Transformation Technique Research Articles

Deep Learning-Driven Parameter Adaptation for Underwater Image Restoration

In this paper we propose a learning-based approach to enhance underwater image quality by optimizing parameters and applying intensity transformations. Our methodology involves training a CNN Regression model on diverse underwater images to learn enhancing parameters, followed by applying intensity transformation techniques. In order to evaluate our approach, we conducted experiments using well-known underwater image datasets found in the literature, comprising real-world subaquatic images and we propose a novel underwater image dataset, composed by 276 images from Amazon turbid water rivers. The results demonstrate that our approach achieves an impressive accuracy rate in three different underwater image datasets. This high level of accuracy showcases the robustness and efficiency of our proposed method in restoring underwater images.

Integrating lipid metabolite analysis with MRI-based transformer and radiomics for early and late stage prediction of oral squamous cell carcinoma

BackgroundOral Squamous Cell Carcinoma (OSCC) presents significant diagnostic challenges in its early and late stages. This study aims to utilize preoperative MRI and biochemical indicators of OSCC patients to predict the stage of tumors.MethodsThis study involved 198 patients from two medical centers. A detailed analysis of contrast-enhanced T1-weighted (ceT1W) and T2-weighted (T2W) MRI were conducted, integrating these with biochemical indicators for a comprehensive evaluation. Initially, 42 clinical biochemical indicators were selected for consideration. Through univariate analysis and multivariate analysis, only those indicators with p-values less than 0.05 were retained for model development. To extract imaging features, machine learning algorithms in conjunction with Vision Transformer (ViT) techniques were utilized. These features were integrated with biochemical indicators for predictive modeling. The performance of model was evaluated using the Receiver Operating Characteristic (ROC) curve.ResultsAfter rigorously screening biochemical indicators, four key markers were selected for the model: cholesterol, triglyceride, very low-density lipoprotein cholesterol and chloride. The model, developed using radiomics and deep learning for feature extraction from ceT1W and T2W images, showed a lower Area Under the Curve (AUC) of 0.85 in the validation cohort when using these imaging modalities alone. However, integrating these biochemical indicators improved the model’s performance, increasing the validation cohort AUC to 0.87.ConclusionIn this study, the performance of the model significantly improved following multimodal fusion, outperforming the single-modality approach.Clinical relevance statementThis integration of radiomics, ViT models, and lipid metabolite analysis, presents a promising non-invasive technique for predicting the staging of OSCC.

Machine Learning-Powered Earthquake Early Warning System

The most devastating natural disasters on earth are earthquakes that causes long-term effects on geography, civilization, and human life. These unpredictable events pose a serious threat to infrastructure. Furthermore, the current Earthquake Early Warning (EEW) systems are facing issues such as limited warning times, false alarms, maintenance costs, high construction costs, and data interpretation. Highlighting these as an urgent need for mitigation measures, there is a need to improve the effectiveness of electronic alerts and public safety measures. For this transformative machine learning techniques and the integration of disparate data, can embark on creating social security and lives protecting from major environmental disasters like earthquakes. This paper has compared various Machine Learning (ML) techniques by training them by using two datasets: one from India and another from India United States Geological from Research World Database to improve the robustness and generality of the earthquake prediction model in the Earthquake Early Warning (EEW) framework. This represents a major advance for earthquake detection and promises to reduce response time. Among various ML Techniques, Random Forest has performed well in earthquake warning with 96.06% accuracy and 98.6% precision.

CFIFusion: Dual‐Branch Complementary Feature Injection Network for Medical Image Fusion

ABSTRACTThe goal of fusing medical images is to integrate the diverse information that multimodal medical images hold. However, the challenges lie in the limitations of imaging sensors and the issue of incomplete modal information retention, which make it difficult to produce images encompassing both functional and anatomical information. To overcome these obstacles, several medical image fusion techniques based on CNN or transformer architectures have been presented. Nevertheless, CNN technique struggles to establish extensive dependencies between the fused and source images, and transformer architecture often overlooks shallow complementary features. To augment both the feature extraction capacity and the stability of the model, we introduce a framework, called dual‐branch complementary feature injection fusion (CFIFusion) technique, a for multimodal medical image fusion framework that combines unsupervised models of CNN model and transformer techniques. Specifically, in our framework, the entire source image and segmented source image are input into an adaptive backbone network to learn global and local features, respectively. To further retain the source images' complementary information, we design a multi‐scale complementary feature extraction framework as an auxiliary module, focusing on calculating feature differences at each level to capture the shallow complementary information. Then, we design a shallow information preservation module tailored for sliced image characteristics. Experimental results on the Harvard whole brain atlas dataset demonstrate that CFIFusion shows greater benefits than recent state‐of‐the‐art algorithms in terms of both subjective and objective evaluations.

Digital Transformation Development Strategies to Improve Sustainable Performance

This article examines the intersection of virtual transformation and sustainable overall performance, highlighting strategies corporations can implement to enhance their operational effectiveness while selling sustainability. It begins with a complete know-how of digital transformation, defining its key additives and the technological improvements riding this change. The significance of sustainable performance is mentioned, emphasizing its growing relevance in these days’s commercial enterprise landscape, wherein stakeholders an increasing number of call for environmentally and socially accountable practices. The article explores how integrating digital transformation initiatives with sustainability dreams can lead to synergistic advantages, growing an aggressive benefit for companies. Implementation techniques for virtual transformation are mentioned, focusing on frameworks that facilitate the adoption of innovative technologies at the same time as ensuring alignment with sustainable overall performance targets. Case research and exemplary practices from various industries illustrate successful digital transformation efforts that have undoubtedly impacted sustainability outcomes. In conclusion, the item offers insights into destiny guidelines for studies and practice, urging groups to embrace virtual transformation no longer merely as a technological improvement but as a holistic technique to attaining long-term sustainability. By exploring these issues, this examines objectives to provide a valuable aid for practitioners and researchers looking to navigate the complexities of digital transformation within the context of sustainable overall performance. Keywords : Digital Transformation, Sustainable Performance, Development Strategies

Heat Transfer Management of Thermal and Electronics Systems using Convective-Radiative Porous Fins with Thermal Contact Resistance and Diabatic Tip

The miniaturization of electronics, electrical, thermal, mechanical and optical devices and systems calls for increasing demand for efficient cooling systems with higher thermal efficiency. The use of passive mode of cooling using fins has provided unprecedented results. However, the previous transient analyses of the extended surfaces have been done without proper considerations of the imperfect contact between the fin base and the prime surface. Also, the tip of the passive device has been assumed to be adiabatic. However, the present article explores the significances of thermal contact resistance and diabatic tip on the transient thermal response of straight fin with temperature-dependent thermal conductivity and magnetic field under radiative-convective conditions. The nonlinear model is analyzed numerically using generalized Integral Transform Techniques and the numerical results are verified by the exact solution for the linear thermal model. The parametric explorations reveal that the adimensional local temperature in the conductive-radiative fin increases when the conductive-convective, conductive-radiative and magnetic field parameters increase. However, under the assumption of perfect thermal contact between the prime surface and the base of the fin, the adimensional temperature in the fin will decrease as the conductive-convective, conductive-radiative and magnetic field parameters increase. The fin temperature is significantly affected by the Biot numbers under the comparably low values of the conductive-convective, conductive-radiative, magnetic field and thermal conductivity parameters. The effects of the fin thermal conductivity parameter along the fin length depend on the respective values of thermal contact Biot number at the base of the fin and the end cooling Biot number at the tip of the fin. When the thermal conductivity parameter is amplified, the fin adimensional temperature increases when thermal contact Biot number at the base of the fin is zero and the end cooling Biot number at the tip of the fin is very large. This study will help in accurate analysis and design of heat sinks and passive devices for heat transfer enhancement for thermal and electronic systems.

Plant Molecular Farming: An Emerging Opportunity For Biopharmaceuticals

Conventional drug production methods are costly. Now, it has been proved that plants are potentially a new source of pharmaceutical proteins, including vaccines, antibodies, blood substitutes, and other therapeutic entities. Unlike mammalian-derived rDNA drugs, plant-derived antibodies, vaccines, and other proteins are particularly advantageous since they are free of mammalian viral vectors and human pathogens. Plants made therapeutics are cheaper, safer, and can be abundantly produced and easily stored. Recombinant proteins and other metabolites are produced in transgenic plants for industrial or pharmaceutical purposes, which are known as molecular farming. Transgenic plants carry one or more foreign genes transferred through the techniques of transformation. Although initially these were produced only in a limited number of plant species (e.g., tobacco, petunia, tomato, etc.), later these could be produced in any plant species, including both dicots and monocots. Transgenic plants resistant to herbicides, insects, viruses, and a host of abiotic stresses have already been produced. These plants have also been produced for improved nutritional quality and are suitable for food processing. The aim of this review paper is to understand plant molecular farming, the advantages and limitations of the process, and biosafety concerns.

Transient behaviors in a spectrum-tailored all-PM NALM mode-locked fiber laser.

We demonstrate an all-polarization-maintaining (PM) dispersion-managed mode-locked ytterbium-doped fiber oscillator based on a nonlinear amplifying loop mirror (NALM). Experimentally, three mode-locking regimes with distinct tailored spectra are achieved with carefully designed cavity setup and adjusted pump power. Using the dispersive Fourier transformation (DFT) technique, the real-time transition dynamics among these regimes are observed as the pump power decreases. It is revealed that the intracavity gain does not decrease continuously during the relaxation period of the gain fiber. Instead, it decreases first and then increases, ultimately leading to the overdriving of NALM. This is attributed to the increase in the population inversion rate of gain fiber under the remaining pump. The dynamic evolution of intracavity gain and non-monotonic transmission properties of NALM collectively induce chaotic pulsation during the transformation. This work not only provides a new perspective for the design and development of novel spectrum-tailored laser sources, but also deepens the understanding of the transient dynamics of ultrafast fiber lasers.

Mass spectrometry imaging of N-linked glycans: Fundamentals and recent advances.

With implications in several medical conditions, N-linked glycosylation is one of the most important posttranslation modifications present in all living organisms. Due to their nontemplate synthesis, glycan structures are extraordinarily complex and require multiple analytical techniques for complete structural elucidation. Mass spectrometry is the most common way to investigate N-linked glycans; however, with techniques such as liquid-chromatography mass spectrometry, there is complete loss of spatial information. Mass spectrometry imaging is a transformative analytical technique that can visualize the spatial distribution of ions within a biological sample and has been shown to be a powerful tool to investigate N-linked glycosylation. This review covers the fundamentals of mass spectrometry imaging and N-linked glycosylation and highlights important findings of recent key studies aimed at expanding and improving the glycomics imaging field.

Sampling and active learning methods for network reliability estimation using K-terminal spanning tree

Network reliability analysis remains a challenge due to the increasing size and complexity of networks. This paper presents a novel sampling method and an active learning method for efficient and accurate network reliability estimation under node failure and edge failure scenarios. The proposed sampling method adopts Monte Carlo technique to sample component lifetimes and the K-terminal spanning tree algorithm to accelerate structure function computation. Unlike existing methods that compute only one structure function value per sample, our method generates multiple component state vectors and corresponding structure function values from each sample. Network reliability is estimated based on survival signatures derived from these values. A transformation technique extends this method to handle both node failure and edge failure. To enhance efficiency of proposed sampling method and achieve adaptability to network topology changes, we introduce an active learning method utilizing a random forest (RF) classifier. This classifier directly predicts structure function values, integrates network behaviors across diverse topologies, and undergoes iterative refinement to enhance predictive accuracy. Importantly, the trained RF classifier can directly predict reliability for variant networks, a capability beyond the sampling method alone. Through investigating several network examples and two practical applications, the effectiveness of both proposed methods is demonstrated.

Pulsating dynamics of noise-like pulses in a fiber laser with nonlinear optical loop mirror

We experimentally investigate the nature of pulsating noise-like pulses (NLPs) in a fiber laser based on the nonlinear optical loop mirror. By adjusting the intra-cavity polarization, three types of pulsating NLPs can be obtained in the cavity. By utilizing the dispersive Fourier transformation technique, the real-time evolution dynamics of NLP pulsation have been investigated in detail. Different from the conventional pulsating behavior, the NLPs undergo remarkable and periodical variations in their width with slight changes in pulse peak powers during pulsation process. We speculate that the wavelength-dependent gain saturation is involved in the pulsating NLP evolution. Quasi-periodic energy oscillations are associated with cyclic generation and subsequent walkoff of wavelength-shifted components, resulting in the different pulsating dynamics of NLP. Moreover, during pulsation process, the NLP splitting could happen with the increasing of energy. All these findings will help to complement our understanding of NLP pulsation in a fiber laser.

Design of a Technology-Based Magic Show System with Virtual User Interfacing to Enhance the Entertainment Effects

The integration of interactive technology into magic performances was explored in this study, with a focus on leveraging virtual user interfacing and interactive video projection techniques to enhance the entertainment effects. A thorough literature review identified transformation techniques between virtual and real forms in the magic performance, along with various digital magic effects. Design principles derived from the review were applied in constructing a magic show system named “FUI Magic”, where FUI stands for fantasy user interface. The system is based on virtual user interfacing, implemented via laser range-finding and video projection techniques. The “FUI Magic” system facilitated the development and presentation of a digital-multimedia magic show titled “Fantasy Doves”, publicly showcased in an exhibition. Statistical evaluation of the feedback from expert interviews and a questionnaire survey revealed positive audience impressions and confirmed the effectiveness of incorporating virtual user-interfacing technology for captivating entertainment. This study affirms the significance of visual design through virtual user interfacing in enhancing technological ambiance and magic effects, suggesting its practicality for further exploration in diverse applications.

Boosting transferability of adversarial samples via saliency distribution and frequency domain enhancement

Deep neural networks (DNNs) perform excellently in various vision tasks, however, they are susceptible to the adversarial samples. These samples are crafted by injecting subtle perturbations into benign images that are barely detectable to humans. Due to the differences between the substitute model and target model, the efficacy of black-box attack still yield suboptimal results. To address this issue, we propose a black-box attack approach, SD-FI2M, based on the momentum iterative fast gradient sign method, which can craft potent adversarial samples by inducing diverse inputs in the frequency domain and the saliency distribution of images. To enhance the diversity of input samples in frequency domain, a spectrum transformation technique utilizing the discrete cosine transform was used to craft adversarial samples with higher transferability. Further, Guided Grad-CAM was employed to obtain the saliency distribution of benign images, reducing indiscriminate damage to image features and alleviating the overfitting problem in adversarial samples. Extensive experiments on ImageNet have verified the superior attack performance and transferability of the proposed method.

Identification of the dynamic model of second-life lithium-ion cells through Subspace System Identification and similarity transformation

When lithium-ion batteries (LiBs) reach the end of their first useful life in electric vehicles (EVs), they can still be used in applications with lower power demands, a process known as second-life. However, to ensure that LiBs – or cells – removed from EVs operate safely, efficiently, and reliably in a second application, various tests and procedures must be applied to study their internal conditions. Some information about the conditions of the lithium-ion cells can be obtained from the parameters of its equivalent circuit model(ECM). However, the tests to obtain blackthe ECM parameters usually require taking the LiBs out of service and testing their cells in laboratory benches with special equipment, which is costly and time-consuming. Therefore, this work proposes a methodology to identify the ECM of lithium-ion cells based on Subspace System Identification (SSI) methods. This procedure can be executed online, while the LiBs are being used in the specified application. SSI methods have been widely adopted to identify the dynamic model of lithium-ion cells, but they return models without a physical meaning, called non-parametric models. This type of model cannot be used to address the internal conditions of a lithium cell, because its parameters do not have any physical meaning. To solve this problem, a novel method for estimating the lithium-ion cell physical parameters, in the ECM format, and from its non-parametric model, is proposed in this work. The procedure is based on similarity transformation techniques and special characteristics of the ECM. To validate the proposed method, nine samples of cells from the same manufacturer were studied. These cells had distinct degradation conditions and were removed from heavy-duty EVs at the end of their first life. The results showed that: (a) a good approximation between the identified model voltage output and the actual cells’ experimental voltage output was achieved, with root mean square error values as small as 2.26 mV; (b) it was possible to identify the ECM of the tested cells and their parameters using the proposed similarity transformation; and (c) the identified parameters can be used to detect the internal conditions of the cells.

Gyro-TWT Frequency Multiplier

Gyrotron-type amplifiers are being developed for various applications requiring high power in the millimeter wavelength range. Calculations of the optimal designs of gyrotron traveling wave tubes (gyro-TWTs), presented in the article, were carried out using the Gyro-K computer program using a coordinate transformation technique, which made it possible to significantly reduce the calculation time of gyroresonance devices by reducing the three-dimensional problem of excitation of an irregular waveguide to a one-dimensional one. Two options for creating a gyro-TWT using the TE02 wave mode are considered: the first one is operating at the first harmonic of the gyrofrequency, the second is a frequency multiplier. The achievable characteristics of devices in the terahertz frequency range are presented, such as gain bands, efficiency, gain factors and distributions of high-frequency fields in longitudinal and transverse sections. The gyro-TWT frequency multiplier has a gain bandwidth of 7.2 %, an efficiency of 17 %, and a gain of 30 dB.

Enhanced daily streamflow forecasting in Northeastern Algeria: integrating hybrid machine learning with advanced wavelet transformation techniques

Enhanced daily streamflow forecasting in Northeastern Algeria: integrating hybrid machine learning with advanced wavelet transformation techniques

Investigation of momentum and heat transfer characteristics of Maxwell fluid cascading over a stretching sheet: A HAM approach

This paper presents a novel investigation into the intricate behaviour of momentum and heat transport phenomena in a non-Newtonian Maxwell fluid flowing over a stretching sheet. Incorporating thermal radiation Rd, magnetic fields M, buoyancy effects λT, and porous media K under convective boundary conditions the study unveils complex fluid behaviours. Energy equation has been obtained by incorporating non-uniform heat source/sink along with viscosity of the fluid as a function of temperature across the domain. Leveraging the Lie Scale transformation technique, the governing non-linear partial differential equations are converted into non-linear ordinary differential equations. With the aid of Homotopy Analysis Method (HAM), a semi-analytical technique, the solutions describing the physical phenomenon of the current model have been obtained. Further, the results are assessed through the graphical analysis of the velocity profile f′η, thermal profile θη, skin friction coefficient CfRe12, and Nusselt number NuRe−12. The obtained results using HAM shows good agreement with the existing literature. The present work offers practical implications for various engineering applications.

An Exploration of the Phonon Frequency Spectrum and Born–von Karman Periodic Boundary Conditions in 1D and 2D Lattice Systems Using a Computational Approach

Periodic boundary conditions (PBCs) are a pivotal concept in the treatment of ideal lattices of infinite extent as a finite lattice. Most undergraduate texts that delve into the analytical treatment of lattice dynamics solve the problem in the reciprocal space with an implicit assumption of PBCs. While the traditional approach centered in the reciprocal or [Formula: see text]-space is elegant for simple one-dimensional lattice structures, the method becomes mathematically cumbersome for more complex lattices with real-world applications, such as a honeycomb lattice involving a [Formula: see text] secular determinant. In this work, we have explored the solution of lattice dynamics in direct space numerically, by constructing unit cells through an explicit implementation of Born–von Karman PBCs and solving the lattice dynamical equations in the displacement–time domain using the fourth-order Runge–Kutta algorithm. The Fast Fourier Transform technique is then used to obtain the phonon frequency spectrum corresponding to the computed instantaneous displacements. The fidelity of the model is validated by the agreement of the computational results obtained for the phonon spectrum at high symmetry points in the Brillouin zone with the analytical values for the given problem. Initially, the dynamics and the phonon spectrum are explored for monatomic and diatomic linear lattices which serve to act as introductory cases for the approach before investigating the 2D monatomic square and honeycomb lattices using the nearest and next-nearest neighbor approximations. A short-range force constant model employing the central forces and angular forces is used for the monatomic square lattice to account for the interatomic interactions. Our computational approach is expected to be physically more intuitive for a student for visualizing the periodicity of a given lattice and understanding its related dynamics. The approach demonstrates the usage of Born–von Karman PBCs in constructing a unit cell that effectively captures the dynamics of a lattice system, complementing the traditional secular determinant formalism. The work also illustrates an instance of how the computer programming can be used as a tool to empower the students to explore abstract concepts in physics. A honing of programming and computational skills of students enables them to interact with dynamic models in physics and gain deeper insight into the real-world applications of physics principles. Target group of this work is the students and educators at undergraduate level who seek a thorough and didactic comprehension of lattice dynamics.

Thermal stresses associated with a thermosensitive multilayered disc analysed due to point heating

In this paper, analytical solutions are presented for temperature and thermal behavior of a thermosensitive multilayered annular disc due to point heat source. Convective heating is applied to both the innermost and outermost layers. The nonlinearity of the thermal diffusivity equation is dealt using Kirchhoff’s transformation technique. A finite integral transform in the form of Bessel’s function is used to deal with the radial variable r. Fourier transform and angular eigen functions are also used to solve the thermal diffusivity equation. Deflection, resultant forces, shearing forces, resultant moments and thermal stresses are obtained. A mathematical representation is formulated for a 3-layered disc, with the inner, middle and outer layers composed of copper, zinc and aluminum respectively. The results are depicted graphically.

Segmentation of retinal microaneurysms in fluorescein fundus angiography images by a novel three-step model.

Microaneurysms serve as early signs of diabetic retinopathy, and their accurate detection is critical for effective treatment. Due to their low contrast and similarity to retinal vessels, distinguishing microaneurysms from background noise and retinal vessels in fluorescein fundus angiography (FFA) images poses a significant challenge. We present a model for automatic detection of microaneurysms. FFA images were pre-processed using Top-hat transformation, Gray-stretching, and Gaussian filter techniques to eliminate noise. The candidate microaneurysms were coarsely segmented using an improved matched filter algorithm. Real microaneurysms were segmented by a morphological strategy. To evaluate the segmentation performance, our proposed model was compared against other models, including Otsu's method, Region Growing, Global Threshold, Matched Filter, Fuzzy c-means, and K-means, using both self-constructed and publicly available datasets. Performance metrics such as accuracy, sensitivity, specificity, positive predictive value, and intersection-over-union were calculated. The proposed model outperforms other models in terms of accuracy, sensitivity, specificity, positive predictive value, and intersection-over-union. The segmentation results obtained with our model closely align with benchmark standard. Our model demonstrates significant advantages for microaneurysm segmentation in FFA images and holds promise for clinical application in the diagnosis of diabetic retinopathy. The proposed model offers a robust and accurate approach to microaneurysm detection, outperforming existing methods and demonstrating potential for clinical application in the effective treatment of diabetic retinopathy.