Previous Techniques Research Articles

A Novel Three-Stage Dehazing Model Using Improved Auto-Color Transfer Method, Adaptive Dehazing, and Adaptive Contrast Enhancement

The problems of underexposure, under-detail-enhancement and residual haze are detected in the previous dehazing techniques. These issues occur due to different reasons and are highly difficult to be resolved using a single algorithm. Therefore, a three-stage dehazing model (TSDM) is proposed in this paper using pre-processing, dehazing and post-processing modules. The improved auto-color transfer (IACT) approach is presented as part of pre-processing to efficiently enhance the hazy image to overcome underexposure. Also, adaptive dehazing (AD) is developed in this work which considers the global characteristics of the hazy image as a parameter, to adaptively enhance the details. Moreover, adaptive contrast enhancement (ACE) is proposed as a post-processing operation that adaptively fuses the dehazed image and its contrast-enhanced image to effectively improve the contrast. However, the IACT operation is performed on the hazy image only when dark regions are detected. Similarly, the ACE is performed only when a dehazed image exhibits residual haze. Based on these prior conditions, the proposed work can be implemented in four distinct ways i.e. using only the AD technique; using IACT and AD approaches, using AD and ACE methods, and using all IACT, AD and ACE algorithms. The proposed TSDM is experimentally analyzed using many databases which shows improved results compared to previous techniques.

TmfimCLIP: Text-Driven Multi-Attribute Face Image Manipulation

Text-to-image conversion has garnered significant research attention, with contemporary methods leveraging the latent space analysis of StyleGAN. However, issues with latent code decoupling, interpretability, and controllability often remain, leading to misaligned image attributes. To address these challenges, we propose a refined approach that segments StyleGAN’s latent code using the Visual Language Model (CLIP). Our method aligns the latent code segments with text embeddings via an image-text alignment module and modulates them through a text injection module. Additionally, we incorporate semantic segmentation loss and mouth loss to constrain operations that affect irrelevant attributes. Compared to previous CLIP-driven techniques, our approach significantly enhances decoupling, interpretability, and controllability. Experiments on the CelebA-HQ and FFHQ datasets validate our model’s efficacy through both qualitative and quantitative comparisons. Our model effectively handles a wide range of style variations, achieving an FID score of 21.15 for facial attributes and an ID metric of 0.88 for hair attributes.

LIA: Latent Image Animator.

Previous animation techniques mainly focus on leveraging explicit structure representations (e.g., meshes or keypoints) for transferring motion from driving videos to source images. However, such methods are challenged with large appearance variations between source and driving data, as well as require complex additional modules to respectively model appearance and motion. Towards addressing these issues, we introduce the Latent Image Animator (LIA), streamlined to animate high-resolution images. LIA is designed as a simple autoencoder that does not rely on explicit representations. Motion transfer in the pixel space is modeled as linear navigation of motion codes in the latent space. Specifically such navigation is represented as an orthogonal motion dictionary learned in a self-supervised manner based on proposed Linear Motion Decomposition (LMD). Extensive experimental results demonstrate that LIA outperforms state-of-the-art on VoxCeleb, TaichiHD, and TED-talk datasets with respect to video quality and spatio-temporal consistency. In addition LIA is well equipped for zero-shot high-resolution image animation. Code, models, and demo video are available at https://github.com/wyhsirius/LIA.

FINSYNC AI: Stock Market Analysis

The goal of the study "FINSYNC AI" is to enhance the accuracy and effectiveness of financial market forecasts by introducing a complex stock market prediction system. This work makes use of the frameworks Gradient Boosting Machine, Random Forest, Long Short-Term Memory (LSTM), and the proposed FINSYNC AI. The process involves acquiring historical stock market data, cleaning and normalizing it, and then feature engineering to produce new features like rolling means and trend components. These preprocessed data are then used to test and train several machine learning models in order to assess their predictive accuracy. Given the inadequacies of previous techniques, the FINSYNC AI is proposed as a means to improve their effectiveness. The project is broken down into three primary modules to guarantee a comprehensive process: creating and refining the framework, obtaining and prepping the data, and the prediction stage and evaluation. The Python implementation of the machine learning tasks makes use of scikit-learn, pandas for data management, yfinance for acquiring stock data, and matplotlib for visualization. The results show that the FINSYNC AI performs better than traditional one, making it a more useful tool for value trading in the stock market

Accuracy-preassigned fixed-time synchronization of switched inertial neural networks with time-varying distributed, leakage and transmission delays

In this paper, the accuracy-preassigned fixed-time synchronization problem of a class of switched inertial neural networks with time-varying distributed, leakage and transmission delays is studied. To this end, a parameterized system solution-based direct analysis method is proposed for the first time. Unlike existing works, this method sets out from the definition of accuracy-preassigned fixed-time synchronization, and does not require variable substitution for inertial item or the construction of any Lyapunov–Krasovskii functional. This not only simplifies the proof process, but also reduces the computational complexity for solving synchronization conditions. Significantly, this paper introduced the time-varying leakage delay into switched inertial neural networks for the first time. Furthermore, the approach utilized in this manuscript stands apart from all previous techniques for achieving fixed-time synchronization. Finally, the reliability of the theoretical results is verified by numerical simulation.

Dynamic Lévy–Brownian marine predator algorithm for photovoltaic model parameters optimization

The dynamic and multimodal nature of photovoltaic (PV) systems makes it challenging to examine all solar photovoltaic characteristics. Consequently, this study recommends a recently developed optimization method called the marine predator algorithm (MPA) for developing reliable PV models. In the traditional MPA, the two main search processes are Lévy flight (LF) and Brownian walk (BW), and the switch across them is unpredictable. This is while the transition between these two mechanisms is naturally continuous and dynamic. To rectify the limitation mentioned above, this research paper presents an innovative, dynamic shift function that effectively modulates the interplay that exists between the BW and LF procedures. By enhancing the changeover pattern between the primary phases of MPA, the suggested dynamic walk substantially boosts the performance of MPA. The dynamic Lévy-Brownian MPA (DLBMPA) is also made to be resilient in dealing with the parameterization limitations of PV Modeling approaches by using a constraint handling technique. The performance of DLBMPA is tested using ten popular optimization methods. Employing the DLBMPA achieved an average RMSE of 9.7 × 10− 4 in the parameter estimation across a number of multiple PV models, including the SDM, DDM, and TDM, where out of the ten optimization algorithms experimented, this was statistically significant (p < 0.05) better. In terms of averaged computation time, DLBMPA was 13 ms and still showed high accuracy in dealing with different irradiance and temperature levels. These improvements allow for MBPA to be credited as having a high efficiency when estimating the PV parameters since its speed of convergence and accuracy level surpass the previous techniques used.

Improving ViT interpretability with patch-level mask prediction

Vision Transformers (ViTs) have demonstrated remarkable performances on various computer vision tasks. Attention scores are often used to explain the decision-making process of ViTs, showing which tokens are more important than others. However, the attention scores have several limitations as an explanation for ViT, such as conflicting with other explainable methods or highlighting unrelated tokens. In order to address this limitation, we propose a novel method for generating a visual explanation map from ViTs. Unlike previous approaches that rely on attention scores, our method leverages ViT features and conducts a single forward pass through our Patch-level Mask prediction (PM) module. Our visual explanation map provides class-dependent and probabilistic interpretation that can identify crucial regions of model decisions. Experimental results demonstrate that our approach outperforms previous techniques in both classification and interpretability aspects. Additionally, it can be applied to the weakly-supervised object localization (WSOL) tasks using pseudo mask labels. Our method requires no extra parameters and necessitates minimal locality supervision, utilizing less than 1% of the ImageNet-1k training dataset.

Improved 3D laser point cloud reconstruction for autonomous mobile robot applications by using SVM-R technique

PurposeIn autonomous mobile robots, high-level accuracy and precision in 3D perception are required for object detection, shape estimation and obstacle distance measurement. However, the existing methods suffer from limitations like inaccurate point clouds, noise in sensor data and synchronization problems between 2D LiDAR and servomotor. These factors can lead to the wrong perception and also introduce noise during sensor registration. Thus, the purpose of this study is to address these limitations and enhance the perception in autonomous mobile robots.Design/methodology/approachA new sensor mounting structure is developed for 3D mapping by using a 2D LiDAR and servomotor. The proposed method uses a support vector machine regression (SVM-R) technique to optimize the waypoints of the servomotor for the point cloud reconstruction process and to obtain a highly accurate and detailed representation of the environment.FindingsThe study includes an analysis of the SVM-R model, including Linear, radial basis function (RBF) and Polynomial kernel. Results show that the Linear kernel performs better with the lowest 3.67 mean absolute error (MAE), 26.24 mean squared error (MSE) and 5.12 root mean squared error (RMSE) values than the RBF and Polynomial kernels. The 2D to 3D point cloud reconstruction shows that the proposed method with a new sensor mounting structure performs better in perception accuracy and achieves an error of 0.45% in measuring the height of the target objects whereas in previous techniques the error was very large.Originality/valueThe study shows the effectiveness of SVM-R in the 3D point cloud reconstruction process and exhibits remarkable performance for object height measurement. Further, the proposed technique is applicable for future advanced visual applications and has a superior performance over the other conventional methods.

Identifying the minimal sets of distance restraints for FRET-assisted protein structural modeling.

Proteins naturally occur in crowded cellular environments and interact with other proteins, nucleic acids, and organelles. Since most previous experimental protein structure determination techniques require that proteins occur in idealized, non-physiological environments, the effects of realistic cellular environments on protein structure are largely unexplored. Recently, Förster resonance energy transfer (FRET) has been shown to be an effective experimental method for investigating protein structure in vivo. Inter-residue distances measured in vivo can be incorporated as restraints in molecular dynamics (MD) simulations to model protein structural dynamics in vivo. Since most FRET studies only obtain inter-residue separations for a small number of amino acid pairs, it is important to determine the minimum number of restraints in the MD simulations that are required to achieve a given root-mean-square deviation (RMSD) from the experimental structural ensemble. Further, what is the optimal method for selecting these inter-residue restraints? Here, we implement several methods for selecting the most important FRET pairs and determine the number of pairs that are needed to induce conformational changes in proteins between two experimentally determined structures. We find that enforcing only a small fraction of restraints, , where is the number of amino acids, can induce the conformational changes. These results establish the efficacy of FRET-assisted MD simulations for atomic scale structural modeling of proteins in vivo.

New crater-pattern matching using a projective invariant for lost-in-space scenario

This paper addresses the challenge of crater matching in lost-in-space scenarios. An innovative algorithm is designed for robust and efficient crater matching by utilizing projective invariant values. Unlike other existing approaches, the proposed algorithm generates grid-shape patterns as a distinctive feature, thereby significantly enhancing recognition accuracy. Notably, the algorithm employs a stepwise matching process involving filtering, matching, and verification to ensure fast and reliable matching. Extensive numerical simulations have been conducted to assess the algorithm’s performance across various scenarios, including variations in crater center positions, shape distortions, and the presence of false craters. The outcomes demonstrate that the proposed algorithm significantly surpasses previous crater matching techniques, delivering superior stability and consistent accuracy even under challenging lost-in-space conditions.

In-Device Ballistic-Electron-Emission Spectroscopy for Accurately In Situ Mapping Energy Level Alignment at Metal-Organic Semiconductors Interface.

Energy level alignment at metal/organic semiconductors (OSCs) interface governs electronic processes in organic electronics devices, making its precise determination essential for understanding carrier transport behaviors and optimizing device performance. However, it is proven that accurately characterizing the energy barrier at metal/OSC interface under operational conditions remains challenging due to the technical limitations of traditional methods. Herein, through integrating highly-improved device constructions with an ingenious derivative-assisted data processing method, this study demonstrates an in-device ballistic-electron-emission spectroscopy using hot-electron transistors to accurately characterize the energy barrier at metal/OSC interface under in-operando conditions. This technique is found that a remarkable improvement in measurement accuracy, reaching up to ±0.03eV, can be achieved-surpassing previous techniques (±0.1-0.2eV). The high accuracy allows us to monitor subtle changes in energy barriers at metal/OSC interface caused by variations in the aggregation state of OSCs, a phenomenon that is theoretically possible but failed to be directly demonstrated through conventional methods. Moreover, this study makes demonstration that this technology is universally applicable to various metal/OSC interfaces consisting of electron-transporting, hole-transporting, and ambipolar OSCs. These findings manifest the great potential of this method to advance both theoretical exploration and technical applications in organic electronics.

A VHF insect-tracking radio-telemeter ASIC with on-chip power-regulation in 65-nm CMOS

This paper demonstrates a VHF radio-telemeter ASIC which for the first time can replace the PCB based telemeter circuit to yield around 37 % miniaturization in size and weight of the state-of-the-art in insect-telemeter and facilitate wireless-tracking of small flying insects in real-time. The chip employs the newly developed all-digital method of ultra-low duty-cycle generation technique, which unlike all the previous techniques does not require any passive components. The ASIC is designed to generate FSK modulated 8-bit coded burst-mode radio-signal at 0.0078 duty-cycle with a signal-window width of 13.1 ms using an on-chip oscillator operating at around 160 MHz. The oscillator frequency was temperature compensated employing a CTAT voltage-supply from a Power-Regulation Unit (PRU) integrated on the ASIC. The PRU steps down the 1.55V supply-voltage for the operation of the ASIC telemeter and consists of CTAT and Band-gap voltage-reference circuits along-with two voltage-regulators, one to drive the digital core-circuitry with temperature compensation and the other to supply the output-driver at constant 1V. The prototype chip was fabricated on a 1mm × 1 mm chip in TSMC 65nmGP CMOS process and the functional performance was experimentally verified against supply-voltage and temperature variation.

A Deep Cryptographic Framework for Securing the Healthcare Network from Penetration.

Ensuring the security of picture data on a network presents considerable difficulties because of the requirement for conventional embedding systems, which ultimately leads to subpar performance. It poses a risk of unauthorized data acquisition and misuse. Moreover, the previous image security-based techniques faced several challenges, including high execution times. As a result, a novel framework called Graph Convolutional-Based Twofish Security (GCbTS) was introduced to secure the images used in healthcare. The medical data are gathered from the Kaggle site and included in the proposed architecture. Preprocessing is performed on the data inserted to remove noise, and the hash 1 value is computed. Using the generated key, these separated images are put through the encryption process to encrypt what they contain. Additionally, to verify the user's identity, the encrypted data calculates the hash 2 values contrasted alongside the hash 1 value. Following completion of the verification procedure, the data are restored to their original condition and made accessible to authorized individuals by decrypting them with the collective key. Additionally, to determine the effectiveness, the calculated results of the suggested model are connected to the operational copy, which depends on picture privacy.

Arthroscopic Congruent-Arc Distal Tibial Allograft Bone Augmentation With Cerclage Fixation for Anterior Shoulder Instability With Bone Loss

Glenoid bone loss is a common sequela of recurrent anterior shoulder instability. When treated surgically, a bone augmentation procedure is more likely to provide a good outcome with a lower risk of recurrent instability compared with a soft-tissue repair alone. There are several bone augmentation options to address glenoid bone loss, including coracoid transfer, iliac crest autograft, and distal tibial allograft (DTA). Both open and arthroscopic DTA techniques have been described using the lateral one-third of the distal tibia oriented with the anterior aspect placed superiorly on the anterior glenoid. This graft orientation places its concavity in the superior-to-inferior direction. It may be preferred to have this concavity maximized toward the anterior direction so that it can further resist anterior displacement. The posterior one-third of the distal tibia may provide a more optimal arc of resistance. However, no previous technique has used the posterior one-third of the distal tibia to reconstruct the “half-pipe” concavity of the anterior glenoid. We describe an all-arthroscopic technique for a posterior third DTA oriented with the posterior tibial edge placed anteriorly on the glenoid, termed the “congruent-arc DTA,” to address anterior shoulder instability with bone loss.

A Novel 3‐D Seismic Scattering and Intrinsic Attenuation Tomography and Its Application to Northern Sumatra

AbstractStudying the subsurface structure of volcanoes is crucial for understanding volcanic mechanisms, current status, and potential risks. However, the intricate physical and chemical processes occurring over geological timescales make it challenging to characterize subsurface features such as volcanic structures and hydrothermal systems. Given the highly attenuating nature of magma, 3‐D scattering and intrinsic attenuation tomography are critical methods for advancing our understanding of tectonic, magmatic, and hydrothermal processes and their interactions. Previous imaging techniques, however, required substantial memory usage and long computational times, limiting their application to only 1‐D velocity models. This paper proposes a novel sensitivity‐kernel calculation method for imaging shear wave seismic scattering and intrinsic attenuation. This method has the advantages of dramatically reducing memory and computational costs, as well as incorporating a 3‐D seismic velocity model. We apply this approach to illustrate the 3‐D scattering and intrinsic attenuation structures beneath the Toba volcano region in Northern Sumatra down to 20 km depth. Our results show high‐intrinsic attenuation anomalies around the Toba caldera, revealing the magma chambers feeding the volcanoes. A conspicuous high‐scattering attenuation anomaly is identified along the Great Sumatran Fault, possibly caused by the fault zone structure. Magmatism also likely contributes to the seismic activity south of the Toba caldera, as evidenced by the overlap of scattering and intrinsic attenuation anomalies.

Robust embedded access control system based on face and encrypted QR with RPi4

&lt;p&gt;&lt;span&gt;Facial-based recognition systems are commonly used for building access control, with the accuracy and computing requirements still being improved. On the other hand, QR codes are gaining rising attention as an input interface to many embedded applications. This paper proposes an embedded access control system that customises both previous techniques to be implemented on the CPU of a low-cost Raspberry Pi 4 computer. The achieved system works smoothly with a frame rate of 8.27 FPS, increasing the accessing control's robustness compared to a system based on face recognition only. It also offers the ability to control the access of unknown faces. In tandem with integration, this strengthens security measures, improves user experience, and outperforms conventional access control approaches, creating an attractive offer for many businesses.&lt;/span&gt;&lt;/p&gt;

Data-driven bifurcation analysis using parameter-dependent trajectories

Identification of bifurcation diagrams in nonlinear systems is of great importance for resilient design and stability analysis of dynamical systems. Data-driven identification of bifurcation diagrams has a significant advantage for large dimensional systems where analysis of the equations is not possible, and for experimental systems where accurate system equations are not available. In this work, a novel forecasting approach to predict bifurcation diagrams in nonlinear systems is presented using system trajectories before instabilities occur. Unlike previous techniques, the proposed method considers a varying bifurcation parameter during the system response to perturbations. Combined with an asymptotic analysis provided by the method of multiple scales eliminates the requirement of using multiple measurements and allows the novel technique to predict the bifurcation diagram using a single system recovery. The proposed approach allows stability analyses of nonlinear systems with limited access to experimental or surrogate data. The novel technique is demonstrated through its application to a Hopf bifurcation, highlighting its inherent advantages. Subsequently, the method is employed in the analysis of an aeroelastic system that shows both supercritical and subcritical Hopf bifurcations. The findings reveal great accuracy, achieved with a reduced number of measurements, while enhancing versatility.

A Proposal for a High-Frequency Earthquake Magnitude (m3Hz) for Seismic Hazard and Rapid Damage Assessment

Abstract Obtaining a size estimate of an earthquake that well represents its potential impact is of primary importance in seismology and earthquake engineering. The magnitude scales currently used in earthquake catalogs, mainly Ms and Mw, are not designed to represent the variability of shaking due to variations in high-frequency radiation. Conversely, ML, which best describes the seismic-wave energy released by an earthquake, is unable to provide an accurate representation of events with magnitudes exceeding approximately 6.5. Therefore, building on and extending the concept of high-frequency magnitude (Atkinson and Hanks, 1995), we propose in this article a new high-frequency magnitude scale m3Hz. This magnitude is estimated by also considering correction factors for the crustal model and site effects. We apply the new magnitude scale m3Hz to two data sets (Central Italy and Japan) and validate its ability to capture high-frequency radiation effects through comparison with available source parameters. Finally, a procedure for estimating m3Hz is also proposed for preinstrumental earthquakes. This procedure addresses a limitation of previous magnitude estimation techniques, such as Me, and is validated for four events for which both macroseismic and instrumental intensity data are available.

Anisotropic Specular Image‐Based Lighting Based on BRDF Major Axis Sampling

AbstractAnisotropic specular appearances are ubiquitous in the environment: brushed stainless steel pans, kettles, elevator walls, fur, or scratched plastics. Real‐time rendering of these materials with image‐based lighting is challenging due to the complex shape of the bidirectional reflectance distribution function (BRDF). We propose an anisotropic specular image‐based lighting method that can serve as a drop‐in replacement for the standard bent normal technique [Rev11]. Our method yields more realistic results with a 50% increase in computation time of the previous technique, using the same high dynamic range (HDR) preintegrated environment image. We use several environment samples positioned along the major axis of the specular microfacet BRDF. We derive an analytic formula to determine the two closest and two farthest points from the reflected direction on an approximation of the BRDF confidence region boundary. The two farthest points define the BRDF major axis, while the two closest points are used to approximate the BRDF width. The environment level of detail is derived from the BRDF width and the distance between the samples. We extensively compare our method with the bent normal technique and the ground truth using the GGX specular BRDF.

3D printer fabrication of boron doped Cu-MWCNT/PLA electrodes: Investigation of its efficiency in electrocatalytic hydrogen production

In this study, three-dimensional (3D) electrodes based on MWCNT-Cu/PLA with high electrocatalytic efficiency, large surface area and different amounts of nano‑boron doped MWCNT-Cu/PLA were fabricated using Fused Granular Fabrication (FGF) method with a 3D printer. When FE-SEM images are examined, MWCNT fibers and copper particles are seen more clearly on the surface due to the formation of more dense conductive networks between the conductive carbon fibers trapped in the PLA structure due to the increasing amount of boron. When XRD results are examined, it is observed that the characteristic peaks belonging to the face-centered cubic structure of Cu are dominant and the crystal size increases as MWCNT addition and boron doping increase. When Raman spectra were analyzed to investigate the internal structural changes of MWCNTs, three characteristic bands corresponding to the D band (defect), G band (graphite band) and G' band (D overtone) of MWCNTs were determined and it was observed that the ID/IG ratio decreased with increasing boron doping compared to MWCNT-Cu/PLA electrode. The B1s spectrum of the 3D 0.8B@MWCNT-Cu/PLA electrode confirmed the presence of boron with the peak corresponding to B@C bonds at 191.2 eV. The obtained 3D electrocatalysts were used as cathodes in an electrolysis cell in aqueous solution containing 1 M KOH and their catalytic efficiency in hydrogen evolution reaction (HER) was tested. In order to increase the activity of the 3D electrodes, unlike the literature, they were activated by electrochemical activation process without using organic solvent and HER measurements of the activated electrodes were performed. Before activation, the charge transfer resistance (Rct) value of the Cu/PLA electrode, which was 6219.00 Ω cm−2, decreased to 681.90 Ω cm−2 after 5 % MWCNT doping. It is seen that with boron doping, a greater decrease in Rct value is realized, reaching values of 66.50–74.56 Ω cm−2. The cathodic Tafel slopes show that the Volmer reaction is the rate-determining step for HER and the rate-determining step is the electrochemical adsorption of hydrogen on the metal surface. In addition, energy efficiency tests of the 3D electrodes in HER were also performed and electrochemically active surface areas were determined. Thus, this study has taken an important step toward using 3D electrocatalysts, which are produced with more cost and more flexible designs, unlike previous production techniques, for hydrogen energy production.