Lightweight Hybrid Research Articles

The Effect of Ice Floe on the Strength, Stability, and Fatigue of Hybrid Flexible Risers in the Arctic Sea

Flexible risers have proven to be a popular choice for deepwater exploration due to their ability to withstand functional and environmental stress while maintaining system integrity. In the challenging arctic conditions, lightweight hybrid composite flexible risers are likely to be employed to mitigate the increase in effective tension. This study investigates the strength and stability performance of production hybrid composite flexible risers with composite pressure armour in the harsh environmental conditions of the Arctic seas. At a water depth of 340 m, the flexible riser was analysed in various global configurations to evaluate the static, dynamic, and lamina-scale performance of its carbon fibre-reinforced thermoplastic polymer composite layer. The drifting ice in the region generated additional load on the riser system, and the effects of this ice on the riser design and its dynamic and lamina-level performances were also analysed. The results indicate that the current riser design incorporating the composite layer is insufficient to ensure system integrity without mitigating the effects of ice loading. The carbon fibre direction in each lamina must be optimised for excess axial stress emanating from the combined action of hoop, axial, and bending stresses. Finally, recommendations on how to improve the life of the lightweight hybrid composite riser in arctic conditions are provided.

Combining CNNs and Markov-like Models for Facial Landmark Detection with Spatial Consistency Estimates.

The accurate localization of facial landmarks is essential for several tasks, including face recognition, head pose estimation, facial region extraction, and emotion detection. Although the number of required landmarks is task-specific, models are typically trained on all available landmarks in the datasets, limiting efficiency. Furthermore, model performance is strongly influenced by scale-dependent local appearance information around landmarks and the global shape information generated by them. To account for this, we propose a lightweight hybrid model for facial landmark detection designed specifically for pupil region extraction. Our design combines a convolutional neural network (CNN) with a Markov random field (MRF)-like process trained on only 17 carefully selected landmarks. The advantage of our model is the ability to run different image scales on the same convolutional layers, resulting in a significant reduction in model size. In addition, we employ an approximation of the MRF that is run on a subset of landmarks to validate the spatial consistency of the generated shape. This validation process is performed against a learned conditional distribution, expressing the location of one landmark relative to its neighbor. Experimental results on popular facial landmark localization datasets such as 300 w, WFLW, and HELEN demonstrate the accuracy of our proposed model. Furthermore, our model achieves state-of-the-art performance on a well-defined robustness metric. In conclusion, the results demonstrate the ability of our lightweight model to filter out spatially inconsistent predictions, even with significantly fewer training landmarks.

Efficient diagnosis of hematologic malignancies using bone marrow microscopic images: A method based on MultiPathGAN and MobileViTv2

Background and objectivesHematologic malignancies, including the associated multiple subtypes, are critically threatening to human health. The timely detection of malignancies is crucial for their effective treatment. In this regard, the examination of bone marrow smears constitutes a crucial step. Nonetheless, the conventional approach to cell identification and enumeration is laborious and time-intensive. Therefore, the present study aimed to develop a method for the efficient diagnosis of these malignancies directly from bone marrow microscopic images. MethodsA deep learning-based framework was developed to facilitate the diagnosis of common hematologic malignancies. First, a total of 2033 microscopic images of bone marrow analysis, including the images for 6 disease types and 1 healthy control, were collected from two Chinese medical websites. Next, the collected images were classified into the training, validation, and test datasets in the ratio of 7:1:2. Subsequently, a method of stain normalization to multi-domains (stain domain augmentation) based on the MultiPathGAN model was developed to equalize the stain styles and expand the image datasets. Afterward, a lightweight hybrid model named MobileViTv2, which integrates the strengths of both CNNs and ViTs, was developed for disease classification. The resulting model was trained and utilized to diagnose patients based on multiple microscopic images of their bone marrow smears, obtained from a cohort of 61 individuals. ResultsMobileViTv2 exhibited an average accuracy of 94.28% when applied to the test set, with multiple myeloma, acute lymphocytic leukemia, and lymphoma revealed as the three diseases diagnosed with the highest accuracy values of 98%, 96%, and 96%, respectively. Regarding patient-level prediction, the average accuracy of MobileViTv2 was 96.72%. This model outperformed both CNN and ViT models in terms of accuracy, despite utilizing only 9.8 million parameters. When applied to two public datasets, MobileViTv2 exhibited accuracy values of 99.75% and 99.72%, respectively, and outperformed previous methods. ConclusionsThe proposed framework could be applied directly to bone marrow microscopic images with different stain styles to efficiently establish the diagnosis of common hematologic malignancies.

Strain rate dependency of lightweight hybrid green composite laminate-an experimental study

Strain rate dependency of lightweight hybrid green composite laminate-an experimental study

A hybrid lightweight encoder-decoder network for automatic bridge crack assessment with real-world interference

Although many learning-based studies have been conducted to detect cracks, there are still many problems in practice, such as slow inference speed due to a large number of hyperparameters required in network architectures and compromised detection accuracy in different environments. To address these issues, the current study employed a Hybrid Lightweight Encoder-Decoder Network (HLEDNet) as an ad-hoc crack segmentation and measurement system on real-world images captured from various concrete bridges. The proposed HLEDNet model was trained and tested with 3000 annotated images with further extensive data augmentation, which achieved 86.92%, 85.71%, 86.31, and 86.01% in precision, recall, F1 score, and mean intersection over union (mIoU), respectively. A crack measurement module was proposed using combined postprocessing techniques, where the R-squared values of the regression lines in crack length and average crack width are 0.9857 and 0.9925, respectively. Finally, an experimental study was undertaken to convert the crack measuring unit from pixel to millimetre.

An Adaptive Lightweight Hybrid Encryption Scheme for Securing the Healthcare Data in Cloud-Assisted Internet of Things

An Adaptive Lightweight Hybrid Encryption Scheme for Securing the Healthcare Data in Cloud-Assisted Internet of Things

Hybrid Attention Asynchronous Cascade Network for Salient Object Detection

The highlighted area or object is defined as the salient region or salient object. For salient object detection, the main challenges are still the clarity of the boundary information of the salient object and the positioning accuracy of the salient object in the complex background, such as noise and occlusion. To remedy these issues, it is proposed that the asynchronous cascade saliency detection algorithm based on a deep network, which is embedded in an encoder–decoder architecture. Moreover, the lightweight hybrid attention module is designed to obtain the explicit boundaries of salient regions. In order to effectively improve location information of salient objects, this paper adopts a bi-directional asynchronous cascade fusion strategy, which generates prediction maps with higher accuracy. The experimental results on five benchmark datasets show that the proposed network HACNet is on a par with the state of the art for image saliency datasets.

Development of light weight sustainable pineapple/kevlar hybridized fiber and peanut husk cellulose toughened vinyl ester biocomposite for unmanned aerial vehicle applications

AbstractThe aim of this study is to develop a light weight hybrid biocomposite using pineapple and Kevlar fiber with peanut husk cellulose in vinyl ester resin for applications in unmanned aerial vehicles. This study focuses on how the silane treatment on fiber and cellulose particle influences the mechanical, fatigue and low velocity impact properties of this hybrid biocomposites. Using hand lay‐up technique, the biocomposite was prepared with cellulose loading ranging from 1 to 5 vol%. The results revealed that the 5 vol% of cellulose added composite had an improved tensile, impact, flexural, hardness and ILSS of 161 MPa, 224 MPa, 6.8 J, 84 shore‐D and 21.4 MPa. Moreover, the biocomposite with the inclusion of 3 vol% cellulose had an improved fatigue life count of 42 697, 29 821, 22 381 and 18 164 at 25%, 50%, 75% and 90% of UTS. Similarly, the 3 vol% cellulose reinforced composite showed an improved low velocity impact toughness of 12.36 J. The obtained results clearly indicated that these mechanically strengthened and highly toughened biocomposites could be used as working material for number of applications, especially in making of UAVs for the aerospace industry, automotive components for the transportation sector and structural material in domestic infrastructure.

Highly boosted homogeneity of polymer matrix composites filled with MXene-derived 2D titanium oxide nanosheet for high-k gate dielectrics

Polymer-based multilayer films with solution processability and specifically tailored material properties are promising candidates as gate dielectrics in functional electronic devices. The dielectric performance of polymer composites that are formed by incorporating nano-scale ceramic or conductive fillers in a polymer matrix can drastically increase dielectric constants. However, producing these polymer composite dielectrics can be challenging because the dispersion homogeneity of fillers and interfacial adhesion between the filler and the matrix are difficult to control. Therefore, developing an adequately aligned structure of incorporated fillers inside the polymer matrix is strongly demanded. Herein, to achieve concurrent improvements in the dielectric constant and dielectric loss, a rational design process for a multilayer-structured gate dielectric layer based on conductive Ti3C2Tx MXene, ceramic TiO2 and poly(vinyl alcohol) (PVA) is proposed. We present a high-k, pinhole-free, and PVA-based nanocomposite dielectric obtained by assembling two-dimensional MXene into a highly aligned three-dimensional rutile TiO2 aerogel. The homogeneous hybrid nanosheet-dispersed polymeric nanocomposite, which attains an exceptionally high dielectric constant of 66.8 (more than one order of magnitude higher than that of pristine PVA), exhibits a relatively low loss tangent, as low as 10−1. Our approach paves the path for developing high-performance Au/PVA-2D TiO2-MXene/p–Si capacitors comprising nanostructured gate dielectrics fabricated using lightweight hybrid aerogel template-assistance.

Hybrid CMOS-RRAM True Random Number Generator Exploiting Coupled Entropy Sources

Compact and reliable ON-chip true random number generators (TRNGs) are inevitable for generating secure cryptographic keys in resource constrained mobile Internet-of-Things (IoT) devices. To this end, in this work, we propose an extremely area-efficient hybrid CMOS-resistive random access memory (RRAM) TRNG exploiting coupled entropy sources: (a) temporal (cycle-to-cycle) and spatial (device-to-device) switching variability of filamentary RRAMs and (b) race condition of the set reset (SR)-latch. Such a coupled entropy source enhances the stochasticity of the generated random bits and eliminates the need for complex and energy-hungry postprocessing circuits. The raw output bits generated utilizing the proposed hybrid CMOS-RRAM TRNG passes all the standard statistical tests of the NIST SP800-22 test suite even in the presence of variations in the operating voltage and temperature. Our work may provide incentive for experimental realization of such lightweight hybrid CMOS-nonvolatile memories (NVM)-based TRNGs harnessing coupled entropy sources.

A resource efficient pseudo random number generator based on sawtooth maps for Internet of Things

AbstractThe strength of cryptographic keys rely on the random number generators (RNGs) to produce random seed values. Unfortunately there are not many RNGs options suitable for Internet of Things (IoTs) scenario, due to limited processing resources and bulk quantity of IoT data that needs to be secured. In this article, we studied sawtooth map which is a chaotic map. However, when implemented on a computer, the sawtooth map results on a non‐chaotic orbit due to the finite precision of computation. This can be avoided if we use the sawtooth map as the local map in a coupled map lattice (CML) system. We explore such coupled map systems for randomness through entropy and statistical analysis. Based on the results, we propose a lightweight hybrid pseudo random number generator (PRNG) based on sawtooth based CML system and SPONGENT hashing. The proposed PRNG is thoroughly tested against statistical attacks, entropy analysis, key space analysis and compared with existing state of the art solutions. The results provide evidence that the proposed PRNG produces random numbers that could produce sufficiently strong cryptographic keys for resource constrained IoT devices.

HTDet: A Hybrid Transformer-Based Approach for Underwater Small Object Detection

As marine observation technology develops rapidly, underwater optical image object detection is beginning to occupy an important role in many tasks, such as naval coastal defense tasks, aquaculture, etc. However, in the complex marine environment, the images captured by an optical imaging system are usually severely degraded. Therefore, how to detect objects accurately and quickly under such conditions is a critical problem that needs to be solved. In this manuscript, a novel framework for underwater object detection based on a hybrid transformer network is proposed. First, a lightweight hybrid transformer-based network is presented that can extract global contextual information. Second, a fine-grained feature pyramid network is used to overcome the issues of feeble signal disappearance. Third, the test-time-augmentation method is applied for inference without introducing additional parameters. Extensive experiments have shown that the approach we have proposed is able to detect feeble and small objects in an efficient and effective way. Furthermore, our model significantly outperforms the latest advanced detectors with respect to both the number of parameters and the mAP by a considerable margin. Specifically, our detector outperforms the baseline model by 6.3 points, and the model parameters are reduced by 28.5 M.

A hybrid modified lightweight algorithm for achieving data integrity and confidentiality

<span lang="EN-US">Encryption algorithms aim to make data secure enough to be decrypted by an attacker. This paper combines the Speck and the Salsa20 to make it difficult for an attacker to exploit any weaknesses in these two algorithms and create a new lightweight hybrid algorithm called Speck-Salsa20 algorithm for data integrity and confidentiality (SSDIC). SSDIC uses less energy and has an efficient throughput. It works well in both hardware and software and can handle a variety of explicit plaintext and key sizes. SSDIC solves the difficulties of the Speck algorithm. The sequence generated by Speck is not random and fails to meet an acceptable success rate when tested in statistical tests. It is processed by generating a random key using the Salsa20 algorithm. Salsa20 is a high-speed secure algorithm that is faster than advanced encryption standard (AES) and can be used on devices with low resources. It uses a 256-bit key hash function. The recovery of the right half of the original key of the Speck algorithm is also handled by modifying the Speck round function and the key schedule. Simulation results show, according to a National Institute of Standards and Technology (NIST) test, the performance achieved by the SSDIC is increased by nearly 66% more than that achieved from the Speck in terms of data integrity and confidentiality.</span>

Online Assembly Inspection Integrating Lightweight Hybrid Neural Network With Positioning Box Matching

Online Assembly Inspection Integrating Lightweight Hybrid Neural Network With Positioning Box Matching

Taguchi dizajn optimizacije zasnovane na eksperimentima i eksperimentalno ispitivanje mehaničkih performansi hibridnih epoksidnih nanokompozita

The rising demand of safety in the aerospace and automobile industry is continuously motivating researchers to develop high strength, lightweight hybrid polymer composites, usually consisting a combination of carbon nanotubes (CNTs) and ceramic nanoparticles in the epoxy matrix. However, the development of such composites are usually hindered by some existing challenges, such as optimization of the concentration of CNTs, nanoparticles and their distribution in viscous epoxy matrices. In order to make the most of the impressive mechanical characteristics of CNTs and SiO2 nanoparticles, ultrasonic dual mixing (UDM) technique was employed to develop MWCNT/SiO2 based hybrid epoxy nanocomposites (HENCs). A well-known approach, such as the Taguchi design of experiment, was used to optimize the concentration of MWCNT, SiO2 nanoparticles in epoxy and curing cycle of epoxy with respect to the tensile strength of the resulting HENCs. Additionally, the tensile strength, Young's Modulus, Strain to failure, and hardness were measured for HENCs. The results revealed that the optimal concentration of 1% MWCNT and 10% SiO2 leads to the maximum increase in tensile strength and other mechanical properties of the HENCs.

HRLE-SARDet: A Lightweight SAR Target Detection Algorithm Based on Hybrid Representation Learning Enhancement

In recent years, deep learning has been widely used in remote sensing, especially in the field of synthetic aperture radar (SAR) image target detection. However, all of these deep learning models continue increasing the network’s depth and width without maintaining a good balance between accuracy and speed. Therefore, in this paper, we propose a hybrid representation learning-enhanced SAR target detection algorithm based on the unique features of SAR images from a lightweight perspective called HRLE-SARDet. First, we design a lightweight and scattering feature extraction backbone that is more suitable for SAR image data. Second, for the multiscale feature discrepancy, we design a new multiscale feature fusion neck. Next, to better extract the scattering information from small targets of SAR images and improve the detection accuracy, we design a lightweight hybrid representation learning enhancement module. Finally, to better fit target detection for SAR image datasets, we redesign a more flexible loss function, which allows for an easy adjustment of the importance of polynomial bases according to the target task and dataset. Extensive experimental results on three SAR image ship target datasets (SSDD, AIR-SARShip-2.0, and HRSID) and a newly released large multiclass target SAR dataset (MSAR-1.0) show that our HRLE-SARDet achieves 98.4%, 79.2%, 92.5%, and 88.4% mAPs (mean Average Precision) with only 1.09 M parameters and 2.5 G floating-point operations (FLOPs) on the SSDD, AIR-SARShip-2.0, HRSID, and MSAR-1.0 datasets, which is an excellent performance.

LHDACT: Lightweight Hybrid Dual Attention CNN and Transformer Network for Remote Sensing Image Change Detection

LHDACT: Lightweight Hybrid Dual Attention CNN and Transformer Network for Remote Sensing Image Change Detection

A Lightweight Hybrid Convolutional Neural Network for Hyperspectral Image Classification

Recent studies have demonstrated the potential of hybrid convolutional models that combine 3D and 2D convolutional neural networks (CNNs) for hyperspectral image (HSI) classification. However, these models do not fully utilize the benefits of hybrid convolution due to inefficient connections between the two types of CNNs. Moreover, most CNNs, including hybrid models, require a significant number of parameters and computational resources for accurate classification, which increases the need for labeled samples and computational cost. Although the common lightweight strategies like depthwise separable convolution (DSC) can reduce parameters and computation compared to normal convolution (NC), they often compromise accuracy. To address these challenges, we propose a lightweight hybrid convolutional neural network (Lite-HCNet) for HSI classification with minimal model parameters and computational effort. Firstly, we design a novel channel attention module (NCAM) and combine it with a convolutional kernel decomposition (CKD) strategy to propose a lightweight and efficient DSC (LE-DSC) deployed in Lite-HCNet. The LE-DSC not only reduces the DSC volume further but also enhances its performance. Secondly, a lightweight and efficient hybrid convolutional layer (LE-HCL) is designed in Lite-HCNet to explore the efficient connection structure between 3D CNNs and 2D CNNs. Experiments show that the Lite-HCNet reduces the required computational cost and practical deployment difficulty while offering advanced performance with a small number of training samples. Furthermore, abundant ablation experiments confirm the superior performance of the designed LE-DSC.

ILSM: Incorporated Lightweight Security Model for Improving QOS in WSN

In the network field, Wireless Sensor Networks (WSN) contain prolonged attention due to afresh augmentations. Industries like health care, traffic, defense, and many more systems espoused the WSN. These networks contain tiny sensor nodes containing embedded processors, Tiny OS, memory, and power source. Sensor nodes are responsible for forwarding the data packets. To manage all these components, there is a need to select appropriate parameters which control the quality of service of WSN. Multiple sensor nodes are involved in transmitting vital information, and there is a need for secure and efficient routing to reach the quality of service. But due to the high cost of the network, WSN components have limited resources to manage the network. There is a need to design a lightweight solution that ensures the quality of service in WSN. In this given manner, this study provides the quality of services in a wireless sensor network with a security mechanism. An incorporated hybrid lightweight security model is designed in which random waypoint mobility (RWM) model and grey wolf optimization (GWO) is used to enhance service quality and maintain security with efficient routing. MATLAB version 16 and Network Stimulator 2.35 (NS2.35) are used in this research to evaluate the results. The overall cost factor is reduced at 60% without the optimization technique and 90.90% reduced by using the optimization technique, which is assessed by calculating the signal-to-noise ratio, overall energy nodes, and communication overhead.

Lightweight Deep Learning Models for High-Precision Rice Seedling Segmentation from UAV-Based Multispectral Images.

Accurate segmentation and detection of rice seedlings is essential for precision agriculture and high-yield cultivation. However, current methods suffer from high computational complexity and poor robustness to different rice varieties and densities. This article proposes 2 lightweight neural network architectures, LW-Segnet and LW-Unet, for high-precision rice seedling segmentation. The networks adopt an encoder-decoder structure with hybrid lightweight convolutions and spatial pyramid dilated convolutions, achieving accurate segmentation while reducing model parameters. Multispectral imagery acquired by unmanned aerial vehicle (UAV) was used to train and test the models covering 3 rice varieties and different planting densities. Experimental results demonstrate that the proposed LW-Segnet and LW-Unet models achieve higher F1-scores and intersection over union values for seedling detection and row segmentation across varieties, indicating improved segmentation accuracy. Furthermore, the models exhibit stable performance when handling different varieties and densities, showing strong robustness. In terms of efficiency, the networks have lower graphics processing unit memory usage, complexity, and parameters but faster inference speeds, reflecting higher computational efficiency. In particular, the fast speed of LW-Unet indicates potential for real-time applications. The study presents lightweight yet effective neural network architectures for agricultural tasks. By handling multiple rice varieties and densities with high accuracy, efficiency, and robustness, the models show promise for use in edge devices and UAVs to assist precision farming and crop management. The findings provide valuable insights into designing lightweight deep learning models to tackle complex agricultural problems.