Based on the data of the cases of severe convection weather such as hail,thunderstorm(thunderstorm gale)and short-time heavy precipitation in recent 10 years,the spatial and temporal distribution characteristics of di...Based on the data of the cases of severe convection weather such as hail,thunderstorm(thunderstorm gale)and short-time heavy precipitation in recent 10 years,the spatial and temporal distribution characteristics of different types of severe convection weather were analyzed.The results show that the frequency of severe convection weather tended to increase,of which short-time heavy precipitation and thunderstorm weather rose,and hail and thunderstorm gale weather decreased.Severe convection weather began to extend in late spring and early autumn.Typical cases were selected to analyze the evolution mechanism,and the conceptual models of severe convective weather caused by cold advection forcing,warm advection forcing and baroclinic frontogenesis were obtained.The key predictors for the potential prediction of severe convection weather were proposed,such as CAPE(convective available potential energy)for hail weather,UH index(maximum ascending helicity)for thunderstorm gale and PWV(precipitable water vapor)for short-time heavy precipitation.ERA5 data were used to get the forecast threshold of the key factor of classified severe convection weather,and it was verified that the threshold was available.Meanwhile,the causes of the error of failure cases were analyzed.For instance,the larger deviation of CAPE was caused by the 2 m deviation of temperature.Supplementary correction method and threshold were given to provide a reference for the objective forecast and early warning of severe convection weather.展开更多
Driven by both the“new engineering”initiative and the energy revolution,the traditional engineering education model can hardly meet the demand of the energy and electric power industry for diversified and interdisci...Driven by both the“new engineering”initiative and the energy revolution,the traditional engineering education model can hardly meet the demand of the energy and electric power industry for diversified and interdisciplinary outstanding engineers.Based on the“industry-university-research-application”four-in-one collaborative education concept,this paper constructs a new training system centered on classified cultivation and classified evaluation.The system aims to solve core problems such as homogeneous training,disconnection between industry and academia,single evaluation method,and insufficient faculty.Through measures including modular courses,the dual-tutor system,and diversified practical platforms,it realizes differentiated and precise talent training,so as to deliver outstanding engineers with the ability to“define problems,break through technologies,and create value”for the energy and electric power industry.展开更多
Objective To develop and evaluate an automated system for digitizing audiograms,classifying hearing loss levels,and comparing their performance with traditional methods and otolaryngologists'interpretations.Design...Objective To develop and evaluate an automated system for digitizing audiograms,classifying hearing loss levels,and comparing their performance with traditional methods and otolaryngologists'interpretations.Designed and Methods We conducted a retrospective diagnostic study using 1,959 audiogram images from patients aged 7 years and older at the Faculty of Medicine,Vajira Hospital,Navamindradhiraj University.We employed an object detection approach to digitize audiograms and developed multiple machine learning models to classify six hearing loss levels.The dataset was split into 70%training(1,407 images)and 30%testing(352 images)sets.We compared our model's performance with classifications based on manually extracted audiogram values and otolaryngologists'interpretations.Result Our object detection-based model achieved an F1-score of 94.72%in classifying hearing loss levels,comparable to the 96.43%F1-score obtained using manually extracted values.The Light Gradient Boosting Machine(LGBM)model is used as the classifier for the manually extracted data,which achieved top performance with 94.72%accuracy,94.72%f1-score,94.72 recall,and 94.72 precision.In object detection based model,The Random Forest Classifier(RFC)model showed the highest 96.43%accuracy in predicting hearing loss level,with a F1-score of 96.43%,recall of 96.43%,and precision of 96.45%.Conclusion Our proposed automated approach for audiogram digitization and hearing loss classification performs comparably to traditional methods and otolaryngologists'interpretations.This system can potentially assist otolaryngologists in providing more timely and effective treatment by quickly and accurately classifying hearing loss.展开更多
Background:Stomach cancer(SC)is one of the most lethal malignancies worldwide due to late-stage diagnosis and limited treatment.The transcriptomic,epigenomic,and proteomic,etc.,omics datasets generated by high-through...Background:Stomach cancer(SC)is one of the most lethal malignancies worldwide due to late-stage diagnosis and limited treatment.The transcriptomic,epigenomic,and proteomic,etc.,omics datasets generated by high-throughput sequencing technology have become prominent in biomedical research,and they reveal molecular aspects of cancer diagnosis and therapy.Despite the development of advanced sequencing technology,the presence of high-dimensionality in multi-omics data makes it challenging to interpret the data.Methods:In this study,we introduce RankXLAN,an explainable ensemble-based multi-omics framework that integrates feature selection(FS),ensemble learning,bioinformatics,and in-silico validation for robust biomarker detection,potential therapeutic drug-repurposing candidates’identification,and classification of SC.To enhance the interpretability of the model,we incorporated explainable artificial intelligence(SHapley Additive exPlanations analysis),as well as accuracy,precision,F1-score,recall,cross-validation,specificity,likelihood ratio(LR)+,LR−,and Youden index results.Results:The experimental results showed that the top four FS algorithms achieved improved results when applied to the ensemble learning classification model.The proposed ensemble model produced an area under the curve(AUC)score of 0.994 for gene expression,0.97 for methylation,and 0.96 for miRNA expression data.Through the integration of bioinformatics and ML approach of the transcriptomic and epigenomic multi-omics dataset,we identified potential marker genes,namely,UBE2D2,HPCAL4,IGHA1,DPT,and FN3K.In-silico molecular docking revealed a strong binding affinity between ANKRD13C and the FDA-approved drug Everolimus(binding affinity−10.1 kcal/mol),identifying ANKRD13C as a potential therapeutic drug-repurposing target for SC.Conclusion:The proposed framework RankXLAN outperforms other existing frameworks for serum biomarker identification,therapeutic target identification,and SC classification with multi-omics datasets.展开更多
An Extended Kalman Filter（EKF） is commonly used to fuse raw Global Navigation Satellite System（GNSS） measurements and Inertial Navigation System（INS） derived measurements. However, the Conventional EKF（CEKF） s...An Extended Kalman Filter（EKF） is commonly used to fuse raw Global Navigation Satellite System（GNSS） measurements and Inertial Navigation System（INS） derived measurements. However, the Conventional EKF（CEKF） suffers the problem for which the uncertainty of the statistical properties to dynamic and measurement models will degrade the performance.In this research, an Adaptive Interacting Multiple Model（AIMM） filter is developed to enhance performance. The soft-switching property of Interacting Multiple Model（IMM） algorithm allows the adaptation between two levels of process noise, namely lower and upper bounds of the process noise. In particular, the Sage adaptive filtering is applied to adapt the measurement covariance on line. In addition, a classified measurement update strategy is utilized, which updates the pseudorange and Doppler observations sequentially. A field experiment was conducted to validate the proposed algorithm, the pseudorange and Doppler observations from Global Positioning System（GPS） and Bei Dou Navigation Satellite System（BDS） were post-processed in differential mode.The results indicate that decimeter-level positioning accuracy is achievable with AIMM for GPS/INS and GPS/BDS/INS configurations, and the position accuracy is improved by 35.8%, 34.3% and 33.9% for north, east and height components, respectively, compared to the CEKF counterpartfor GPS/BDS/INS. Degraded performance for BDS/INS is obtained due to the lower precision of BDS pseudorange observations.展开更多
The participation of ordinary devices in networking has created a world of connected devices rapidly.The Internet of Things(IoT)includes heterogeneous devices from every field.There are no definite protocols or standa...The participation of ordinary devices in networking has created a world of connected devices rapidly.The Internet of Things(IoT)includes heterogeneous devices from every field.There are no definite protocols or standards for IoT communication,and most of the IoT devices have limited resources.Enabling a complete security measure for such devices is a challenging task,yet necessary.Many lightweight security solutions have surfaced lately for IoT.The lightweight security protocols are unable to provide an optimum protection against prevailing powerful threats in cyber world.It is also hard to deploy any traditional security protocol on resource-constrained IoT devices.Software-defined networking introduces a centralized control in computer networks.SDN has a programmable approach towards networking that decouples control and data planes.An SDN-based intrusion detection system is proposed which uses deep learning classifier for detection of anomalies in IoT.The proposed intrusion detection system does not burden the IoT devices with security profiles.The proposed work is executed on the simulated environment.The results of the simulation test are evaluated using various matrices and compared with other relevant methods.展开更多
The cervical spine injury represents a potential devastating disease with 6% associated in-hospital mortality （lain et al., 2015）. Neurological deterioration ranging from complete spinal cord injury （SCI） to incom...The cervical spine injury represents a potential devastating disease with 6% associated in-hospital mortality （lain et al., 2015）. Neurological deterioration ranging from complete spinal cord injury （SCI） to incomplete SCI or single radiculopathy are potential consequences of the blunt trauma over this region. The subaxial cervical spine accounts the vast majority of cervical injuries, making up two thirds of all cervical fractures （Alday, 1996）. Few classifications （Holdsworth, 1970; White et al., 1975; Mien et al., 1982; Denis, 1984; Vaccaro et al., 2007） have been proposed to describe injuries of the cervical spine for several reasons. First, to delineate the best treatment in each case; second, to determinate an accurate neurological prognosis, and third, to establish a standard way to communicate and describe specific characteristics of cervical injuries patterns. Classical systems are primarily descriptive and no single system has gained widespread use, largely because of restrictions in clinical relevance and its complexity.展开更多
In the era of Big data,learning discriminant feature representation from network traffic is identified has as an invariably essential task for improving the detection ability of an intrusion detection system(IDS).Owin...In the era of Big data,learning discriminant feature representation from network traffic is identified has as an invariably essential task for improving the detection ability of an intrusion detection system(IDS).Owing to the lack of accurately labeled network traffic data,many unsupervised feature representation learning models have been proposed with state-of-theart performance.Yet,these models fail to consider the classification error while learning the feature representation.Intuitively,the learnt feature representation may degrade the performance of the classification task.For the first time in the field of intrusion detection,this paper proposes an unsupervised IDS model leveraging the benefits of deep autoencoder(DAE)for learning the robust feature representation and one-class support vector machine(OCSVM)for finding the more compact decision hyperplane for intrusion detection.Specially,the proposed model defines a new unified objective function to minimize the reconstruction and classification error simultaneously.This unique contribution not only enables the model to support joint learning for feature representation and classifier training but also guides to learn the robust feature representation which can improve the discrimination ability of the classifier for intrusion detection.Three set of evaluation experiments are conducted to demonstrate the potential of the proposed model.First,the ablation evaluation on benchmark dataset,NSL-KDD validates the design decision of the proposed model.Next,the performance evaluation on recent intrusion dataset,UNSW-NB15 signifies the stable performance of the proposed model.Finally,the comparative evaluation verifies the efficacy of the proposed model against recently published state-of-the-art methods.展开更多
An important problem in wireless communication networks (WCNs) is that they have a minimum number of resources, which leads to high-security threats. An approach to find and detect the attacks is the intrusion detecti...An important problem in wireless communication networks (WCNs) is that they have a minimum number of resources, which leads to high-security threats. An approach to find and detect the attacks is the intrusion detection system (IDS). In this paper, the fuzzy lion Bayes system (FLBS) is proposed for intrusion detection mechanism. Initially, the data set is grouped into a number of clusters by the fuzzy clustering algorithm. Here, the Naive Bayes classifier is integrated with the lion optimization algorithm and the new lion naive Bayes (LNB) is created for optimally generating the probability measures. Then, the LNB model is applied to each data group, and the aggregated data is generated. After generating the aggregated data, the LNB model is applied to the aggregated data, and the abnormal nodes are identified based on the posterior probability function. The performance of the proposed FLBS system is evaluated using the KDD Cup 99 data and the comparative analysis is performed by the existing methods for the evaluation metrics accuracy and false acceptance rate (FAR). From the experimental results, it can be shown that the proposed system has the maximum performance, which shows the effectiveness of the proposed system in the intrusion detection.展开更多
To improve the identification for visual defect of TFF-LCD, a new machine vision system is proposed, which is superior to human eye inspection. The system respectively employs a CCD camera to capture the image of TFT-...To improve the identification for visual defect of TFF-LCD, a new machine vision system is proposed, which is superior to human eye inspection. The system respectively employs a CCD camera to capture the image of TFT-LCD panel and an image processing system to identify potential visual defects. Image pre-processing, such as average filtering and geometric correction, was performed on the captured image, and then a candidate area of defect was segmented from the background. Feature information extracted from the area of interest entered a fuzzy rule-based classifier that simulated the defect inspection of TFT-LCD undertaken by experienced technicians. Experiment results show that the machine vision system can obtain fast, objective and accurate inspection compared with subjective and inaccurate human eye inspection.展开更多
This paper demonstrates a Geographic Information Systems （GIS） procedure of classifying and mapping forest management category in Baihe Forestry Burea, Jilin Province, China. Within the study area, Baihe Forestry Bu...This paper demonstrates a Geographic Information Systems （GIS） procedure of classifying and mapping forest management category in Baihe Forestry Burea, Jilin Province, China. Within the study area, Baihe Forestry Bureau land was classified into a two-hierarchy system. The top-level class included the non-forest and forest. Over 96% of land area is forest in the study area, which was further divided into key ecological service forest （KES）, general ecological service forest （GES）, and commodity forest （COM）. COM covered 45.0% of the total land area and was the major forest management type in Baihe Forest Bureau. KES and GES accounted for 21.2% and 29.9% of the total land area, respectively. The forest management zones designed with GIS in this study were then compared with the forest management zones established using the hand draw by the local agency. There were obvious differences between the two products. It suggested that the differences had some to do with the data sources, basic unit and mapping procedures. It also suggested that the GIS method was a useful tool in integrating forest inventory data and other data for classifying and mapping forest zones to meet the needs of the classified forest management system.展开更多
In this paper, based on hourly precipitation observations in 1977e2013 in the Beijing area, China, hourly precipitation in summer (June?August) is classified into three categories: light (below the 50th percentile val...In this paper, based on hourly precipitation observations in 1977e2013 in the Beijing area, China, hourly precipitation in summer (June?August) is classified into three categories: light (below the 50th percentile values), moderate (the 50th to 95th percentile values), and heavy (above the 95th percentile values). Results reveal that both light and moderate precipitation decreased significantly during the research period and thereby caused the decrease in summer totals. By contrast, pronounced trends failed to be detected in the heavy category. Since 2004, the contribution of heavy rainfall to the summer total precipitation in the urban area increased as compared to the suburban area, which is opposite to light rainfall. There are obvious differences in the diurnal variations of classified precipitation. Light precipitation shows a double peak structure in the early morning and at night, while moderate and heavy rainfall show a single peak at night. Light precipitation at the early morning peak time decreased significantly in the whole Beijing area. Compared with the suburban area, light precipitation in the urban area occurred less frequently whereas heavy precipitation occurred more frequently at evening peak time after 2004. The asymmetry of the rainfall is obvious, especially, for heavy precipitation. The asymmetry of heavy precipitation events in the urban area exhibits a significant increasing trend.展开更多
The nature of the slurry from the stone-crushing and sand-making processes is analyzed to develop a novel separation process.The process comprises hydro-cyclone separation followed by screening of the fines,clarificat...The nature of the slurry from the stone-crushing and sand-making processes is analyzed to develop a novel separation process.The process comprises hydro-cyclone separation followed by screening of the fines,clarification,and filtration.Recovering fine sand and clean wastewater for recycle is demonstrated.The +0.045 mm fine sand fraction and à0.045 mm ultra-fine clay in the slurry are separated and recovered.Fine sand that was previously lost and wasted is now recoverable.The cleaned and reused water is as much as 94% of the total.展开更多
The key objective of intrusion detection systems(IDS)is to protect the particular host or network by investigating and predicting the network traffic as an attack or normal.These IDS uses many methods of machine learn...The key objective of intrusion detection systems(IDS)is to protect the particular host or network by investigating and predicting the network traffic as an attack or normal.These IDS uses many methods of machine learning(ML)to learn from pastexperience attack i.e.signatures based and identify the new ones.Even though these methods are effective,but they have to suffer from large computational costs due to considering all the traffic features,together.Moreover,emerging technologies like the Internet of Things(Io T),big data,etc.are getting advanced day by day;as a result,network traffics are also increasing rapidly.Therefore,the issue of computational cost needs to be addressed properly.Thus,in this research,firstly,the ML methods have been used with the feature selection technique(FST)to reduce the number of features by picking out only the important ones from NSL-KDD,CICIDS2017,and CIC-DDo S2019datasets later that helped to build IDSs with lower cost but with the higher performance which would be appropriate for vast scale network.The experimental result demonstrated that the proposed model i.e.Decision tree(DT)with Recursive feature elimination(RFE)performs better than other classifiers with RFE in terms of accuracy,specificity,precision,sensitivity,F1-score,and G-means on the investigated datasets.展开更多
Currently, cybersecurity and cyber resilience are emerging and urgent issues in nextgeneration air traffic surveillance systems, which depend primarily on Automatic Dependent Surveillance-Broadcast(ADS-B) owing to its...Currently, cybersecurity and cyber resilience are emerging and urgent issues in nextgeneration air traffic surveillance systems, which depend primarily on Automatic Dependent Surveillance-Broadcast(ADS-B) owing to its low cost and high accuracy. Unfortunately, ADS-B is prone to cyber-attacks. To verify the ADS-B positioning data of aircraft, multilateration(MLAT)techniques that use Time Differences of Arrivals(TDoAs) have been proposed. MLAT exhibits low accuracy in determining aircraft positions. Recently, a novel technique using a theoretically calculated TDoA fingerprint map has been proposed. This technique is less dependent on the geometry of sensor deployment and achieves better accuracy than MLAT. However, the accuracy of the existing technique is not sufficiently precise for determining aircraft positions and requires a long computation time. In contrast, this paper presents a reliable surveillance framework using an Actual TDoA-Based Augmentation System(ATBAS). It uses historically recorded real-data from the OpenSky network to train our TDoA fingerprint grid network. Our results show that the accuracy of the proposed ATBAS framework in determining the aircraft positions is significantly better than those of the MLAT and expected TDoA techniques by 56.93% and 48.86%, respectively. Additionally, the proposed framework reduced the computation time by 77% compared with the expected TDoA technique.展开更多
Recently,the spread of COVID-19 virus infection and the increase of people number with chronic diseases have attracted great attention all over the world.The detection and control of such diseases based on patient dem...Recently,the spread of COVID-19 virus infection and the increase of people number with chronic diseases have attracted great attention all over the world.The detection and control of such diseases based on patient demographic data are considered to be a major problem.The key issue in the solution to these problems is the development of methods and algorithms to forecast wellness and categorise patients according to their healthy and unhealthy states.In this paper,a comprehensive analysis of machine learning approaches in the field of diagnosing COVID-19 has been conducted,and for the detection of chronic diseases in patients,to identify symptoms of COVID-19 virus infection in advance,and control the situation a healthcare system has been proposed.The constructed system provides real-time monitoring of chronic diseases and COVID-19 virus infection in patients.The proposed system consists of five layers:IoT sensor layer,Data transmission layer,Fog layer,Cloud layer,the Application layer.The system architecture in the Fog layer uses machine learning and deep learning algorithms to diagnose patients'diseases,to generate and send diagnostic and emergency alerts to users.The classification module of the system's Fog layer categorises the patient's health status into healthy and unhealthy classes.In this module,to classify medical data the Decision Tree,Random Forest,SVM,Gradient Boosting,Logistic Regression algorithms are used.The COVID-19 dataset is used to test the effectiveness of the methods.The best results from the comparative analysis of the methods are obtained from the Decision Tree,Random Forest,and Gradient Boosting algorithms,which are recognised data points with high accuracy and on the accuracy metric reached 1.0,0.99,1.0 values,respectively.The classification of the other two SVM and Logistic Regression algorithms provided the worst results,and the accuracy score of both classifiers obtained a 0.89 value.展开更多
Automated grading of colon biopsy images across all magnifications is challenging because of tailored segmentation and dependent features on each magnification.This work presents a novel approach of robust magnificati...Automated grading of colon biopsy images across all magnifications is challenging because of tailored segmentation and dependent features on each magnification.This work presents a novel approach of robust magnification-independent colon cancer grading framework to distinguish colon biopsy images into four classes:normal,well,moderate,and poor.The contribution of this research is to develop a magnification invariant hybrid feature set comprising cartoon feature,Gabor wavelet,wavelet moments,HSV histogram,color auto-correlogram,color moments,and morphological features that can be used to characterize different grades.Besides,the classifier is modeled as a multiclass structure with six binary class Bayesian optimized random forest(BO-RF)classifiers.This study uses four datasets(two collected from Indian hospitals—Ishita Pathology Center(IPC)of 4X,10X,and 40X and Aster Medcity(AMC)of 10X,20X,and 40X—two benchmark datasets—gland segmentation(GlaS)of 20X and IMEDIATREAT of 10X)comprising multiple microscope magnifications.Experimental results demonstrate that the proposed method outperforms the other methods used for colon cancer grading in terms of accuracy(97.25%-IPC,94.40%-AMC,97.58%-GlaS,99.16%-Imediatreat),sensitivity(0.9725-IPC,0.9440-AMC,0.9807-GlaS,0.9923-Imediatreat),specificity(0.9908-IPC,0.9813-AMC,0.9907-GlaS,0.9971-Imediatreat)and F-score(0.9725-IPC,0.9441-AMC,0.9780-GlaS,0.9923-Imediatreat).The generalizability of the model to any magnified input image is validated by training in one dataset and testing in another dataset,highlighting strong concordance in multiclass classification and evidencing its effective use in the first level of automatic biopsy grading and second opinion.展开更多
This paper presents a vision-based fingertip-writing character recognition system. The overall system is implemented through a CMOS image camera on a FPGA chip. A blue cover is mounted on the top of a finger to simpli...This paper presents a vision-based fingertip-writing character recognition system. The overall system is implemented through a CMOS image camera on a FPGA chip. A blue cover is mounted on the top of a finger to simplify fingertip detection and to enhance recognition accuracy. For each character stroke, 8 sample points (including start and end points) are recorded. 7 tangent angles between consecutive sampled points are also recorded as features. In addition, 3 features angles are extracted: angles of the triangle consisting of the start point, end point and average point of all (8 total) sampled points. According to these key feature angles, a simple template matching K-nearest-neighbor classifier is applied to distinguish each character stroke. Experimental result showed that the system can successfully recognize fingertip-writing character strokes of digits and small lower case letter alphabets with an accuracy of almost 100%. Overall, the proposed finger-tip-writing recognition system provides an easy-to-use and accurate visual character input method.展开更多
Land cover classification of mountainous environments continues to be a challenging remote sensing problem,owing to landscape complexities exhibited by the region.This study explored a multiple classifier system(MCS)a...Land cover classification of mountainous environments continues to be a challenging remote sensing problem,owing to landscape complexities exhibited by the region.This study explored a multiple classifier system(MCS)approach to the classification of mountain land cover for the Khumbu region in the Himalayas using Sentinel-2 images and a cloud-based model framework.The relationship between classification accuracy and MCS diversity was investigated,and the effects of different diversification and combination methods on MCS classification performance were comparatively assessed for this environment.We present ten MCS models that implement a homogeneous ensemble approach,using the high performing Random Forest(RF)algorithm as the selected classifier.The base classifiers of each MCS model were developed using different combinations of three diversity techniques:(1)distinct training sets,(2)Mean Decrease Accuracy feature selection,and(3)‘One-vs-All’problem reduction.The base classifier predictions of each RFMCS model were combined using:(1)majority vote,(2)weighted argmax,and(3)a meta RF classifier.All MCS models reported higher classification accuracies than the benchmark classifier(overall accuracy with 95% confidence interval:87.33%±0.97%),with the highest performing model reporting an overall accuracy(±95% confidence interval)of 90.95%±0.84%.Our key findings include:(1)MCS is effective in mountainous environments prone to noise from landscape complexities,(2)problem reduction is indicated as a stronger method over feature selection in improving the diversity of the MCS,(3)although the MCS diversity and accuracy have a positive correlation,our results suggest this is a weak relationship for mountainous classifications,and(4)the selected diversity methods improve the discriminability of MCS against vegetation and forest classes in mountainous land cover classifications and exhibit a cumulative effect on MCS diversity for this context.展开更多
A new remote sensing image coding scheme based on the wavelet transform and classified vector quantization (CVQ) is proposed. The original image is first decomposed into a hierarchy of 3 layers including 10 subimages ...A new remote sensing image coding scheme based on the wavelet transform and classified vector quantization (CVQ) is proposed. The original image is first decomposed into a hierarchy of 3 layers including 10 subimages by DWT. The lowest frequency subimage is compressed by scalar quantization and ADPCM. The high frequency subimages are compressed by CVQ to utilize the similarity among different resolutions while improving the edge quality and reducing computational complexity. The experimental results show that the proposed scheme has a better performance than JPEG, and a PSNR of reconstructed image is 31～33 dB with a rate of 0.2 bpp.展开更多
基金Supported by the Open-end Funds of Key Laboratory for Disaster Prevention and Mitigation of Qinghai Province(QFZ-2021-Z04)。
文摘Based on the data of the cases of severe convection weather such as hail,thunderstorm(thunderstorm gale)and short-time heavy precipitation in recent 10 years,the spatial and temporal distribution characteristics of different types of severe convection weather were analyzed.The results show that the frequency of severe convection weather tended to increase,of which short-time heavy precipitation and thunderstorm weather rose,and hail and thunderstorm gale weather decreased.Severe convection weather began to extend in late spring and early autumn.Typical cases were selected to analyze the evolution mechanism,and the conceptual models of severe convective weather caused by cold advection forcing,warm advection forcing and baroclinic frontogenesis were obtained.The key predictors for the potential prediction of severe convection weather were proposed,such as CAPE(convective available potential energy)for hail weather,UH index(maximum ascending helicity)for thunderstorm gale and PWV(precipitable water vapor)for short-time heavy precipitation.ERA5 data were used to get the forecast threshold of the key factor of classified severe convection weather,and it was verified that the threshold was available.Meanwhile,the causes of the error of failure cases were analyzed.For instance,the larger deviation of CAPE was caused by the 2 m deviation of temperature.Supplementary correction method and threshold were given to provide a reference for the objective forecast and early warning of severe convection weather.
文摘Driven by both the“new engineering”initiative and the energy revolution,the traditional engineering education model can hardly meet the demand of the energy and electric power industry for diversified and interdisciplinary outstanding engineers.Based on the“industry-university-research-application”four-in-one collaborative education concept,this paper constructs a new training system centered on classified cultivation and classified evaluation.The system aims to solve core problems such as homogeneous training,disconnection between industry and academia,single evaluation method,and insufficient faculty.Through measures including modular courses,the dual-tutor system,and diversified practical platforms,it realizes differentiated and precise talent training,so as to deliver outstanding engineers with the ability to“define problems,break through technologies,and create value”for the energy and electric power industry.
文摘Objective To develop and evaluate an automated system for digitizing audiograms,classifying hearing loss levels,and comparing their performance with traditional methods and otolaryngologists'interpretations.Designed and Methods We conducted a retrospective diagnostic study using 1,959 audiogram images from patients aged 7 years and older at the Faculty of Medicine,Vajira Hospital,Navamindradhiraj University.We employed an object detection approach to digitize audiograms and developed multiple machine learning models to classify six hearing loss levels.The dataset was split into 70%training(1,407 images)and 30%testing(352 images)sets.We compared our model's performance with classifications based on manually extracted audiogram values and otolaryngologists'interpretations.Result Our object detection-based model achieved an F1-score of 94.72%in classifying hearing loss levels,comparable to the 96.43%F1-score obtained using manually extracted values.The Light Gradient Boosting Machine(LGBM)model is used as the classifier for the manually extracted data,which achieved top performance with 94.72%accuracy,94.72%f1-score,94.72 recall,and 94.72 precision.In object detection based model,The Random Forest Classifier(RFC)model showed the highest 96.43%accuracy in predicting hearing loss level,with a F1-score of 96.43%,recall of 96.43%,and precision of 96.45%.Conclusion Our proposed automated approach for audiogram digitization and hearing loss classification performs comparably to traditional methods and otolaryngologists'interpretations.This system can potentially assist otolaryngologists in providing more timely and effective treatment by quickly and accurately classifying hearing loss.
基金the Deanship of Research and Graduate Studies at King Khalid University,KSA,for funding this work through the Large Research Project under grant number RGP2/164/46.
文摘Background:Stomach cancer(SC)is one of the most lethal malignancies worldwide due to late-stage diagnosis and limited treatment.The transcriptomic,epigenomic,and proteomic,etc.,omics datasets generated by high-throughput sequencing technology have become prominent in biomedical research,and they reveal molecular aspects of cancer diagnosis and therapy.Despite the development of advanced sequencing technology,the presence of high-dimensionality in multi-omics data makes it challenging to interpret the data.Methods:In this study,we introduce RankXLAN,an explainable ensemble-based multi-omics framework that integrates feature selection(FS),ensemble learning,bioinformatics,and in-silico validation for robust biomarker detection,potential therapeutic drug-repurposing candidates’identification,and classification of SC.To enhance the interpretability of the model,we incorporated explainable artificial intelligence(SHapley Additive exPlanations analysis),as well as accuracy,precision,F1-score,recall,cross-validation,specificity,likelihood ratio(LR)+,LR−,and Youden index results.Results:The experimental results showed that the top four FS algorithms achieved improved results when applied to the ensemble learning classification model.The proposed ensemble model produced an area under the curve(AUC)score of 0.994 for gene expression,0.97 for methylation,and 0.96 for miRNA expression data.Through the integration of bioinformatics and ML approach of the transcriptomic and epigenomic multi-omics dataset,we identified potential marker genes,namely,UBE2D2,HPCAL4,IGHA1,DPT,and FN3K.In-silico molecular docking revealed a strong binding affinity between ANKRD13C and the FDA-approved drug Everolimus(binding affinity−10.1 kcal/mol),identifying ANKRD13C as a potential therapeutic drug-repurposing target for SC.Conclusion:The proposed framework RankXLAN outperforms other existing frameworks for serum biomarker identification,therapeutic target identification,and SC classification with multi-omics datasets.
基金co-supported by the National Key Research and Development Program of China(No.2016YFC0803103)Beijing Advanced Innovation Center for Future Urban Design(No.UDC2016050100)Beijing Postdoctoral Research Foundation
文摘An Extended Kalman Filter（EKF） is commonly used to fuse raw Global Navigation Satellite System（GNSS） measurements and Inertial Navigation System（INS） derived measurements. However, the Conventional EKF（CEKF） suffers the problem for which the uncertainty of the statistical properties to dynamic and measurement models will degrade the performance.In this research, an Adaptive Interacting Multiple Model（AIMM） filter is developed to enhance performance. The soft-switching property of Interacting Multiple Model（IMM） algorithm allows the adaptation between two levels of process noise, namely lower and upper bounds of the process noise. In particular, the Sage adaptive filtering is applied to adapt the measurement covariance on line. In addition, a classified measurement update strategy is utilized, which updates the pseudorange and Doppler observations sequentially. A field experiment was conducted to validate the proposed algorithm, the pseudorange and Doppler observations from Global Positioning System（GPS） and Bei Dou Navigation Satellite System（BDS） were post-processed in differential mode.The results indicate that decimeter-level positioning accuracy is achievable with AIMM for GPS/INS and GPS/BDS/INS configurations, and the position accuracy is improved by 35.8%, 34.3% and 33.9% for north, east and height components, respectively, compared to the CEKF counterpartfor GPS/BDS/INS. Degraded performance for BDS/INS is obtained due to the lower precision of BDS pseudorange observations.
基金The authors are grateful to MANF UGC,Government of India,for providing financial support under MANF-UGC(MANF-2015-17-JAM-60,506)programme to carry out this work.
文摘The participation of ordinary devices in networking has created a world of connected devices rapidly.The Internet of Things(IoT)includes heterogeneous devices from every field.There are no definite protocols or standards for IoT communication,and most of the IoT devices have limited resources.Enabling a complete security measure for such devices is a challenging task,yet necessary.Many lightweight security solutions have surfaced lately for IoT.The lightweight security protocols are unable to provide an optimum protection against prevailing powerful threats in cyber world.It is also hard to deploy any traditional security protocol on resource-constrained IoT devices.Software-defined networking introduces a centralized control in computer networks.SDN has a programmable approach towards networking that decouples control and data planes.An SDN-based intrusion detection system is proposed which uses deep learning classifier for detection of anomalies in IoT.The proposed intrusion detection system does not burden the IoT devices with security profiles.The proposed work is executed on the simulated environment.The results of the simulation test are evaluated using various matrices and compared with other relevant methods.
文摘The cervical spine injury represents a potential devastating disease with 6% associated in-hospital mortality （lain et al., 2015）. Neurological deterioration ranging from complete spinal cord injury （SCI） to incomplete SCI or single radiculopathy are potential consequences of the blunt trauma over this region. The subaxial cervical spine accounts the vast majority of cervical injuries, making up two thirds of all cervical fractures （Alday, 1996）. Few classifications （Holdsworth, 1970; White et al., 1975; Mien et al., 1982; Denis, 1984; Vaccaro et al., 2007） have been proposed to describe injuries of the cervical spine for several reasons. First, to delineate the best treatment in each case; second, to determinate an accurate neurological prognosis, and third, to establish a standard way to communicate and describe specific characteristics of cervical injuries patterns. Classical systems are primarily descriptive and no single system has gained widespread use, largely because of restrictions in clinical relevance and its complexity.
基金This work was supported by the Research Deanship of Prince Sattam Bin Abdulaziz University,Al-Kharj,Saudi Arabia(Grant No.2020/01/17215).Also,the author thanks Deanship of college of computer engineering and sciences for technical support provided to complete the project successfully。
文摘In the era of Big data,learning discriminant feature representation from network traffic is identified has as an invariably essential task for improving the detection ability of an intrusion detection system(IDS).Owing to the lack of accurately labeled network traffic data,many unsupervised feature representation learning models have been proposed with state-of-theart performance.Yet,these models fail to consider the classification error while learning the feature representation.Intuitively,the learnt feature representation may degrade the performance of the classification task.For the first time in the field of intrusion detection,this paper proposes an unsupervised IDS model leveraging the benefits of deep autoencoder(DAE)for learning the robust feature representation and one-class support vector machine(OCSVM)for finding the more compact decision hyperplane for intrusion detection.Specially,the proposed model defines a new unified objective function to minimize the reconstruction and classification error simultaneously.This unique contribution not only enables the model to support joint learning for feature representation and classifier training but also guides to learn the robust feature representation which can improve the discrimination ability of the classifier for intrusion detection.Three set of evaluation experiments are conducted to demonstrate the potential of the proposed model.First,the ablation evaluation on benchmark dataset,NSL-KDD validates the design decision of the proposed model.Next,the performance evaluation on recent intrusion dataset,UNSW-NB15 signifies the stable performance of the proposed model.Finally,the comparative evaluation verifies the efficacy of the proposed model against recently published state-of-the-art methods.
文摘An important problem in wireless communication networks (WCNs) is that they have a minimum number of resources, which leads to high-security threats. An approach to find and detect the attacks is the intrusion detection system (IDS). In this paper, the fuzzy lion Bayes system (FLBS) is proposed for intrusion detection mechanism. Initially, the data set is grouped into a number of clusters by the fuzzy clustering algorithm. Here, the Naive Bayes classifier is integrated with the lion optimization algorithm and the new lion naive Bayes (LNB) is created for optimally generating the probability measures. Then, the LNB model is applied to each data group, and the aggregated data is generated. After generating the aggregated data, the LNB model is applied to the aggregated data, and the abnormal nodes are identified based on the posterior probability function. The performance of the proposed FLBS system is evaluated using the KDD Cup 99 data and the comparative analysis is performed by the existing methods for the evaluation metrics accuracy and false acceptance rate (FAR). From the experimental results, it can be shown that the proposed system has the maximum performance, which shows the effectiveness of the proposed system in the intrusion detection.
文摘To improve the identification for visual defect of TFF-LCD, a new machine vision system is proposed, which is superior to human eye inspection. The system respectively employs a CCD camera to capture the image of TFT-LCD panel and an image processing system to identify potential visual defects. Image pre-processing, such as average filtering and geometric correction, was performed on the captured image, and then a candidate area of defect was segmented from the background. Feature information extracted from the area of interest entered a fuzzy rule-based classifier that simulated the defect inspection of TFT-LCD undertaken by experienced technicians. Experiment results show that the machine vision system can obtain fast, objective and accurate inspection compared with subjective and inaccurate human eye inspection.
基金Foundation project： This research was jointly supported by the National Natural Science Foundation of China （70373044＆30470302）, China＇s Ministry of Science and Technology （04EFN216600328）, and Northeast Rejuvenation Program of the Chinese Academy of Sciences.
文摘This paper demonstrates a Geographic Information Systems （GIS） procedure of classifying and mapping forest management category in Baihe Forestry Burea, Jilin Province, China. Within the study area, Baihe Forestry Bureau land was classified into a two-hierarchy system. The top-level class included the non-forest and forest. Over 96% of land area is forest in the study area, which was further divided into key ecological service forest （KES）, general ecological service forest （GES）, and commodity forest （COM）. COM covered 45.0% of the total land area and was the major forest management type in Baihe Forest Bureau. KES and GES accounted for 21.2% and 29.9% of the total land area, respectively. The forest management zones designed with GIS in this study were then compared with the forest management zones established using the hand draw by the local agency. There were obvious differences between the two products. It suggested that the differences had some to do with the data sources, basic unit and mapping procedures. It also suggested that the GIS method was a useful tool in integrating forest inventory data and other data for classifying and mapping forest zones to meet the needs of the classified forest management system.
基金This work was supported by the National Natural Science Foundation of China (41575094) and Special Scientific Research Fund of Meteorological Public Welfare Profession of China (GYHY201506014).
文摘In this paper, based on hourly precipitation observations in 1977e2013 in the Beijing area, China, hourly precipitation in summer (June?August) is classified into three categories: light (below the 50th percentile values), moderate (the 50th to 95th percentile values), and heavy (above the 95th percentile values). Results reveal that both light and moderate precipitation decreased significantly during the research period and thereby caused the decrease in summer totals. By contrast, pronounced trends failed to be detected in the heavy category. Since 2004, the contribution of heavy rainfall to the summer total precipitation in the urban area increased as compared to the suburban area, which is opposite to light rainfall. There are obvious differences in the diurnal variations of classified precipitation. Light precipitation shows a double peak structure in the early morning and at night, while moderate and heavy rainfall show a single peak at night. Light precipitation at the early morning peak time decreased significantly in the whole Beijing area. Compared with the suburban area, light precipitation in the urban area occurred less frequently whereas heavy precipitation occurred more frequently at evening peak time after 2004. The asymmetry of the rainfall is obvious, especially, for heavy precipitation. The asymmetry of heavy precipitation events in the urban area exhibits a significant increasing trend.
基金financially supported by Xinkaiyuan Crushed Stones Co. Ltd
文摘The nature of the slurry from the stone-crushing and sand-making processes is analyzed to develop a novel separation process.The process comprises hydro-cyclone separation followed by screening of the fines,clarification,and filtration.Recovering fine sand and clean wastewater for recycle is demonstrated.The +0.045 mm fine sand fraction and à0.045 mm ultra-fine clay in the slurry are separated and recovered.Fine sand that was previously lost and wasted is now recoverable.The cleaned and reused water is as much as 94% of the total.
文摘The key objective of intrusion detection systems(IDS)is to protect the particular host or network by investigating and predicting the network traffic as an attack or normal.These IDS uses many methods of machine learning(ML)to learn from pastexperience attack i.e.signatures based and identify the new ones.Even though these methods are effective,but they have to suffer from large computational costs due to considering all the traffic features,together.Moreover,emerging technologies like the Internet of Things(Io T),big data,etc.are getting advanced day by day;as a result,network traffics are also increasing rapidly.Therefore,the issue of computational cost needs to be addressed properly.Thus,in this research,firstly,the ML methods have been used with the feature selection technique(FST)to reduce the number of features by picking out only the important ones from NSL-KDD,CICIDS2017,and CIC-DDo S2019datasets later that helped to build IDSs with lower cost but with the higher performance which would be appropriate for vast scale network.The experimental result demonstrated that the proposed model i.e.Decision tree(DT)with Recursive feature elimination(RFE)performs better than other classifiers with RFE in terms of accuracy,specificity,precision,sensitivity,F1-score,and G-means on the investigated datasets.
文摘Currently, cybersecurity and cyber resilience are emerging and urgent issues in nextgeneration air traffic surveillance systems, which depend primarily on Automatic Dependent Surveillance-Broadcast(ADS-B) owing to its low cost and high accuracy. Unfortunately, ADS-B is prone to cyber-attacks. To verify the ADS-B positioning data of aircraft, multilateration(MLAT)techniques that use Time Differences of Arrivals(TDoAs) have been proposed. MLAT exhibits low accuracy in determining aircraft positions. Recently, a novel technique using a theoretically calculated TDoA fingerprint map has been proposed. This technique is less dependent on the geometry of sensor deployment and achieves better accuracy than MLAT. However, the accuracy of the existing technique is not sufficiently precise for determining aircraft positions and requires a long computation time. In contrast, this paper presents a reliable surveillance framework using an Actual TDoA-Based Augmentation System(ATBAS). It uses historically recorded real-data from the OpenSky network to train our TDoA fingerprint grid network. Our results show that the accuracy of the proposed ATBAS framework in determining the aircraft positions is significantly better than those of the MLAT and expected TDoA techniques by 56.93% and 48.86%, respectively. Additionally, the proposed framework reduced the computation time by 77% compared with the expected TDoA technique.
基金This study was funded by the project“Management of big data resources in the electronic socio-technological environment,the formation of electronic demography and the development of its intellectual analysis technologies”.
文摘Recently,the spread of COVID-19 virus infection and the increase of people number with chronic diseases have attracted great attention all over the world.The detection and control of such diseases based on patient demographic data are considered to be a major problem.The key issue in the solution to these problems is the development of methods and algorithms to forecast wellness and categorise patients according to their healthy and unhealthy states.In this paper,a comprehensive analysis of machine learning approaches in the field of diagnosing COVID-19 has been conducted,and for the detection of chronic diseases in patients,to identify symptoms of COVID-19 virus infection in advance,and control the situation a healthcare system has been proposed.The constructed system provides real-time monitoring of chronic diseases and COVID-19 virus infection in patients.The proposed system consists of five layers:IoT sensor layer,Data transmission layer,Fog layer,Cloud layer,the Application layer.The system architecture in the Fog layer uses machine learning and deep learning algorithms to diagnose patients'diseases,to generate and send diagnostic and emergency alerts to users.The classification module of the system's Fog layer categorises the patient's health status into healthy and unhealthy classes.In this module,to classify medical data the Decision Tree,Random Forest,SVM,Gradient Boosting,Logistic Regression algorithms are used.The COVID-19 dataset is used to test the effectiveness of the methods.The best results from the comparative analysis of the methods are obtained from the Decision Tree,Random Forest,and Gradient Boosting algorithms,which are recognised data points with high accuracy and on the accuracy metric reached 1.0,0.99,1.0 values,respectively.The classification of the other two SVM and Logistic Regression algorithms provided the worst results,and the accuracy score of both classifiers obtained a 0.89 value.
基金This work was partially supported by the Research Groups Program(Research Group Number RG-1439-033),under the Deanship of Scientific Research,King Saud University,Riyadh,Saudi Arabia.
文摘Automated grading of colon biopsy images across all magnifications is challenging because of tailored segmentation and dependent features on each magnification.This work presents a novel approach of robust magnification-independent colon cancer grading framework to distinguish colon biopsy images into four classes:normal,well,moderate,and poor.The contribution of this research is to develop a magnification invariant hybrid feature set comprising cartoon feature,Gabor wavelet,wavelet moments,HSV histogram,color auto-correlogram,color moments,and morphological features that can be used to characterize different grades.Besides,the classifier is modeled as a multiclass structure with six binary class Bayesian optimized random forest(BO-RF)classifiers.This study uses four datasets(two collected from Indian hospitals—Ishita Pathology Center(IPC)of 4X,10X,and 40X and Aster Medcity(AMC)of 10X,20X,and 40X—two benchmark datasets—gland segmentation(GlaS)of 20X and IMEDIATREAT of 10X)comprising multiple microscope magnifications.Experimental results demonstrate that the proposed method outperforms the other methods used for colon cancer grading in terms of accuracy(97.25%-IPC,94.40%-AMC,97.58%-GlaS,99.16%-Imediatreat),sensitivity(0.9725-IPC,0.9440-AMC,0.9807-GlaS,0.9923-Imediatreat),specificity(0.9908-IPC,0.9813-AMC,0.9907-GlaS,0.9971-Imediatreat)and F-score(0.9725-IPC,0.9441-AMC,0.9780-GlaS,0.9923-Imediatreat).The generalizability of the model to any magnified input image is validated by training in one dataset and testing in another dataset,highlighting strong concordance in multiclass classification and evidencing its effective use in the first level of automatic biopsy grading and second opinion.
文摘This paper presents a vision-based fingertip-writing character recognition system. The overall system is implemented through a CMOS image camera on a FPGA chip. A blue cover is mounted on the top of a finger to simplify fingertip detection and to enhance recognition accuracy. For each character stroke, 8 sample points (including start and end points) are recorded. 7 tangent angles between consecutive sampled points are also recorded as features. In addition, 3 features angles are extracted: angles of the triangle consisting of the start point, end point and average point of all (8 total) sampled points. According to these key feature angles, a simple template matching K-nearest-neighbor classifier is applied to distinguish each character stroke. Experimental result showed that the system can successfully recognize fingertip-writing character strokes of digits and small lower case letter alphabets with an accuracy of almost 100%. Overall, the proposed finger-tip-writing recognition system provides an easy-to-use and accurate visual character input method.
文摘Land cover classification of mountainous environments continues to be a challenging remote sensing problem,owing to landscape complexities exhibited by the region.This study explored a multiple classifier system(MCS)approach to the classification of mountain land cover for the Khumbu region in the Himalayas using Sentinel-2 images and a cloud-based model framework.The relationship between classification accuracy and MCS diversity was investigated,and the effects of different diversification and combination methods on MCS classification performance were comparatively assessed for this environment.We present ten MCS models that implement a homogeneous ensemble approach,using the high performing Random Forest(RF)algorithm as the selected classifier.The base classifiers of each MCS model were developed using different combinations of three diversity techniques:(1)distinct training sets,(2)Mean Decrease Accuracy feature selection,and(3)‘One-vs-All’problem reduction.The base classifier predictions of each RFMCS model were combined using:(1)majority vote,(2)weighted argmax,and(3)a meta RF classifier.All MCS models reported higher classification accuracies than the benchmark classifier(overall accuracy with 95% confidence interval:87.33%±0.97%),with the highest performing model reporting an overall accuracy(±95% confidence interval)of 90.95%±0.84%.Our key findings include:(1)MCS is effective in mountainous environments prone to noise from landscape complexities,(2)problem reduction is indicated as a stronger method over feature selection in improving the diversity of the MCS,(3)although the MCS diversity and accuracy have a positive correlation,our results suggest this is a weak relationship for mountainous classifications,and(4)the selected diversity methods improve the discriminability of MCS against vegetation and forest classes in mountainous land cover classifications and exhibit a cumulative effect on MCS diversity for this context.
文摘A new remote sensing image coding scheme based on the wavelet transform and classified vector quantization (CVQ) is proposed. The original image is first decomposed into a hierarchy of 3 layers including 10 subimages by DWT. The lowest frequency subimage is compressed by scalar quantization and ADPCM. The high frequency subimages are compressed by CVQ to utilize the similarity among different resolutions while improving the edge quality and reducing computational complexity. The experimental results show that the proposed scheme has a better performance than JPEG, and a PSNR of reconstructed image is 31～33 dB with a rate of 0.2 bpp.
