analysis (KFDA). In Computer Vision, KFDA has been shown to be competitive with other state-of-the-art techniques. In the KFDA-based framework, we exploit the complex structure of grey matter, white matter and cerebro-spinal fluid intensity clusters to find an optimal classification. We illustrate our novel technique using a dataset of early normal brain development in the age range from 10 days to 4.5 years. The SSIM metric, an objective measure of an image quality as perceived by the Human Visual System, is used to evaluate the quality of brain segmentation. SSIM comparison of the quality of classification obtained by the KFDA-based and the Expectation-Maximization algorithms shows the superior performance of the proposed technique.

Monte Carlo (MC)-based estimation of dose-averaged linear energy transfer (LETd) in proton therapy is accurate but computationally intensive. Although deep learning (DL) models offer efficient alternatives, few have been optimized for multi-source inputs or validated on anatomically complex tumors, such as nasopharyngeal carcinoma (NPC). This study developed and evaluated a multi-source DL framework for voxel-level LETd prediction in proton therapy for NPC, aiming to identify optimal input configurations and assess both prediction accuracy and clinical reliability. This study analyzed 100 patients with NPC, yielding 500 proton therapy fields. Ground-truth LETd maps were generated using GPU-based Monte Carlo simulations. A deep learning model was trained with different combinations of four input features: dose distribution, CT image, beam mask, and radiological depth, resulting in eight configurations (Mdose, Mdose+ct, Mdose+beam, Mdose+depth, Mdose+beam+depth, Mdose+ct+depth, Mdose+ct+beam, and Mdose+ct+beam+depth). The dataset was randomly divided into training (300 fields, 60 patients), validation (100 fields, 20 patients), and testing (100 fields, 20 patients) sets. Model performance was evaluated using mean absolute error (MAE), peak signal-to-noise ratio (PSNR), normalized cross-correlation (NCC), structural similarity index measure (SSIM), and 3D Gamma analysis. A 3D U-Net baseline was retrained under identical conditions for comparison. Model interpretability was examined with Gradient-weighted Class Activation Mapping (Grad-CAM), and prediction confidence was assessed using uncertainty maps. The Mdose+ct model achieved the best performance (MAE 0.21±0.03keV/µm, PSNR 28.99±1.33, NCC 0.98±0.01, SSIM 0.971±0.007, Gamma 3%/3mm: 94.69±3.69%), outperforming the 3D U-Net baseline (MAE 0.38±0.52keV/µm). Grad-CAM visualizations confirmed that the model focused on physically relevant regions, while uncertainty maps highlighted high-variability zones at LETd transitions. The model generated voxel-level LETd predictions in approximately 7 s per case, which was substantially faster than GPU-based MC simulations (30 s per case). The proposed multi-source deep learning framework enables rapid and accurate voxel-level LETd prediction for NPC. Combining dose with CT yielded the most reliable input configuration, and incorporating interpretability and uncertainty analyses enhances confidence in its suitability for biologically guided proton therapy planning.

More and more high-spatial resolution satellite images are produced with the improvement of satellite technology. However, the quality of images is not always satisfactory for application. Due to the impact of complicated atmospheric conditions and complex radiation transmission process in imaging process the images often suffer deterioration. In order to assess the quality of remote sensing images over urban areas, we proposed a general purpose image quality assessment methods based on feature extraction and machine learning. We use two types of features in multi scales. One is from the shape of histogram the other is from the natural scene statistics based on Generalized Gaussian distribution (GGD). A 20-D feature vector for each scale is extracted and is assumed to capture the RS image quality degradation characteristics. We use SVM to learn to predict image quality scores from these features. In order to do the evaluation, we construct a median scale dataset for training and testing with subjects taking part in to give the human opinions of degraded images. We use ZY3 satellite images over Wuhan area (a city in China) to conduct experiments. Experimental results show the correlation of the predicted scores and the subjective perceptions.