SSIM Values Research Articles

A generative adversarial network with multiscale and attention mechanisms for underwater image enhancement

Underwater images collected are often of low clarity and suffer from severe color distortion due to the marine environment and Illumination conditions. This directly impacts tasks such as marine ecological monitoring and underwater target detection, which rely on image processing. Therefore, enhancing Underwater images to improve their quality is necessary. A generative adversarial network with an encoder-decoder structure is proposed to improve the quality of Underwater images. The network consists of a generative network and an adversarial network. The generative network is responsible for enhancing the images, while the adversarial network determines whether the input is an enhanced image or a real high-quality image. In the generative network, we first design a residual convolution module to extract more texture and edge information from underwater images. Next, we design a multi-scale dilated convolution module to capture underwater features at different scales. Then, we design a feature fusion adaptive attention module to reduce the interference of redundant features and enhance the local perception capabilities. Finally, we construct the generative network using these modules along with conventional modules. In the adversarial network, we first design a multi-scale feature extraction module to improve the feature extraction ability. We then use the multi-scale feature extraction module along with conventional convolution modules to design the adversarial network. Additionally, we propose an improved loss function by introducing color loss into the conventional loss function. The improved loss function can better measure the color discrepancy between the enhanced image and the real image. It is useful to reduce color distortion in the enhanced images. In experimental simulations, the images enhanced by the proposed methods have the highest PSNR, SSIM, and UIQM values, indicating that the proposed method has superior Underwater image enhancement capabilities compared to other methods.

Utility of 3D facial reconstruction for forensic identification: a focus on facial soft tissue thickness and customized techniques.

Facial reconstruction, a crucial method in forensic identification, finds particular significance in cases where conventional means of identification are unavailable. This study addresses a significant gap in the field of forensic facial reconstruction focusing on facial soft tissue thickness (FSTT) and facial reconstruction techniques specifically tailored to the Thai population. By developing and implementing the 3D (three-dimensional) facial reconstruction program and compiling an extensive dataset of FSTT, this research makes substantial progress in advancing forensic facial reconstruction methodologies employing the combination Manchester Method, 3D skull images obtained through cone beam computed tomography (CBCT) scans were reconstructed using Autodesk Maya software. A dataset comprising 100 Thai cadavers underwent FSTT measurements via ultrasound (US) for 53 landmarks, with subsequent facial comparisons of 10 samples between reconstructed faces and real photographs conducted using the facial pool comparison and the structural similarity index (SSIM). The accuracy of facial pool comparison ranged from 30 to 80%, reflecting a wide range due to human errors. Thus, incorporating computerized assessment is necessary to minimize human bias. SSIM values ranged from 0.76 to 0.89, indicating strong similarity between reconstructed and real faces and validating the reconstruction process's accuracy. These findings suggest that the facial soft tissue thickness database of the Thai population used in this study can effectively support 3D computerized facial reconstruction. Moreover, this study sets the stage for future advancements in facial reconstruction methodologies tailored to diverse populations, emphasizing the ongoing need for comprehensive data gathering and technique refinement to enhance accuracy and applicability in forensic investigations.

Edge-Aware Dual-Task Image Watermarking Against Social Network Noise

In the era of widespread digital image sharing on social media platforms, deep-learning-based watermarking has shown great potential in copyright protection. To address the fundamental trade-off between the visual quality of the watermarked image and the robustness of watermark extraction, we explore the role of structural features and propose a novel edge-aware watermarking framework. Our primary innovation lies in the edge-aware secret hiding module (EASHM), which achieves adaptive watermark embedding by aligning watermarks with image structural features. To realize this, the EASHM leverages knowledge distillation from an edge detection teacher and employs a dual-task encoder that simultaneously performs edge detection and watermark embedding through maximal parameter sharing. The framework is further equipped with a social network noise simulator (SNNS) and a secret recovery module (SRM) to enhance robustness against common image noise attacks. Extensive experiments on three public datasets demonstrate that our framework achieves superior watermark imperceptibility, with PSNR and SSIM values exceeding 40.82 dB and 0.9867, respectively, while maintaining an over 99% decoding accuracy under various noise attacks, outperforming existing methods by significant margins.

Image Denoising Using Adaptive Weighted Low-Rank Matrix Recovery

This paper introduces a new image denoising method using adaptive weighted low-rank matrix recovery to tackle the challenges of separating low-rank information from noise and improving performance affected by empirical hyperparameters. We start by using image nonlocal similarity to build a low-rank denoising model, then apply the Gerschgorin theory to precisely determine the rank of the low -rank matrix. With this rank estimation, we use adaptive weighting along with singular value decomposition and weighted soft-thresholding to solve the denoising model, resulting in the denoised image. Experiments show our algorithm surpasses traditional denoising methods in average PSNR and SSIM. Specifically, for images contaminated with high-intensity noise (with a variance of 100), our algorithm achieves average PSNR and SSIM values of 24.66dB and 0.7267, respectively. Additionally, our algorithm exhibits superior performance in denoising images with real noise and is also applicable to color image denoising.

Two self-supervised image denoiser designs with discrete wavelet transform and non-local means-based algorithms

Image denoising is crucial in applications like medical imaging and photography, where restoring high-quality images from noisy data is essential. Traditional techniques often struggle with complex noise patterns, while deep learning-based methods typically rely on clean-noisy image pairs for training, limiting their practicality. Additionally, deep learning approaches face challenges such as the lack of ground truth clean images, sensitivity to specific noise types, and the introduction of artifacts during processing. In this work, we propose two novel self-supervised denoising approaches: a Discrete Wavelet Transform (DWT)-based model and a Non-Local Means (NLM)-based model. The DWT-based approach employs wavelet decomposition to separate image details across multiple frequency scales, selectively suppressing high-frequency noise via soft thresholding while preserving low-frequency components. The resulting wavelet coefficients are used to create pseudo-clean targets for training a U-Net architecture, ensuring effective denoising while maintaining structural integrity. The NLM-based approach leverages redundancy in image patches by applying the NLM algorithm to generate pseudo-clean targets through patch similarity-based averaging. These targets train a U-Net model with a custom loss function that balances Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index (SSIM), optimizing perceptual quality. Both models are trained on 5,000 noisy images from the ImageNet validation set without relying on clean references. Validated on synthetic Gaussian and Poisson noise at varying magnitudes, the DWT-based model achieved Mean PSNR and SSIM values of 31.07 and 0.9279, respectively, while the NLM-based model attained 30.17 and 0.9303. These results demonstrate the robustness and effectiveness of the proposed methods, making them suitable for real-world applications such as medical diagnostics and low-light photography.

Enhancing Cropland Mapping with Spatial Super-Resolution Reconstruction by Optimizing Training Samples for Image Super-Resolution Models

Mixed pixels often hinder accurate cropland mapping from remote sensing images with coarse spatial resolution. Image spatial super-resolution reconstruction technology is widely applied to address this issue, typically transforming coarse-resolution remote sensing images into fine spatial resolution images, which are then used to generate fine-resolution land cover maps using classification techniques. Deep learning has been widely used for image spatial super-resolution reconstruction; however, collecting training samples is often difficult for cropland mapping. Given that the quality of spatial super-resolution reconstruction directly impacts classification accuracy, this study aims to assess the impact of different types of training samples on image spatial super-resolution reconstruction and cropland mapping results by employing a Residual Channel Attention Network (RCAN) model combined with a spatial attention mechanism. Four types of samples were used for spatial super-resolution reconstruction model training, namely fine-resolution images and their corresponding coarse-resolution images, including original Sentinel-2 and degraded Sentinel-2 images, original GF-2 and degraded GF-2 images, histogram-matched GF-2 and degraded GF-2 images, and registered original GF-2 and Sentinel-2 images. The results indicate that the samples acquired by the histogram-matched GF-2 and degraded GF-2 images can resolve spectral band mismatches when simulating training samples from fine spatial resolution imagery, while the other three methods have limitations in their inability to fully address spectral and spatial mismatches. The histogram-matched method yielded the best image quality with PSNR, SSIM, and QNR values of 42.2813, 0.9778, and 0.9872, respectively, and produced the best mapping results, achieving an overall accuracy of 0.9306. By assessing the impact of training samples on image spatial super-resolution reconstruction and classification, this study addresses data limitations and contributes to improving the accuracy of cropland mapping, which is crucial for agricultural management and decision-making.

A Secure Data Hiding For H.264 Video Based on Chaotic Map Methods and RC4 Algorithm

Video steganography is the common field of data security in digital communication where data include text, image, and video. Compressed video is the suitable modern source for hiding huge secret data that will trust security. This work proposed a compressed video steganography scheme for hiding cipher secret data in high security levels. The proposed work consists of three phases, protected secret data (Text English and Arabic, Images, and Video) based on proposed chaotic maps module and RC4 algorithm. The chaotic maps module is a combination between (honen map and sine map) with simple operations to protected plain data. Second phase, hidden cipher secret data between the quantized discrete cosine transform (QDCT) and entropy coding from the inter-prediction mode at the H.264/AVC video frames using Least Significant Bit (LSB) replacement. Third phase, transmit and receive secret data. The experimental result shows that the proposed scheme is high-security level where was cipher secret data with the average entropy value of (7.993), an average PSNR value of (8.790 dB), an average correlation value of (3.4x 10-3), key length is 228 bit and key space is 2228. Moreover, measured Randomness to generation key by NIST were succeeded in all the tests and Robustness, imperceptibility, and embedding capacity are high while maintaining visual quality, where the average PSNR value of (36.26 dB), and the average SSIM value of (0.95).

Deep learning based super-resolution for CBCT dose reduction in radiotherapy.

Cone-beam computed tomography (CBCT) is a crucial daily imaging modality in image-guided and adaptive radiotherapy. However, the use of ionizing radiation in CBCT imaging increases the risk of secondary cancers, which is particularly concerning for pediatric patients. Deep learning super-resolution has shown promising results in enhancing the resolution of photographic and medical images but has not yet been explored in the context of CBCT dose reduction. To facilitate CBCT imaging dose reduction, we propose using an enhanced super-resolution generative adversarial network (ESRGAN) in both the projection and image domains to restore the image quality of low-dose CBCT. An extensive projection database, containing 2997 CBCT scans from head and neck cancer patients, was used to train two different ESRGAN models to generate super-resolution CBCTs. One model operated in the projection domain, using pairs of simulated low-resolution (low-dose) and original high-resolution (high-dose) projections and yielded CBCTSRpro. The other model operated in the image domain, using pairs of axial slices from reconstructed low-resolution and high-resolution CBCTs (CBCTLR and CBCTHR) and resulted in CBCTSRimg. Super-resolution CBCTs were evaluated in terms of image similarity (MAE, ME, PSNR, and SSIM), noise characteristics, spatial resolution, and registration accuracy, using the original CBCT as a reference. To test the perceptual difference between the original and super-resolution CBCT, we performed a visual Turing test. Visually, both super-resolution approaches in the projection and image domains improved the image quality of low-dose CBCTs. This was confirmed by the visual Turing test, that showed low accuracy, sensitivity, and specificity, indicating almost no perceptual difference between CBCTHR and the super-resolution CBCTs. CBCTSRimg (accuracy: 0.55, sensitivity: 0.59, specificity: 0.50) performed slightly better than CBCTSRpro (accuracy: 0.59, sensitivity: 0.61, specificity: 0.57). Image similarity metrics were affected by varying noise levels and did not reflect the visual improvements, with MAE/ME/PSNR/SSIM values of 110.4HU/2.9HU/40.4dB/0.82 for CBCTLR, 136.6HU/-0.4HU/38.6dB/0.77 for CBCTSRpro, and 128.2HU/1.9HU/39.0dB/0.80 for CBCTSRimg. In terms of spatial resolution, quantified by calculating 10% levels of the task transfer function, both CBCTSRpro and CBCTSRimg outperformed CBCTLR and nearly matched the reference CBCTHR (CBCTLR: 0.66lp/mm, CBCTSRpro: 0.88lp/mm, CBCTSRimg: 0.95lp/mm, CBCTHR: 1.01lp/mm). Noise characteristics of CBCTSRimg and CBCTSRpro were comparable to the reference CBCTHR. Registration parameters showed negligible differences for all CBCTs (CBCTLR, CBCTSRpro, CBCTSRimg), with average absolute differences in registration parameters being below 0.4° for rotations and below 0.06mm for translations (CBCTHR as reference). This study demonstrates that deep learning can be a valuable tool for CBCT dose reduction in CBCT-guided radiotherapy by acquiring low-dose CBCTs and restoring the image quality using deep learning super-resolution. The results suggest that higher quality images can be generated when super-resolution is performed in the image domain compared to the projection domain.

Mitigating high frame rate demands in shear wave elastography using radial basis function-based reconstruction: An experimental phantom study.

Shear wave elastography (SWE) is a technique that quantifies tissue stiffness by assessing the speed of shear waves propagating after being excited by acoustic radiation force. SWE allows the quantification of elastic tissue properties and serves as an adjunct to conventional ultrasound techniques, aiding in tissue characterization. To capture this transient propagation of the shear wave, the ultrasound device must be able to reach very high frame rates. In this paper, our aim is to relax the high frame rate requirement for SWE imaging. To this end, we propose lower frame rate SWE imaging followed by employing a 2-dimensional (2D) radial basis functions (RBF)-based interpolation. More specifically, the process involves obtaining low frame rate data and then temporal upsampling to reach a synthetic high frame rate data by inserting the 'UpS-1' image frames with missing values between two successive image frames (UpS: Upsampling rate). Finally, we apply the proposed interpolation technique to reconstruct the missing values within the incomplete high frame rate data. The results obtained from employing the proposed model on two experimental datasets indicate that we can relax the frame rate requirement of SWE imaging by a factor of 4 while maintaining shear wave speed (SWS), group velocity, and phase velocity estimates closely align with the high frame rate SWE model so that the error is less than 3%. Furthermore, analysis of the structural similarity index (SSIM) and root mean squared error (RMSE) on the 2D-SWS maps highlights the efficacy of the suggested technique in enhancing local SWS estimates, even at a downsampling (DS) factor of 4. For DS≤4, the SSIM values between the 2D-SWS maps produced by the proposed technique and those generated by the original high frame rate data consistently remain above 0.94. Additionally, the RMSE values is below 0.37 m/s, indicating promising performance of the proposed technique in reconstruction of SWS values.

Feedback projection synchronization of discrete chaotic systems and its application to speech encryption

Abstract Chaotic systems are widely used in secure communication due to their sensitivity to initial values, unpredictability, and complex motion trajectories. In this paper, we study the encryption method of chaotic synchronization and introduce a scaling factor based on traditional feedback control synchronization to achieve more accurate projection synchronization. The effectiveness and robustness of the method in chaotic systems are verified through theoretical proofs and numerical simulations. A chaotic masked speech encryption system utilizing bit similarity is designed; the structural similarity index (SSIM) of the decrypted signal with the original signal is as high as 0.992866, while the SSIM value of the encrypted signal with the original signal is only 0.000030, proving the efficiency and security of the encryption process. Additionally, we analyzed the data transmission process of the encryption system. The fusion of the control signal and the encryption sequence into one transmission sequence in the channel not only saves hardware and software design resources but also reduces inter-channel interference and conflict, improving the reliability and stability of the transmission. Experimental results show that the system performs well in terms of data transmission security and anti-interference capability.

Color Reproduction of Chinese Painting Under Multi-Angle Light Source Based on BRDF

It is difficult to achieve high-precision color reproduction using traditional color reproduction methods when the angle is changed, and, for large-sized artefacts, it is also significantly difficult to collect a large amount of data and reproduce the colors. In this paper, we use three Bidirectional Reflectance Distribution Function (BRDF) modeling methods based on spectral imaging techniques, namely, the five-parameter model, the Cook–Torrance model and the segmented linear interpolation model. We investigated the color reproduction of color chips with matte surfaces and Chinese paintings with rough surfaces under unknown illumination angles. Experiments have shown that all three models can effectively perform image reconstruction under small illumination angle intervals. The segmented linear interpolation model exhibits a higher stability and accuracy in color reconstruction under small and large illumination angle intervals; it can not only reconstruct color chips and Chinese painting images under any illumination angle, but also achieve high-quality image color reconstruction standards in terms of objective data and intuitive perception. The best test model (segmented linear interpolation) performs well in reconstruction, reconstructing Chinese paintings at 65° and 125° with an illumination angle interval of 10°. The average RMSE of the selected reference color blocks is 0.0450 and 0.0589, the average CIEDE2000 color difference is 1.07 and 1.50, and the SSIM values are 0.9227 and 0.9736, respectively. This research can provide a theoretical basis and methodological support for accurate color reproduction as well as the large-sized scientific prediction of artifacts at any angle, and has potential applications in cultural relic protection, art reproduction, and other fields.

ENHANCING UNDERWATER IMAGE QUALITY: EVALUATING COMBINATIVE APPROACHES FOR EFFECTIVE IN SEAGRASS BED ECOSYSTEM

The Complex underwater characteristics, challenges for image processing tasks. These images often have poor visibility due to low contrast, light scattering and various types of interference. There is a lack of exploration into the effectiveness of existing underwater image enhancement methods, particularly in the context of seagrass ecosystems, allows for further investigation. This study aims to explore and evaluate the effectiveness of various methods in underwater image enhancement, including Colour Balanced, CLAHE, and Unsharp Masking and their combinations, starting with converting video data from UTS devices into two-dimensional images. Furthermore, the quality of images taken from underwater cameras placed in a complex and wild seagrass meadow environment was improved using the proposed method, and the quality was evaluated by the SSIM value. The results show that the CLAHE method has the highest average SSIM value of 0.898. Meanwhile, the combined Color Balanced-CLAHE method achieved an SSIM value of 0.683 in a separate evaluation. This combination is an innovative approach to address complex underwater image quality problems, providing a more specific and adaptive solution. Overall, the proposed method is able to improve the visual quality of images on aspects such as clarity, color, and visibility of objects in the image

Abstract B010: Improving pancreatic cancer image classification with transfer learning: CD45 vs. PANCK

Abstract Pancreatic ductal adenocarcinoma (PDAC) is known for its high aggressiveness and is the fourth most common cause of death of cancer-related death in the United States. It is highly malignant and often goes undiagnosed until advanced stages due to its late-onset symptoms, resulting in poor prognosis. Using quantitative analyses of histological features, such as the expression of prognostic biomarker, can provide valuable insight to guide treatment decisions, monitor disease progression and predict outcomes. Previous researchers have found that CD45 and PANCK play a huge role in characterizing the tumor microenvironment of PDAC. One study using single-cell RNA sequencing compared the microenvironments of healthy pancreata and pancreatic tumors, revealing differences in cellular composition. Tumor samples showed a higher proportion of immune cells, including CD45+ cells, compared to heathy samples and utilized PANCK to differentiate between epithelial and non-epithelial cells. This study aims to enhance the accuracy of classifying pancreatic cancer pathological images based on prognostic markers using transfer learning techniques. We curated and pre-processed a dataset of pancreatic cancer pathological images with different prognostic markers. Using the pre-trained convolutional neural network (CNN) model VGG16, we fine-tuned it with our dataset to better differentiate between two prognostic markers, CD45 and PANCK. The model's performance was evaluated using a similarity matrix to illustrate distinctions among the two prognostic markers. Our results demonstrate that the transfer learning approach significantly improves classification accuracy, achieving confusion matrix validation results of 90% between CD45 and PANCK prognostic markers. SSIM values were relatively low, highlighting the ability of our model to differentiate between the two prognostic markers. The low SSIM value also shows variability within the two prognostic markers, indicating heterogeneity in structural changes due to pancreatic cancer. These findings indicate that transfer learning can be a powerful tool for improving the classification of pathological images and that there is notable variability in the structural features associated with different prognostic markers in pancreatic cancer. Citation Format: Miracle Thomas. Improving pancreatic cancer image classification with transfer learning: CD45 vs. PANCK [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Tumor-body Interactions: The Roles of Micro- and Macroenvironment in Cancer; 2024 Nov 17-20; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2024;84(22_Suppl):Abstract nr B010.

PoWBWM: Proof of work consensus cryptographic blockchain-based adaptive watermarking system for tamper detection applications

Image tamper detection is a challenging area in multimedia research. The advances in photography technology have made it possible to capture real-time high-dynamic-range (HDR) images through an iPhone or an Android device, which highlights the need for rigorous research on HDR images for tamper detection, and localization. The algorithms built for standard images may affect the watermark visibility when it is applied to HDR images as this produces high perceptual variations in the image compared to the original. To tackle this problem, we presented a Proof of Work consensus blockchain watermarking scheme combined with a convolution attention model (PoWBWM) system for tamper detection and localization. The system utilizes a Convolution Attention model (CoAtNet) to generate robust watermarks. A quaternion graph-based transform (QGBT) for embedding, ensuring imperceptibility and robustness. A fuzzy inference system optimizes embedding regions and factors based on human visual system characteristics. The system's security is enhanced through blockchain's proof-of-work (consensus) mechanism, providing a semi-blind watermarking scheme that authenticates ownership and detects tampering efficiently. The security is ensured only when the embedded hash key is authentic with its previous block to proceed further extraction process. The proposed algorithm's performance is evaluated in terms of its visibility by Peak-Signal-to-Noise-Ratio (PSNR), Structural Similarity Index (SSIM), and the perceptual quality of an HDR image is additionally measured by the Visual Dynamic Predictor (VDP) metric. On the other hand, the robustness performance is measured by Normalized Correlation Coefficient (NCC) and Bit Error Rate (BER). The experimental results for CASIA images achieved the highest PSNR value of 63.84 dB, and the SSIM value of 1.000, whereas the maximum VDP value obtained for HDR images is 98.02. In comparison with the existing system, the experimental findings of the suggested model show an effective tamper detection watermarking system as well as a robust against both intentional and unintentional attacks with an average NCC value of 0.98.

Asymmetric Large Kernel Distillation Network for efficient single image super-resolution.

Recently, significant advancements have been made in the field of efficient single-image super-resolution, primarily driven by the innovative concept of information distillation. This method adeptly leverages multi-level features to facilitate high-resolution image reconstruction, allowing for enhanced detail and clarity. However, many existing approaches predominantly emphasize the enhancement of distilled features, often overlooking the critical aspect of improving the feature extraction capabilities of the distillation module itself. In this paper, we address this limitation by introducing an asymmetric large-kernel convolution design. By increasing the size of the convolution kernel, we expand the receptive field, which enables the model to more effectively capture long-range dependencies among image pixels. This enhancement significantly improves the model's perceptual ability, leading to more accurate reconstructions. To maintain a manageable level of model complexity, we adopt a lightweight architecture that employs asymmetric convolution techniques. Building on this foundation, we propose the Lightweight Asymmetric Large Kernel Distillation Network (ALKDNet). Comprehensive experiments conducted on five widely recognized benchmark datasets-Set5, Set14, BSD100, Urban100, and Manga109-indicate that ALKDNet not only preserves efficiency but also demonstrates performance enhancements relative to existing super-resolution methods. The average PSNR and SSIM values show improvements of 0.10 dB and 0.0013, respectively, thereby achieving state-of-the art performance.

[formula omitted]-norm regularized quaternion matrix completion using sparse representation and approximate QSVD

Color image completion is a challenging problem in computer vision. Despite the effectiveness of low-rank quaternion matrix completion (LRQMC) methods, their reliance on quaternion singular value decomposition (QSVD) makes them computationally expensive. We propose a novel method based on quaternion Qatar Riyal decomposition (QQR) and quaternion L2,1-norm, named QLNM-QQR, which reduces computational complexity by avoiding the need for calculating the QSVD of large quaternion matrices. Furthermore, we introduce two improvements: IRQLNM-QQR, which utilizes iteratively reweighted quaternion L2,1-norm minimization, and QLNM-QQR-SR, which integrates sparse regularization. Additionally, an analysis of the convergence of the proposed algorithms is conducted. Our experiments show that IRQLNM-QQR outperforms QLNM-QQR on both natural and medical color images. The QLNM-QQR-SR method achieves higher PSNR and SSIM values than state-of-the-art methods, with PSNR averaging 4–8 dB higher than non-sparse methods and 0.4 dB higher than the sparse-based TNN-SR method.

Enhancing Black-and-White Image Colorization Using Cycle GAN: A Study on Unpaired Image Translation

Abstract. This paper explores the application of Cycle-Consistent Generative Adversarial Network (Cycle GAN) for the colorization of black-and-white images, addressing the challenges in image to the image translation and it was not requiring paired datasets. Cycle GAN, a generative adversarial network, uses cycle consistency loss to map grayscale images to color, effectively preserving structural integrity. The model is trained on unpaired datasets containing black-and-white and color images, enabling it to generate colorized versions from grayscale inputs. Performance evaluation is conducted using Two significant evaluation metrics: Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM). Results demonstrate that Cycle GAN effectively preserves image structure and detail, achieving high PSNR and SSIM values, particularly in scenes with distinct color boundaries. However, challenges arise in complex and low-resolution scenes where color accuracy and detail reconstruction may degrade. Additionally, this paper assesses the impact of different loss functions, learning rate schedules, and training strategies on the models output quality. Future work will focus on addressing these limitations by incorporating attention mechanisms and other advanced techniques to improve Cycle GAN's ability to handle intricate image features and complex scenes.

Research on the generation and evaluation of bridge defect datasets for underwater environments utilizing CycleGAN networks

The surface cracks on the underwater structures critically damages the overall reliability of the structures and reduces their strength. It is significant to monitor these cracks in timely manner. Recently, deep learning algorithms have been used for large scale data study and predictions. However, deep supervised learning algorithms need to get training on large scale data set which is time consuming and difficult to apply on the underwater structures. Therefore, it is highly needed to address these issues. Current research proposes an improved cycle-constraint generative adversarial algorithm for the timely detection of surface cracks in underwater structures. It utilizes an enhanced cycle-consistent generative adversarial network (CycleGAN). The proposed algorithm uses image processing techniques including DeblurGAN and Dark channel prior methods to get quality of dataset from underwater structures. The proposed Algorithm introduces a novel cross-domain VGG-cosine similarity assessment to precisely evaluate the performance of proposed algorithm to retain crack information etc. Moreover, performance of proposed algorithm is evaluated through both qualitative and quantitative methods. The quantitative results are directly obtained from the visual results are presented which are generated by the proposed Algorithm. Whereas, the performance of proposed algorithm based on quantitative results is obtained from metrics including PSNR, SSIM, and FID. Experimental results indicates that the proposed algorithm outperforms the original CycleGAN. End results indicate that the proposed algorithm decreased the value of FID by 20 % and increased the values of PSNR and SSIM by 2.37 % and 3.33 % respectively. Quantitative and qualitative results of the proposed algorithm give significant advantages during creating of surface crack images.

Inpainting of damaged temple murals using edge- and line-guided diffusion patch GAN.

Mural paintings are vital cultural expressions, enriching our lives by beautifying spaces, conveying messages, telling stories, and evoking emotions. Ancient temple murals degrade over time due to natural aging, physical damage, etc. Preserving these cultural treasures is challenging. Image inpainting is often used for digital restoration, but existing methods typically overlook naturally degraded areas, using randomly generated binary masks or small, narrow regions for repair. This study proposes a novel architecture to reconstruct large areas of naturally degraded murals, maintaining intrinsic details, avoiding color bias, and preserving artistic excellence. The architecture integrates generative adversarial networks (GANs) and the diffusion model, including a whole structure formation network (WSFN), a semantic color network (SCN), and a diffusion mixture distribution (DIMD) discriminator. The WSFN uses the original image, a line drawing, and an edge map to capture mural details, which are then texturally inpainted in the SCN using gated convolution for enhanced results. Special attention is given to globally extending the receptive field for large-area inpainting. The model is evaluated using custom-degraded mural images collected from Tamil Nadu temples. Quantitative analysis showed superior results than state-of-the-art methods, with SSIM, MSE, PSNR, and LPIPS values of 0.8853, 0.0021, 29.8826, and 0.0426, respectively.

Design of Shape Forming Elements for Architected Composites via Bayesian Optimization and Genetic Algorithms: A Concept Evaluation.

This article presents the first use of shape forming elements (SFEs) to produce architected composites from multiple materials in an extrusion process. Each SFE contains a matrix of flow channels connecting input and output ports, where materials are routed between corresponding ports. The mathematical operations of rotation and shifting are described, and design automation is explored using Bayesian optimization and genetic algorithms to select fifty or more parameters for minimizing two objective functions. The first objective aims to match a target cross-section by minimizing the pixel-by-pixel error, which is weighted with the structural similarity index (SSIM). The second objective seeks to maximize information content by minimizing the SSIM relative to a white image. Satisfactory designs are achieved with better objective function values observed in rectangular rather than square flow channels. Validation extrusion of modeling clay demonstrates that while SFEs impose complex material transformations, they do not achieve the material distributions predicted by the digital model. Using the SSIM for results comparison, initial stages yielded SSIM values near 0.8 between design and simulation, indicating a good initial match. However, the control of material processing tended to decline with successive SFE processing with the SSIM of the extruded output dropping to 0.023 relative to the design intent. Flow simulations more closely replicated the observed structures with SSIM values around 0.4 but also failed to predict the intended cross-sections. The evaluation highlights the need for advanced modeling techniques to enhance the predictive accuracy and functionality of SFEs for biomedical, energy storage, and structural applications.