Efficient Extraction Method Research Articles

Spectral analysis of intricate orbital angular momentum modes in multiplexing communication using a residual neural network

Vortex beams carrying orbital angular momentum (OAM) modes offer a compelling avenue for increasing optical communication capacity. Despite significant efforts in OAM mode (de)multiplexing, challenges persist in signal demodulation and noise monitoring of multiplexed channels. This challenge largely stems from the absence of efficient feature extraction methods capable of analyzing the OAM spectrum, which delineates mode components and intensity weights. Herein, we introduce a novel strategy for OAM mode spectrum analysis using a residual neural network incorporating interference techniques. By introducing interference between target vortex beams and a spherical wave, resulting in the mapping of OAM distributions onto the stripe features observed in the interferogram. To effectively extract spectrum details from these interferograms, we construct a residual neural network exhibiting high feature processing capabilities, which enables precise analysis of intricate OAM mode spectrum. We demonstrate the effectiveness of this OAM spectrum analysis by obtaining the intensity weights for 15 superimposed modes ranging from −10 to +10, achieving a mean-square-error of 2.3 × 10−4. To validate its practicality, we showcase an 80-channel three-dimensional multiplexing communication system including five OAM modes, two polarizations, and eight wavelengths. The 4 Tbit/s quadrature-phase-shift-keying signals are successfully demodulated with a bit-error rate below 3.34 × 10−6. Additionally, the measured optical signal-to-noise ratio reaches 11.5 dB, indicating the potential applications of this OAM spectrum analysis strategy in signal demodulation and noise monitoring.

Process Optimization of an In-Situ Bioleaching Section with Associated Membrane Filtration in a Field Test Laboratory

The winning of critical raw materials from secondary resources, e.g., from abandoned mines, mining residues, electronic waste or low-grade ores, is a potential source with promising outcomes due to innovative and more efficient extraction methods. The research and education mine “Reiche Zeche” at the TU Bergakademie Freiberg, Germany offers a scientific lab in a real application environment for an in-situ bioleaching section from a low-grade sulfide ore vein with on-site associated membrane downstream processing. The evaluation and resumption of previous research activities showed both the feasibility and the potential for further up-scaling. But there was also potential evaluated for improving the effectiveness, especially in terms of individual process elements within the leaching and membrane cycle as well as regarding microbiology. Based on this, further optimization was carried out and effectiveness was evaluated and compared to the prior state. The results regarding the leaching outcome proved that process optimization leads to stable and continuous leaching process operations as well as to improved and more favorable conditions for the microorganisms due to the implementation of a bioreactor and changing the whole leaching operation from a direct into an indirect process. Furthermore, cleaning in place (CIP) resulted in the maintenance of selectivity despite discontinuous membrane process operation.

MetaTron: advancing biomedical annotation empowering relation annotation and collaboration

BackgroundThe constant growth of biomedical data is accompanied by the need for new methodologies to effectively and efficiently extract machine-readable knowledge for training and testing purposes. A crucial aspect in this regard is creating large, often manually or semi-manually, annotated corpora vital for developing effective and efficient methods for tasks like relation extraction, topic recognition, and entity linking. However, manual annotation is expensive and time-consuming especially if not assisted by interactive, intuitive, and collaborative computer-aided tools. To support healthcare experts in the annotation process and foster annotated corpora creation, we present MetaTron. MetaTron is an open-source and free-to-use web-based annotation tool to annotate biomedical data interactively and collaboratively; it supports both mention-level and document-level annotations also integrating automatic built-in predictions. Moreover, MetaTron enables relation annotation with the support of ontologies, functionalities often overlooked by off-the-shelf annotation tools.ResultsWe conducted a qualitative analysis to compare MetaTron with a set of manual annotation tools including TeamTat, INCEpTION, LightTag, MedTAG, and brat, on three sets of criteria: technical, data, and functional. A quantitative evaluation allowed us to assess MetaTron performances in terms of time and number of clicks to annotate a set of documents. The results indicated that MetaTron fulfills almost all the selected criteria and achieves the best performances.ConclusionsMetaTron stands out as one of the few annotation tools targeting the biomedical domain supporting the annotation of relations, and fully customizable with documents in several formats—PDF included, as well as abstracts retrieved from PubMed, Semantic Scholar, and OpenAIRE. To meet any user need, we released MetaTron both as an online instance and as a Docker image locally deployable.

Extraction and detection of mitragynine in Kratom leaves by high-performance liquid chromatography

Mitragyna speciosa, also known as Kratom, is an evergreen tree native to Southeast Asia that has been used in traditional medicine to relieve pain. The primary active alkaloid, mitragynine, found in Kratom leaves affects the same µ-receptor as the opioids. The potential effects of Kratom leaves on humans draw international concerns. The purpose of this study is to develop an efficient extraction method for mitragynine from dried Kratom leaves. Mitragynine was extracted from grounded Kratom leaves by liquid-liquid extraction and detected by an HPLC system with a C-18 column. The extracted sample was eluted with a mobile phase of 0.05% formic acid and acetonitrile (50:50, v/v), flow rate = 1.0 mL/min. Mitragynine was detected at 223 nm. The amount of mitragynine in four different strains of Kratom leaves was determined (wt/wt). Intra-day (n = 3) and inter-day(n = 3) studies show excellent precision with RSD < 2.1% and RSD < 3.6% respectively.

Strategies for using magnetic beads in enhanced deep eutectic solvent-mechanochemical extraction of natural products from orange peels

To address the challenges of low recovery, prolonged extraction times, and environmental pollution caused by toxic solvents in traditional extraction methods, magnetic bead-enhanced deep eutectic solvent mechanochemical extraction was developed for extracting natural products from orange peels. The extraction efficiencies of deep eutectic solvents were experimentally evaluated, and theoretical methods were used to guide solvent selection. Choline chloride-ethylene glycol demonstrated the highest efficiency under the optimal extraction conditions: a molar ratio of 1:2, no water content, a solid-liquid ratio of 0.08 g/mL, and an extraction time of 60 s. The synergy between the deep eutectic solvent and magnetic bead-enhanced the mechanochemical extraction efficiencies. The study also examined the effects of different magnetic bead types and orange peel powder particle sizes on extraction efficiency, finding that a 0.11 mm particle size combined with CIP@SiO2 yielded the best results. Overall, this study holds promise as an environmentally friendly and efficient extraction method.

ULTRASOUND-ASSISTED DEEP EUTECTIC SOLVENT EXTRACTION OF PHENOLIC COMPOUNDS FROM ROSE MYRTLE

Rose myrtle fruits contain high levels of phenolics and other bioactive compounds. To increase the extraction efficiency of phenolic compounds, ultrasound-assisted- deep eutectic solvents (DESs) was proposed. First of all, six different DESs were used in this research and compared with traditional solvents (ethanol, methanol, and water), and the most suitable solvents for extracting phenolic compounds from rose myrtle fruits was choline chloride: urea (molar ratio of 1:2). Then, the Box–Behnken design was used to find the optimal extraction conditions (ultrasonic power (X1), solid to liquid ratio (X2), and extraction time (X3)) for the achievement of high total phenolic contents (TPC) and antioxidant activity (IC50). The results showed that the optimum ultrasound assisted-DES extraction parameters to achieve the highest TPC and antioxidant activity (IC50) (85.19 mgGAE/g fresh fruit and 2.31 mg/g, respectively) were X1 of 233 W, X2 of 440 mg/ml, and X3 of 55 min. Compared with traditional maceration extraction, ultrasound-assisted-DES has a higher TPC and antioxidant activity (IC50) and shorter extraction time. Scanning electron microscopy images (SEM) revealed that the dissolution of bioactive components was enhanced from the disrupted cells after ultrasound-assisted-DES extraction. Consequently, the combination of ultrasound-assisted-DESs extraction provide an excellent extraction method for efficient extraction of bioactive components of rose myrtle fruit.

Probing the Interaction Mechanism of Sodium Oleate and Dodecyl Amine with Quartz Surfaces in the Presence of Ca2+ Ions by AFM Force Measurement.

Quartz is a key raw material in high-tech fields (such as photovoltaics and semiconductor microelectronics), and the most efficient extraction method of quartz is mineral flotation. Quartz activation plays a crucial role in mineral flotation. However, the mechanism underlying the process remains unclear, and the role of additional metal ions is controversial. In this study, the interaction forces between the quartz surface, the dodecylamine (DDA) cation/sodium oleate (NaOL) anion mixed collectors, and Ca2+ were analyzed using atomic force microscopy in order to systematically explore the activation process of quartz flotation. The results confirmed that the activation process was initialized from NaOL, which was adsorbed on the surface of a calcium-covered quartz surface. The existence of DDA inhibited the binding of Ca2+ to NaOL so that more Ca2+ was adsorbed on the quartz surface to provide the adsorption site for NaOL. Moreover, the best adsorption condition of Ca2+ + NaOL + DDA mixed solution was analyzed by quartz crystal microbalance with dissipation, and it demonstrated that the most stable chemisorption environment on the quartz surface was at pH 11.0. In these circumstances, Ca2+ could first adsorb in a point-like manner on the quartz surface, which was then adsorbed with a mixture of NaOL and DDA. This result showed that, at a specific pH, Ca2+ could encourage the coadsorption of cationic/anionic mixed collectors on quartz. This work provides an important new understanding of the intermolecular interactions that take place during complex mineral flotation processes between chemical additives and mineral surfaces. The methodology used in this study can be easily adapted to different interfacial processes, including water treatment, membrane technology, bioengineering, and oil production.

Single-tone Aerodynamic Noise Source Separation for Gas Turbines

The aerodynamic noise in the gas turbines primarily originates from the pulsations generated by the air within compressors, and it is a combination of single-tone noise caused by the pressure fluctuation of blades and broadband noise generated by interactions between turbulent fluid and blades. Therefore, the separation of single-tone noise is vital in predicting the contribution of each noise source for noise reduction and health monitoring. In current practice, the Vold-Kalman filter (VKF) is widely employed for single-tone noise separation. However, the optimal VKF bandwidth is often difficult to determine, thereby impeding the separation accuracy of single-tone noise. To solve this problem, this paper proposes a single-tone noise separation method by utilizing computed order tracking based on double criteria VKF (DC-VKF). Firstly, the order spectrum of the original signal is obtained by order analysis. Secondly, the relative error of order amplitude (REOA) and total relative standard deviation (TRSD) of the extracted signal are calculated under different bandwidths and corresponding curves are separately carried out. The optimal bandwidth is determined by identifying the first bandwidth that meets either the REOA or TRSD threshold, which are respectively set based on trend of curves. Finally, the corresponding order components are separated by VKF with optimized bandwidth. The effectiveness of DC-VKF is validated on simulation and gas turbine datasets. This paper provides an efficient extraction method for the separation of single-tone aerodynamic noise of gas turbines, and provides a scientific basis for noise monitoring, evaluation, and reduction.

High enrichment and measurement of heterocyclic aromatic hydrocarbons from environmental waters with magnetic resorcinol-formaldehyde nanocomposites coupled with high performance liquid chromatography

Heterocyclic aromatic hydrocarbons are concerned pollutants with carcinogenic toxicity, which exist universally in various environmental matrices and have great harm to environmental and human health. In present work, magnetic resorcinol-formaldehyde composites (Fe3O4@SiO2@R–F) were fabricated via aldol condensation reaction under alkaline condition. The prepared magnetic materials were examined and analyzed with Fourier transform infrared spectroscopy and other related instruments. The Fe3O4@SiO2@R–F composites were utilized to develop an efficient magnetic solid phase extraction (MSPE) method for extracting six heteropolyclic aromatic hydrocarbons from environmental water samples including carbazole (CB), 7-methylquinoline (7-MQL), 9-methylcarbazole (9-MCB), dibenzothiophene (DBT), 4-methyldibenzothiophene (4-MDBT), and 4,6-dimethyldibenzothiophene (4,6-DMDBT). The analytes were analyzed by high performance liquid chromatography-ultraviolet variable wavelength detector (HPLC-VWD). The main factors affecting MSPE were optimized. With the optimal parameters, 9-MCB and 4-MDBT have good linearity over the concentration range of 0.1–300 μg L−1, and 7-MQL, CB, DBT and 4,6-DMDBT have good linearity over the concentration range of 0.5–300 μg L−1. The limits of detection were over the concentration range of 0.012–0.031 μg L−1. This method was successfully employed to measure real waters, and the spiked recoveries ranged from 89.4% to 99.9%. The results confirmed that the developed method was reliable, robust and could be employed as a usefully alternate way for analyzing such pollutants in waters.

Fabrication of a bio-based polymer adsorbent and its application for extraction and determination of glycosides from Huangqi Liuyi Decoction

Huangqi Liuyi Decoction, a famous classical Chinese prescription, shows significant curative effect on diabetes and its complications, in which calycosin-7-glucoside, liquiritin and glycyrrhizic acid are the main components that playing these mentioned pharmacological activity, under the synergistic action of various other ingredients in the decoction. However, there are significant differences in the content of active compounds in Chinese medicinal materials, which mainly due to origin, picking seasons, and processing methods. Hence, the accurate content of the glycosides is the prerequisite for ensuring the pharmacological efficacy. Aiming at establishing an efficient extraction and determination method for accurate quantitative analysis of calycosin-7-glucoside, liquiritin and glycyrrhizic acid in Huangqi Liuyi Decoction, an on line solid-phase extraction-high-performance liquid chromatography method was developed, using a homemade bio-based monolithic adsorbent. The bio-based adsorbent was prepared in a stainless steel tube, using bio-monomers of methyleugenol and S-allyl-L-cysteine, which effectively reduced the dependence of the polymer field on non-renewable fossil resources and reduced carbon emissions. Furthermore, the prepared adsorbent owned abundant chemical groups, which can produce interactions of hydrogen bond, dipole-dipole, π-π and hydrophobic force with the target glycosides, thus improving the specific recognition ability of the adsorbent. The experiments were carried out on an LC-3000 HPLC instrument with a six-way valve. Methodology validation indicates that the recovery is in the range of 97.0%-103.4% with the RSD in the range of 1.6%-4.0%, due to the specific selectivity of the bio-based monolithic adsorbent for these three glycosides, and good matrix-removal ability for Huangqi Liuyi decoction. The limit of detection is 0.17, 0.50 and 0.33μg/mL for calycosin-7-glucoside, liquiritin and glycyrrhizic acid, respectively, and the limit of quantitation is 0.50, 1.50 and 1.00μg/mL, respectively, with the linear range of 2-200μg/mL for calycosin-7-glucoside, and 5-500μg/mL for liquiritin and glycyrrhizic acid. The present work provided a simple and efficient method for the extraction and determination of glycosides in complex medicinal plants.

Cross-Modality Interaction-Based Traffic Accident Classification

Traffic accidents on the road lead to serious personal and material damage. Furthermore, preventing secondary accidents caused by traffic accidents is crucial. As various technologies for detecting traffic accidents in videos using deep learning are being researched, this paper proposes a method to classify accident videos based on a video highlight detection network. To utilize video highlight detection for traffic accident classification, we generate information using the existing traffic accident videos. Moreover, we introduce the Car Crash Highlights Dataset (CCHD). This dataset contains a variety of weather conditions, such as snow, rain, and clear skies, as well as multiple types of traffic accidents. We compare and analyze the performance of various video highlight detection networks in traffic accident detection, thereby presenting an efficient video feature extraction method according to the accident and the optimal video highlight detection network. For the first time, we have applied video highlight detection networks to the task of traffic accident classification. In the task, the most superior video highlight detection network achieves a classification performance of up to 79.26% when using video, audio, and text as inputs, compared to using video and text alone. Moreover, we elaborated the analysis of our approach in the aspects of cross-modality interaction, self-attention and cross-attention, feature extraction, and negative loss.

Numerical modelling of vibratory extraction of monopiles from clayey seabed

Understanding and optimizing the extraction of offshore monopiles from the seabed in decommissioning operations give both environmental benefits and cost reduction. The literature highlights the significance of utilizing a vibratory hammer for the complete removal of monopiles in granular or non-cohesive seabed, e.g., sandy soil, offering a cost-effective alternative to the conventional cutting method. However, the available knowledge for applying this method to cohesive soil is limited. The present study examines vibratory pile extraction from cohesive soil, for both soft and stiff clays, with the aim of assessing the resulting extraction force required for pile removal. Coupled Eulerian-Lagrangian (CEL) analyses with two domains of pile and soil are performed using the finite element package ABAQUS. A modified Mohr-Coulomb (MCC) soil model is coded into a VUSDFLD subroutine to properly simulate the soil softening and shear strain rate dependency of seabed soil in an Eulerian domain. The importance of shear strain and softening effects is underscored for soft clay by demonstrating an approximately 20% reduction in the force needed to initiate cyclic extraction of the monopile compared to scenarios where these effects are not included. However, incorporating these effects does not affect the behavior of stiff clay. The present numerical findings show that adopting vibratory extraction method leads to more soil resistance in cohesive soils at the beginning of the extraction. Considering the added weight og the vibratory hammer which should be clamped to the monopile top head leads to exceedance of the required force than the one required for removal of the monopile without vibration (i.e., the pulling-out method). This demonstrates that the vibratory extraction method is not an efficient method for monopile extraction from clayey seabed despite its beneficial effects reported in the literature for sandy soil.

W-FENet: Wavelet-based Fourier-Enhanced Network Model Decomposition for Multivariate Long-Term Time-Series Forecasting

Accurately predicting the future trend of a time series holds immense importance for decision-making and planning across various domains, including energy planning, weather forecasting, traffic warning, and other practical applications. Recently, deep learning methods based on transformers and time convolution networks (TCN) have achieved a surprising performance in long-term sequence prediction. However, the attention mechanism for calculating global correlation is highly complex, and TCN methods do not fully consider the characteristics of time-series data. To address these challenges, we introduce a new learning model named wavelet-based Fourier-enhanced network model decomposition (W-FENet). Specifically, we have used trend decomposition and wavelet transform to decompose the original data. This processed time-series data can then be more effectively analyzed by the model and mined for different components in the series, as well as capture the local details and overall trendiness of the series. An efficient feature extraction method, Fourier enhancement-based feature extraction (FEMEX), is introduced in our model. The mechanism converts time-domain information into frequency-domain information through a Fourier enhancement module, and the obtained frequency-domain information is better captured by the model than the original time-domain information in terms of periodicity, trend, and frequency features. Experiments on multiple benchmark datasets show that, compared with the state-of-the-art methods, the MSE and MAE of our model are improved by 11.1 and 6.36% on average, respectively, covering three applications (i.e. ETT, Exchange, and Weather).

Automated sepsis detection with vancomycin- and allantoin-polydopamine magnetic nanoparticles

Rapid and accurate identification of the bacteria responsible for sepsis is paramount for effective patient care. Molecular diagnostic methods, such as polymerase chain reaction (PCR), encounter challenges in sepsis due to inhibitory compounds in the blood, necessitating their removal for precise analysis. In this study we present an innovative approach that utilizes vancomycin (Van) and allantoin (Al)-conjugated polydopamine (PDA)-coated magnetic nanoparticles (MNPs) for the rapid and automated enrichment of bacteria and their DNA extraction from blood without inducing clumping and aggregation of blood. Al/Van-PDA-MNPs, facilitated by IMS, eliminate the need for preliminary sample treatments, providing a swift and efficient method for bacterial concentration and DNA extraction within an hour. Employing Al/Van-PDA-MNPs within an automated framework has markedly improved our ability to pre-concentrate various Gram-negative and Gram-positive bacteria directly from blood samples. This advancement has effectively reduced the detection threshold to 102 colony-forming unit/mL by both PCR and quantitative PCR. The method's expedited processing time, combined with its precision, positions it as a feasible diagnostic tool for diverse healthcare settings, ranging from small clinics to large hospitals. Furthermore, the innovative application of nanoparticles for DNA extraction holds promising potential for advancing sepsis diagnostics, enabling earlier interventions and improving patient outcomes.

Improved Method for Extracting Nitrites in Soil

Soil nitrite (NO2−) is an important reactive intermediate in many nitrogen transformation processes, but it is unstable under acidic conditions and may be lost as gaseous N. The canonical extraction method of soil NO2− using a potassium chloride (KCl) solution greatly underestimates its concentration. To reflect the concentration more accurately, we optimized the extraction method of soil NO2− for three agricultural soils differing in soil texture and pH, an alkalic fluvo-aquic soil and acidic Mollisol and Ultisol soils, respectively. Both extractable soil ammonium (NH4+) and nitrate (NO3−) were systematically investigated to optimize the simultaneous extraction of soil inorganic nitrogen. The effects of different extractants (deionized water (DIW), un-buffered 2 mol L−1 KCl, and pH-buffered 2 mol L−1 KCl), shaking time (10 and 30 min), and storage duration of the extracts (stored at −20 °C for 1 day, and at 4 °C for 1, 3, and 6 days) on the determination of soil inorganic nitrogen were investigated. The results showed that the un-buffered KCl extractant significantly underestimated soil NO2− concentration compared to DIW. The highest recovery of NO2− was obtained by extracting with DIW at 10 min of shaking for all three soils. Compared with DIW, the concentration of NH4+ and NO3− in soil extracted from the KCl solution increased significantly. Furthermore, the soil inorganic nitrogen content of extracts stored at 4 °C for one day was closer to the direct measurements of fresh samples than with the other storage methods. Overall, the recommended analysis method for soil NO2− was extraction by DIW, shaking for 10 min, and filtering with a 0.45 µm filter, while soil NH4+ and NO3− were extracted with a KCl solution and shaken for 30 min. The extract should be stored at 4 °C and analyzed within 24 h. Our study provides an efficient extraction method for soil NO2− and supports studies on the biogeochemical nitrogen cycle, e.g., in the investigation of soil nitrous acid (HONO) and nitric oxide (NO) emissions.

Development and validation of an acid/alkaline digestion method for efficient microplastic extraction from wastewater treatment plant effluents: Sulfuric acid concentration and contact time do matter

Accurate analysis of microplastic particles (MPs) in environmental samples requires removal of interferences during sample preparation. Wastewater samples are interference-rich and thus particularly challenging, with concentrated sulfuric acid currently deemed impractical as a reagent. Therefore, this study aimed to establish a straightforward, effective, and safe method employing concentrated sulfuric acid and potassium hydroxide to eliminate interferents from effluent samples obtained from wastewater treatment plants (WWTPs). We found that 80 % sulfuric acid at room temperature with a brief contact time of 5 min was viable through a qualitative spot test involving 37 plastics categorized into three types (I, II, and III) based on their polymer structure's oxygen position. A quantitative assessment revealed that treatments involving H2SO4 and KOH (20 %, 24 h, 48 °C), either separately or in combination, had no discernible physical impact on the overall plastics, except for a subtle one for Type III plastics (e.g., nylon and PMMA) known to be labile under harsh pH conditions. This acid/alkaline digestion (AAD) method, incorporating such conditions for H2SO4 and KOH treatments, yielded a high mass removal efficacy (97.8 ± 2.4 %, n = 13) for eliminating natural particle interferents for primary, secondary, and tertiary effluent samples. Furthermore, the AAD method allowed for the determination of MPs in effluents with high surrogate particle recoveries (e.g., 95.1 % for larger than 500 μm size fraction). This method is readily adaptable to create appropriate protocols for different types of environmental matrices.

An innovative optimized protocol for high-quality genomic DNA extraction from recalcitrant Shea tree (Vitellaria paradoxa, C.F. Gaertn) plant and its suitability for downstream applications.

It is not always easy to find a universal protocol for the extraction of genomic DNA (gDNA) from plants. Extraction of gDNA from plants such as shea with a lot of polysaccharides in their leaves is done in two steps: a first step to remove the polysaccharides and a second step for the extraction of the gDNA. In this work, we designed a protocol for extracting high-quality gDNA from shea tree and demonstrate its suitability for downstream molecular applications. Fifty milligrams of leaf and root tissues were used to test the efficiency of our protocol. The quantity of gDNA was measured with the NanoDrop spectrometer and the quality was checked on agarose gel. Its suitability for use in downstream applications was tested with restriction enzymes, SSRs and RAPD polymerase chain reactions and Sanger sequencing. The average yield of gDNA was 5.17; 3.96; 2.71 and 2.41µg for dry leaves, dry roots, fresh leaves and fresh roots respectively per 100mg of tissue. Variance analysis of the yield showed significant difference between all tissue types. Leaf gDNA quality was better compared to root gDNA at the absorbance ratio A260/280 and A260/230. The minimum amplifiable concentration of leaf gDNA was 1pg/µl while root gDNA remained amplifiable at 10pg/µl. Genomic DNA obtained was also suitable for sequencing. This protocol provides an efficient, convenient and cost effective DNA extraction method suitable for use in various vitellaria paradoxa genomic studies.

Development of an intelligent multi-scale defect detection system for aspheric lenses with high accuracy and efficiency

Existing methods struggle to fulfill the demand for fast and precise detection of multi-scale faults in aspheric lenses. A high accuracy and efficiency intelligent detection system for aspheric lenses is developed, which utilizes four-step phase-shifted grating optical imaging and Fourier low-pass filtering as image acquisition and preprocessing systems. Accurate and efficient defect feature extraction methods are designed based on regression loss and lightweight feature selection networks, and the model's ability to learn multi-scale defect information is improved. Finally, experimental tests were conducted based on the self-constructed multi-scale defect datasets of aspheric lenses. The results show that the detection system achieves a global accuracy of 96.3 %, an accuracy of 63.8 % for tiny defects detection, as well as a single-image detection time of 24.9 ms, which is a significant improvement in terms of accuracy and efficiency compared with existing methods. Therefore, the intelligent detection system can be widely applied to defect detection in aspheric lenses.

Bilingual video captioning model for enhanced video retrieval

Many video platforms rely on the descriptions that uploaders provide for video retrieval. However, this reliance may cause inaccuracies. Although deep learning-based video captioning can resolve this problem, it has some limitations: (1) traditional keyframe extraction techniques do not consider video length/content, resulting in low accuracy, high storage requirements, and long processing times; (2) Arabic language support in video captioning is not extensive. This study proposes a new video captioning approach that uses an efficient keyframe extraction method and supports both Arabic and English. The proposed keyframe extraction technique uses time- and content-based approaches for better quality captions, fewer storage space requirements, and faster processing. The English and Arabic models use a sequence-to-sequence framework with long short-term memory in both the encoder and decoder. Both models were evaluated on caption quality using four metrics: bilingual evaluation understudy (BLEU), metric for evaluation of translation with explicit ORdering (METEOR), recall-oriented understudy of gisting evaluation (ROUGE-L), and consensus-based image description evaluation (CIDE-r). They were also evaluated using cosine similarity to determine their suitability for video retrieval. The results demonstrated that the English model performed better with regards to caption quality and video retrieval. In terms of BLEU, METEOR, ROUGE-L, and CIDE-r, the English model scored 47.18, 30.46, 62.07, and 59.98, respectively, whereas the Arabic model scored 21.65, 36.30, 44.897, and 45.52, respectively. According to the video retrieval, the English and Arabic models successfully retrieved 67% and 40% of the videos, respectively, with 20% similarity. These models have potential applications in storytelling, sports commentaries, and video surveillance.

Cellulose nanocrystals from marine algae Cladophora glomerata by using microwave-assisted extraction

Algae of the order Cladophorales are the source of a unique nanocellulose with high crystallinity and a large aspect ratio, enabling broad surface modification. Cellulose nanocrystals (CNCs) are obtained via acid hydrolysis of nanocellulose, which is highly crystalline. However, the production of CNCs from Cladophorales algae is limited and still uses a conventional heating method. Thus, this study aimed to develop a microwave-assisted extraction (MAE) method for fast and efficient extraction of CNCs from Cladophora glomerata algae. Additionally, we replaced the use of hypochlorite with H2O2, which is more environmentally friendly, and compared the CNCs obtained from the conventional methods with our new method. The functional structure of CNCs was confirmed by Fourier-transform infrared spectroscopy. Single-step H2O2 bleaching with MAE yielded the smallest-sized CNCs. Our developed method resulted in the production of CNCs with a high crystallinity index, high thermal stability, and high purity of native cellulose. Additionally, none of the CNCs were toxic to primary normal human dermal fibroblasts. The properties of the isolated CNCs may make them useful materials in pharmaceutical and cosmetic formulations.