Classification Function Research Articles

ConCave-Convex procedure for support vector machines with Huber loss for text classification

The classical support vector machine (SVM) adopts the linear Hinge loss whereas the least squares SVM (LS-SVM) employs the quadratically growing least squares loss function. The robust Ramp loss function is employed in Ramp loss SVM (RSVM) that truncates the Hinge loss function and becomes flat a specified point afterwards, thus, increases robustness to outliers. Recently proposed SVM with pinball loss (pin-SVM) utilizes pinball loss function that maximizes the margin between the class hyperplanes based on quantile distance. Huber loss function is the generalization of linear Hinge loss and quadratic loss. Huber loss solves sensitivity issues of least squares loss to noise and outlier. In this work, we employ the robust Huber loss function for SVM classification for improved generalization performance. The cost function of the proposed approach consists of one convex and one non-convex part, which might sometimes provide local optimum solution instead of a global optimum. We suggest a ConCave-Convex Procedure (CCCP) to resolve this issue. Additionally, the proximal cost is scaled for each class sample based on their class size to reduce the effect of the class imbalance problem. Thus, it can be claimed that the proposed approach incorporates class imbalance learning as well. Extensive experimental analysis establishes efficacy of the proposed method. Furthermore, a sequential minimal optimization (SMO) procedure for high dimensional HSVM is proposed and its performance is tested on two text classification datasets.

A deep neural network model with spectral correlation function for electrocardiogram classification and diagnosis of atrial fibrillation

Atrial Fibrillation (AF) is a common type of irregular heartbeat, and early detection can significantly improve treatment outcomes and prognoses. Single-lead Electrocardiogram (ECG) devices are under extensive scrutiny for monitoring patients' heart health worldwide. Standardized ECG signal monitoring has demonstrated a significant reduction in mortality rates associated with severe cardiovascular diseases. However, the automatic detection method for AF requires significant improvement. This study presents a novel approach that utilizes the cyclostationary analysis of ECG signals, uncovering a spectral hidden periodicity between the QRS-T (the main wave components representing electrical activity in the heart) complexes of the ECG signal through the Spectral Correlation Function (SCF). To validate the proposed method's performance, the single ECG's SCF coefficients are applied to the Convolutional Recurrent Neural Network (CRNN), which consists of convolutional and long short-term memory (LSTM) layers, on the 2017 PhysioNet challenge dataset. The obtained results demonstrate that the proposed approach efficiently represents ECG signals through SCF coefficients, leading to the accurate detection of AF with an average accuracy of 92.76% and an average F1-score of 89.1%.

Rapid post-earthquake damage assessment of building portfolios through deep learning-based component-level image recognition

Frequent seismic events significantly heighten the likelihood of the building structure being impacted throughout its operational lifecycle. Seismic effects on these structures result in a notable reduction in structural safety and serviceability. Enabling the post-earthquake recovery of building structures necessitates a precise and swift assessment of earthquake-induced damage. The conventional method for assessing building damage involves collecting data through close manual observation or contact inspection. Although it can obtain relatively accurate results by combining evaluation specifications, it is time-consuming and labour-intensive. Deep learning (DL) is data-driven and employs computational methods for data processing. The image classification function of convolutional neural network (CNN) in DL brings great convenience to image data processing. It has been applied to various aspects of structural health monitoring/post-disaster assessment. However, most of the current studies focus on the component level and lack of a comprehensive perspective on the whole structure, which is not conducive to the judgment of the overall condition. Consequently, this paper proposes a rapid assessment method of the overall damage level of post-earthquake buildings based on component images and DL. Two component-level image classification models, component type and damage level, are firstly trained. Then, the images of the buildings to be evaluated after the earthquake are collected and classified. Combined with the weights of different component types and damage levels proposed, the overall condition scores and grades of the structures can be obtained by weighted calculation. The proposed method realizes an overall evaluation of the structural damage condition after seismic events, and further extends to the building portfolios, providing actionable guidance for subsequent personnel and fund allocation, rescue, and maintenance measures. Due to limitations in the dataset, such as the potential biases in the training dataset, the trained model is not perfect and faces challenges in distinguishing between minor and moderate damage. In the future, the dataset should update and ideally cover a wide range of building types, component types, and damage levels to ensure the model's robustness and applicability to various real-world scenarios.

The regulatory role of tRNA-derived small RNAs in the prognosis of gastric cancer

In recent years, tRNA-derived small RNAs (tsRNAs) including tRNA-derived stress-induced RNAs (tiRNAs) and tRNA-derived fragments (tRFs), with specific structure and enriched in body fluids, have been found to have specific biological functions. In this paper, the biogenesis, classification, subcellular localization, and biological functions of tsRNAs were summarized. It has been proved that tsRNAs affected tumor cells in proliferation, apoptosis, migration and invasion, and played roles in regulating the occurrence and development of various tumors. In gastric cancer (GC), the imbalance of tsRNAs, such as tRF-33-P4R8YP9LON4VDP, tRF-17-WS7K092, tRF-23-Q99P9P9NDD and others, was closely related to the clinicopathological characteristics of GC patients. Some tsRNAs, such as tRF-23-Q99P9P9NDD, tRF-31-U5YKFN8DYDZDD, and tRF-27-FDXXE6XRK45 promoted the proliferation, migration and invasion of GC cells. Other tsRNAs, such as tRF-41-YDLBRY73W0K5KKOVD, tRF-18-79MP9PO4, and tRF-Glu-TTC-027 inhibited the proliferation, migration and invasion of GC cells. The tsRNAs played roles in the occurrence of GC were through several signaling pathways, such as phosphoinositide 3-kinase (PI3K)-AKT serine/threonine kinase (AKT), Wnt-β-Catenin, and mitogen-activated protein kinase (MAPK) pathways. These findings may provide new strategies for the diagnosis and treatment of GC.

B-TBM: A Novel Deep Learning Model with Enhanced Loss Function for HAZOP Risk Classification Using Natural Language Statistical Laws

HAZOP is a paradigm of industrial safety, and the introduction of deep learning-based HAZOP text categorization marks the arrival of an intelligent era of safety analysis. However, existing risk analysis methods have limitations in processing complex texts and extracting deep risk features. To solve this problem, this paper proposes a novel HAZOP risk event classification model based on BERT, BiLSTM, and TextCNN. The complexity of HAZOP text is revealed by introducing statistical laws of natural language, such as Zipf’s law and Heaps’ law, and the outputs of different levels of BERT are further combined linearly to collaborate with BiLSTM and TextCNN to capture long-term dependency and local contextual information for a more accurate classification task. Meanwhile, an improved loss function is proposed to effectively solve the deficiencies of the traditional cross-entropy loss function in the mislabeling process and improve the generalization ability of the model. It is experimentally demonstrated that the accuracy of the model is improved by 3% to 4% compared to the traditional BERT model in the task of severity and possibility classification of HAZOP reports. This study not only improves the accuracy and efficiency of HAZOP risk analysis, but also provides new ideas and methods for the application of natural language processing in industrial safety.

Alteration of Cecal Microbiota by Antimicrobial Peptides Enhances the Rational and Efficient Utilization of Nutrients in Holstein Bulls.

We previously observed that supplementation with antimicrobial peptides facilitated the average daily weight gain, net meat, and carcass weights of Holstein bulls. To expand our knowledge of the possible impact of antimicrobial peptides on cecum microbiota, further investigations were conducted. In this study, 18 castrated Holstein bulls with insignificant weight differences and 10 months of age were split randomly into two groups. The control group (CK) was fed a basic diet, whereas the antimicrobial peptide group (AP) was supplemented with 8 g of antimicrobial peptides for 270 days. After slaughter, metagenomic and metabolomic sequencing analyses were performed on the cecum contents. The results showed significantly higher levels of amylase, cellulase, protease, and lipase in the CK than in the AP group (P ≤ 0.05). The levels of β-glucosidase and xylanase (P ≤ 0.05), and acetic and propionic acids (P ≤ 0.01), were considerably elevated in the AP than in the CK group. The metagenome showed variations between the two groups only at the bacterial level, and 3258 bacteria with differences were annotated. A total of 138 differential abundant genes (P < 0.05) were identified in the CAZyme map, with 65 genes more abundant in the cecum of the AP group and 48 genes more abundant in the cecum of the CK group. Metabolomic analysis identified 68 differentially expressed metabolites. Conjoint analysis of microorganisms and metabolites revealed that Lactobacillus had the greatest impact on metabolites in the AP group and Brumimicrobium in the CK group. The advantageous strains of the AP group Firmicutes bacterium CAG:110 exhibited a strong symbiotic relationship with urodeoxycholic acid and hyodeoxycholic acid. This study identified the classification characteristics, functions, metabolites, and interactions of cecal microbiota with metabolites that contribute to host growth performance. Antimicrobial peptides affect the cecal microorganisms, making the use of nutrients more efficient. The utilization of hemicellulose in the cecum of ruminants may contribute more than cellulose to their production performance.

The relationship of hemorheological blood values and blood velocity of microcirculatory bloodstream in rats` skin vessels

Non-invasive study of blood rheology is relevant, but quite complex issue. When systemic blood viscosity and hematocrit levels deviate, blood flow indicators in different parts of microvasculature change. Purpose of the study – research of blood flow characteristics in skin microcirculatory bloodstream of rats obtained by high-frequency ultrasound Dopplerography (HFUD) with given changes in rheological blood indicators. The studies were carried out on pubescent male Wistar rats. 3 experimental groups were formed. Group 1 (n = 21) “Hemodilution” – viscosity 1.99 ± 0.02 mPa*s, hematocrit 31.48 ± 0.31%. Group 2 (n = 32) “Reference values” – animals with unchanged blood levels – viscosity 2.84 ± 0.03 mPa*s, hematocrit 41.60 ± 0.3%. Group 3 (n = 32) “Erythrocytosis” – viscosity 3.95 ± 0.04 mPa*s, hematocrit 54.56 ± 0.23%. Dynamic blood viscosity in vitro studies were carried out on oscillatory viscometer. In order to evaluate hematocrit level heparinized whole blood was centrifuged in glass capillaries using; hematocrit values were assessed taking into account sedimentation of formed elements column using a hematocrit reader card. Blood flow in skin microcirculatory bloodstream of rats` left thigh area was estimated by HFUD method using Minimax-Doppler-K hardware and software system, with ultrasound transducer (frequency 20 MHz). Statistical analysis showed the models are correct. Blood indicators of the animals in three experimental groups differed statistically and significantly in terms of blood viscosity and hematocrit. Discriminant analysis was used to determine the relations between rheological blood parameters and characteristics of blood velocity in microcirculatory bloodstream which made it possible to identify the most significant characteristics of blood flow that tend to change depending on altered blood composition. These include: mean systolic velocity Vas (p 0.01), mean velocity Vam (p 0.001), mean volume velocity Qam (p 0.001), vascular resistance index RI (p 0.01) and the percentage of blood cells moving in low-speed H' (p = 0.03). The reliability of selected characteristics was checked with one-way analysis of variance; and their significance in determining membership in “Hemodilution”, “Reference values” or “Erythrocytosis” groups according to HFUD data was confirmed. Based on this analysis classification functions were generated for non-invasive dynamic blood viscosity determination according to ultrasound Dopplerography data.

Fixing imbalanced binary classification: An asymmetric Bayesian learning approach.

Most statistical and machine learning models used for binary data modeling and classification assume that the data are balanced. However, this assumption can lead to poor predictive performance and bias in parameter estimation when there is an imbalance in the data due to the threshold election for the binary classification. To address this challenge, several authors suggest using asymmetric link functions in binary regression, instead of the traditional symmetric functions such as logit or probit, aiming to highlight characteristics that would help the classification task. Therefore, this study aims to introduce new classification functions based on the Lomax distribution (and its variations; including power and reverse versions). The proposed Bayesian functions have proven asymmetry and were implemented in a Stan program into the R workflow. Additionally, these functions showed promising results in real-world data applications, outperforming classical link functions in terms of metrics. For instance, in the first example, comparing the reverse power double Lomax (RPDLomax) with the logit link showed that, regardless of the data imbalance, the RPDLomax model assigns effectively lower mean posterior predictive probabilities to failure and higher probabilities to success (21.4% and 63.7%, respectively), unlike Logistic regression, which does not clearly distinguish between the mean posterior predictive probabilities for these two classes (36.0% and 39.5% for failure and success, respectively). That is, the proposed asymmetric Lomax approach is a competitive model for differentiating binary data classification in imbalanced tasks against the Logistic approach.

An anthropomorphic diagnosis system of pulmonary nodules using weak annotation-based deep learning

The accurate categorization of lung nodules in CT scans is an essential aspect in the prompt detection and diagnosis of lung cancer. The categorization of grade and texture for nodules is particularly significant since it can aid radiologists and clinicians to make better-informed decisions concerning the management of nodules. However, currently existing nodule classification techniques have a singular function of nodule classification and rely on an extensive amount of high-quality annotation data, which does not meet the requirements of clinical practice. To address this issue, we develop an anthropomorphic diagnosis system of pulmonary nodules (PN) based on deep learning (DL) that is trained by weak annotation data and has comparable performance to full-annotation based diagnosis systems. The proposed system uses DL models to classify PNs (benign vs. malignant) with weak annotations, which eliminates the need for time-consuming and labor-intensive manual annotations of PNs. Moreover, the PN classification networks, augmented with handcrafted shape features acquired through the ball-scale transform technique, demonstrate capability to differentiate PNs with diverse labels, including pure ground-glass opacities, part-solid nodules, and solid nodules. Through 5-fold cross-validation on two datasets, the system achieved the following results: (1) an Area Under Curve (AUC) of 0.938 for PN localization and an AUC of 0.912 for PN differential diagnosis on the LIDC-IDRI dataset of 814 testing cases, (2) an AUC of 0.943 for PN localization and an AUC of 0.815 for PN differential diagnosis on the in-house dataset of 822 testing cases. In summary, our system demonstrates efficient localization and differential diagnosis of PNs in a resource limited environment, and thus could be translated into clinical use in the future.

A Comparative Study of Metaheuristic Feature Selection Algorithms for Respiratory Disease Classification.

The correct diagnosis and early treatment of respiratory diseases can significantly improve the health status of patients, reduce healthcare expenses, and enhance quality of life. Therefore, there has been extensive interest in developing automatic respiratory disease detection systems. Most recent methods for detecting respiratory disease use machine and deep learning algorithms. The success of these machine learning methods depends heavily on the selection of proper features to be used in the classifier. Although metaheuristic-based feature selection methods have been successful in addressing difficulties presented by high-dimensional medical data in various biomedical classification tasks, there is not much research on the utilization of metaheuristic methods in respiratory disease classification. This paper aims to conduct a detailed and comparative analysis of six widely used metaheuristic optimization methods using eight different transfer functions in respiratory disease classification. For this purpose, two different classification cases were examined: binary and multi-class. The findings demonstrate that metaheuristic algorithms using correct transfer functions could effectively reduce data dimensionality while enhancing classification accuracy.

Defining the Etiology of Renal Allograft Dysfunction Using Banff 2019 Classification: Correlation with Post-Transplant Duration and Creatinine Levels-A Comprehensive Analysis of 200 Renal Biopsies at a Tertiary Care Medical Center Hospital.

Chronic kidney disease is a growing global health issue, contributing significantly to morbidity and mortality. The incidence of end-stage renal disease (ESRD) is approximately 100 per million population. Renal transplantation remains the cornerstone treatment for ESRD, with a projected 20-year survival rate of 60%. We aim to define the etiology of renal allograft dysfunction using the Banff 2019 classification by analyzing 200 renal allograft biopsies in correlation with creatinine levels across post-transplant time frames. 200 renal allograft biopsies are analyzed using the recent Banff 2019 classification with creatinine levels and post-transplant duration correlation. The study included 150 (75%) male patients and 50 (25%) female patients, with the majority 78 (39%) representing the age group of 16-30 years. 36 (18%) biopsies were within 3-month post-transplant, while 92 (46%) were 2-year post-transplant. According to the Banff 2019 classification, 92 (46.0%) transplant rejection biopsies were identified, with most 54 (27%) exhibiting antibody-mediated rejection (Category 2), including 40 (20%) active acute antibody-mediated rejection (ABMR) and 14 (7.0%) chronic active ABMR. T-cell-mediated rejection (TCMR; Category 4) represented 12 (6%) biopsies, including 10 (5%) acute TCMR and 2 (1%) chronic active TCMR. Category 5, the miscellaneous group, represented 100 (50%) biopsies, out of which 32 (16%) exhibited calcineurin inhibitor (CNI) toxicity, 38 (19%) acute tubular necrosis, and 8 (4%) thrombotic microangiopathy. A notable variation in the dysfunction distribution across different post-transplant time frames indicated a temporal evolution in the underlying causes of allograft dysfunction. Specific Banff categories showed a robust association with renal dysfunction, potentially contributing to the elevation of creatinine levels and renal function deterioration. Our study highlights the intricate pathophysiology of renal allograft dysfunction. Most biopsies were attributed to ABMR whereas one-third of biopsies exhibited mixed lesions (ABMR and TCMR or ABMR and calcineurin inhibitor toxicity (CNIT)). Additionally, this study suggests that renal allograft rejection remains a significant contributor to graft dysfunction. A complex interplay between histological findings, Banff classification, and renal function is noted. A significant difference in the distribution of dysfunction across post-transplant time frames is noted suggesting a temporal evolution in the etiology of allograft dysfunction. Certain Banff categories demonstrate a stronger association with renal dysfunction that may influence creatinine level increase and renal function deterioration. In correspondence to the recent Banff 2019 guidelines for diagnosing ABMR, we emphasize the role of C4d staining on immunofluorescence or immunohistochemistry in allograft biopsies as imperative for timely diagnosis and immunosuppressant therapy adjustment, ultimately enhancing graft survival. Further research is needed to elucidate the underlying mechanisms driving renal dysfunction in different Banff categories, ultimately informing personalized management strategies for patients with renal allograft dysfunction. In line with the Banff 2019 guidelines for diagnosing ABMR, this study highlights the critical role of C4d staining through immunofluorescence or immunohistochemistry in allograft biopsies for early diagnosis and timely adjustment of immunosuppressive therapy, ultimately improving graft survival.

A novel binary data classification algorithm based on the modified reaction-diffusion predator-prey system with Holling-II function.

In this study, we propose a modified reaction-diffusion prey-predator model with a Holling-II function for binary data classification. In the model, we use u and v to represent the densities of prey and predators, respectively. We modify the original equation by substituting the term v with f-v to obtain a stable and clear nonlinear decision surface. By employing a finite difference method for numerical solution of the original model, we conduct various experiments in two-dimensional and three-dimensional spaces to validate the feasibility of the classifier. Additionally, with consideration for wide real applications, we perform classification experiments on electroencephalogram signals, demonstrating the effectiveness and robustness of the classifier in binary data classification.

SkipResNet: Crop and Weed Recognition Based on the Improved ResNet

Weeds have a detrimental effect on crop yield. However, the prevailing chemical weed control methods cause pollution of the ecosystem and land. Therefore, it has become a trend to reduce dependence on herbicides; realize a sustainable, intelligent weed control method; and protect the land. In order to realize intelligent weeding, efficient and accurate crop and weed recognition is necessary. Convolutional neural networks (CNNs) are widely applied for weed and crop recognition due to their high speed and efficiency. In this paper, a multi-path input skip-residual network (SkipResNet) was put forward to upgrade the classification function of weeds and crops. It improved the residual block in the ResNet model and combined three different path selection algorithms. Experiments showed that on the plant seedling dataset, our proposed network achieved an accuracy of 95.07%, which is 0.73%, 0.37%, and 4.75% better than that of ResNet18, VGG19, and MobileNetV2, respectively. The validation results on the weed–corn dataset also showed that the algorithm can provide more accurate identification of weeds and crops, thereby reducing land contamination during the weeding process. In addition, the algorithm is generalizable and can be used in image classification in agriculture and other fields.

Rethinking the loss function in image classification

Rethinking the loss function in image classification

The long-term post-surgical outcome of intermediate anorectal malformation in our department.

Posterior sagittal anorectoplasty and laparoscopic-assisted anorectal pull-through are preferred for anorectal malformation (ARM) today, while careful pull-through procedures with sacroperineal approach yield excellent outcomes. This study focuses on a pull-through procedure emphasizing continence mechanism preservation and compares outcomes with historical studies with various procedures. Bowel function of patients with intermediate ARM followed up for over 10years post-surgically was assessed. Data collected included ARM type with the Krickenbeck classification, comorbidities, complications, post-surgical examinations, follow-up, and bowel function at the latest clinic visit. The literature review collected original articles including more than 10 post-anorectoplasty cases which were followed for over 10years. Eleven cases were identified, with a median age at anorectoplasty and follow-up length of 6.9months and 14.4years. Two fistula recurrences required surgical treatment. Long-term incontinence and constipation were observed in 9% and 45% of the cohort, respectively. Good rectal angulation and a positive rectoanal inhibitory reflex were confirmed in most cases examined. A literature review identified eight studies with various outcome-measuring instruments. Outcomes of the introduced pull-through procedure were favorable, while the literature review highlights the variation in outcomes of various anorectoplasty. Level IV.

Activation function optimization scheme for image classification

Activation function has a significant impact on the dynamics, convergence, and performance of deep neural networks. The search for a consistent and high-performing activation function has always been a pursuit during deep learning model development. Existing state-of-the-art activation functions are manually designed with human expertise except for Swish. Swish was developed using a reinforcement learning-based search strategy. In this study, we propose an evolutionary approach for optimizing activation functions specifically for image classification tasks, aiming to discover functions that outperform current state-of-the-art options. Through this optimization framework, we obtain a series of high-performing activation functions denoted as Exponential Error Linear Unit (EELU). The developed activation functions are evaluated for image classification tasks from two perspectives: (1) five state-of-the-art neural network architectures, such as ResNet50, AlexNet, VGG16, MobileNet, and Compact Convolutional Transformer, which cover computationally heavy to light neural networks, and (2) eight standard datasets, including CIFAR10, Imagenette, MNIST, Fashion MNIST, Beans, Colorectal Histology, CottonWeedID15, and TinyImageNet which cover from typical machine vision benchmark, agricultural image applications to medical image applications. Finally, we statistically investigate the generalization of the resultant activation functions developed through the optimization scheme. With a Friedman test, we conclude that the optimization scheme is able to generate activation functions that outperform the existing standard ones in 92.8% cases among 28 different cases studied, and −x⋅erf(e−x) is found to be the best activation function for image classification generated by the optimization scheme.

Is a cork a legal shortcut? – A comparison of the measured and assumed amount of rotation in freestyle tricks

ABSTRACT In snowboard freestyle disciplines, the amount of rotation is commonly determined as the sum of rotations around all board axes and is the most important indicator of the trick difficulty across all snowboard freestyle disciplines. Based on the type of rotation, tricks can be classified as flatspins, corks and flips. It is not yet known whether the type of rotation of a trick can influence the actual amount of rotation. Therefore, the aim of this study was to determine the amount of deviation, defined as difference between measured and assumed amount of rotation as a function of trick classification, using kinematic motion analysis. The amount of deviation was positive for flatspins (median: 21°; min: −4°; max: 49°) and negative for corks (median: −25°; min: −89°; max: 12°) and flips (median: −28°; min: −94°; max: 13°). Our results demonstrate that there are ways of execution where riders perform corks and flips with a shortcut and flatspins with a detour. This should be taken into account by judges, coaches and riders. Further research is needed to investigate how the shortcut can be influenced.

Characterizing Solar Center-to-limb Radial-velocity Variability with SDO

Stellar photospheric inhomogeneities are a significant source of noise, which currently precludes the discovery of Earth-mass planets orbiting Sun-like stars with the radial-velocity (RV) method. To complement several previous studies that have used ground- and spaced-based facilities to characterize the RV of the Sun, here we characterize the center-to-limb variability (CLV) of solar RVs arising from various solar-surface inhomogeneities observed by the Solar Dynamics Observatory (SDO)/Helioseismic Magnetic Imager and SDO/Atmospheric Imaging Assembly. By using various SDO observables to classify pixels and calculate line-of-sight velocities as a function of pixel classification and limb angle, we show that each identified feature type, including the umbrae and penumbrae of sunspots, quiet-Sun magnetoconvective cells, magnetic network, and plage, exhibit distinct and complex CLV signatures, including a notable limb-angle dependence in the observed suppression of convective blueshift for magnetically active regions. We discuss the observed distributions of velocities by identified region type and limb angle, offer interpretations of the physical phenomena that shape these distributions, and emphasize the need to understand the RV signatures of these regions as astrophysical signals, rather than simple (un)correlated noise processes.

A C4 Software for Anti Drone System

Mini unmanned aerial vehicles (UAVs), commonly known as small drones, have seen outstanding advancements in recent years and have been used in a variety of fields. However, their potential misuse for illegal activities and the risks they pose to safety and privacy have raised concerns. To address these issues, we propose a Command, Control, Communications, and Computers (C4) software able to manage and control anti-drone systems. Our software solution includes an easy-to-use dashboard that processes and displays video data from surveillance sensors. It incorporates AI-powered functionalities, including object detection, target tracking, and classification of small drones. At the inference stage, the network models for drone detection and classification functionalities have achieved an accuracy exceeding 96 %. We have evaluated the effectiveness of our solution by deploying it in a no-fly zone, where it successfully identified and tracked drones in near real-time. The proposed control system provides unified information with which the entire anti-drone process can be managed starting from the detection of each threat.

Comprehensive analysis of pepper (Capsicum annuum) RAV genes family and functional identification of CaRAV1 under chilling stress

BackgroundDespite its known significance in plant abiotic stress responses, the role of the RAV gene family in the response of Capsicum annuum to chilling stress remains largely unexplored.ResultsIn this study, we identified and characterized six members of the CaRAV gene subfamily in pepper plants through genome-wide analysis. Subsequently, the CaRAV subfamily was classified into four branches based on homology with Arabidopsis thaliana, each exhibiting relatively conserved domains within the branch. We discovered that light response elements accounted for the majority of CaRAVs, whereas low-temperature response elements were specific to the NGA gene subfamily. After pepper plants were subjected to chilling stress, qRT‒PCR analysis revealed that CaRAV1, CaRAV2 and CaNGA1 were significantly induced in response to chilling stress, indicating that CaRAVs play a role in the response to chilling stress. Using virus-induced gene silencing (VIGS) vectors, we targeted key members of the CaRAV gene family. Under normal growth conditions, the MDA content and SOD enzyme activity of the silenced plants were slightly greater than those of the control plants, and the REC activity was significantly greater than that of the control plants. The levels of MDA and electrolyte leakage were greater in the silenced plants after they were exposed to chilling stress, and the POD and CAT enzyme activities were significantly lower than those in the control, which was particularly evident under repeated chilling stress. In addition, the relative expression of CaPOD and CaCAT was greater in V2 plants upon repeated chilling stress, especially CaCAT was significantly greater in V2 plants than in the other two silenced plants, with 3.29 and 1.10 increases within 12 and 24 h. These findings suggest that CaRAV1 and CaNGA1 positively regulate the response to chilling stress.ConclusionsSilencing of key members of the CaRAV gene family results in increased susceptibility to chilling damage and reduced antioxidant enzyme activity in plants, particularly under repeated chilling stress. This study provides valuable information for understanding the classification and putative functions of RAV transcription factors in pepper plants.