检索结果-南通市图书馆

A Large Chinese Text dataset in the Wild

在线全文

学校读者我要写书评

暂无评论

Journal of Computer Science & Technology 2019年第3期34卷 509-521页

作者： Tai-Ling Yuan Zhe Zhu Kun Xu Cheng-Jun Li Tai-Jiang Mu Shi-Min Hu Department of Computer Science and Technology Tsinghua UniversityBeijing 100084China Department of Radiology Duke UniversityNorth Carolina 27708U.S.A. Tencent Technology(Beijing)Co.Ltd. Beijing 100080China

In this paper,we introduce a very large Chinese text dataset,in the *** optical character recognition(OCR)in document images is well studied and many commercial tools are available,the detection and recognition of text in natural images is still a challenging problem,especially for some more complicated character sets such as Chinese *** of training data has always been a problem,especially for deep learning methods which require massive training *** this paper,we provide details of a newly created dataset of Chinese text with about 1 million Chinese characters from 3850 unique ones annotated by experts in over 30 000 street view *** is a challenging dataset with good diversity containing planar text,raised text,text under poor illumination,distant text,partially occluded text,*** each character,the annotation includes its underlying character,bounding box,and six *** attributes indicate the charactcr's background complexity,appearance,style,*** the dataset,we give baseline results using state-of-the-art methods for tliree tasks:character recognition(top-1 accuracy of 80.5%),character detection(AP of 70.9%),and text line detection(AED of 22.1).The dataset,source code,and trained models are publicly available.

关键词： Chinese text dataset Chinese text detection Chinese text recognition

InStereo2K: a large real dataset for stereo matching in indoor scenes

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2020年第11期63卷 136-146页

作者： Wei BAO Wei WANG Yuhua XU Yulan GUO Siyu HONG Xiaohu ZHANG School of Electrical Engineering and Automation Hefei University of Technology School of Electronics and Communication Engineering Sun Yat-sen University College of Electronic Science and Technology National University of Defense Technology School of Aeronautics and Astronautics Sun Yat-sen University

Deep neural networks have shown great success in stereo matching in recent years. On the KITTI datasets, most top performing methods are based on neural networks. However, on the Middlebury datasets,these methods usually do not perform well. The KITTI datasets are collected in outdoor scenes while the Middlebury datasets are collected in indoor scenes. It is commonly believed that the community still lacks a large labelled dataset for stereo matching in indoor scenes. In this paper, we introduce a new stereo dataset called InS tereo2K. It contains 2050 pairs of stereo images with highly accurate groundtruth disparity maps, including 2000 pairs for training and 50 pairs for test. Experimental results show that our dataset can significantly improve the performance of several latest networks(including StereoNet and PSMNet) on the Middlebury 2014 dataset. The large scale, high accuracy and rich diversity of the proposed InS tereo2K dataset provide new opportunities to researchers in the area of stereo matching and beyond. It also takes end-to-end stereo matching methods a step towards practical applications.

关键词： stereo matching depth estimation convolutional neural network dataset

A new dataset of dog breed images and a benchmark for fine-grained classification

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Computational Visual Media 2020年第4期6卷 477-487页

作者： Ding-Nan Zou Song-Hai Zhang Tai-Jiang Mu Min Zhang Department of Computer Science and Technology BNRistTsinghua UniversityBeijing 100084China NaJiu Company Hunan 410022China Harvard Medical School Brigham and Women’s HospitalBostonMA 02115USA

In this paper, we introduce an image dataset for fine-grained classification of dog breeds: the Tsinghua Dogs dataset. It is currently the largest dataset for fine-grained classification of dogs, including 130 dog breeds and 70,428 real-world images. It has only one dog in each image and provides annotated bounding boxes for the whole body and head. In comparison to previous similar datasets, it contains more breeds and more carefully chosen images for each breed. The diversity within each breed is greater,with between 200 and 7000+ images for each *** of the whole body and head makes the dataset not only suitable for the improvement of finegrained image classification models based on overall features, but also for those locating local informative parts. We show that dataset provides a tough challenge by benchmarking several state-of-the-art deep neural models. The dataset is available for academic purposes at https://***/ThuDogs/.

关键词： fine-grained classification dog dataset benchmark

Distribution Network Planning Enhancement via Network Reconfiguration and DG Integration Using dataset Approach and Water Cycle Algorithm

在线全文

学校读者我要写书评

暂无评论

Journal of Modern Power Systems and Clean Energy 2020年第1期8卷 86-93页

作者： Munir Azam Muhammad Hazlie Mokhlis Kanendra Naidu Adil Amin John Fredy Franco Mohamadariff Othman Department of Electrical Engineering Faculty of EngineeringUniversity of Malaya50603 Kuala LumpurMalaysia Electrical Technology Section University Kuala LumpurBritish Malaysian InstituteBt.8Jalan Sungai Pusu53100 GombakSelangor Darul EhsanMalaysia Sao Paulo State University Campus of Rosana19274-000RosanaBrazil

The integration of network reconfiguration and distributed generation(DG)can enhance the performances of overall ***,proper sizing and siting of DG need to be determined,otherwise it will cause degradation in system ***,determining proper sizing and siting of DG together with network reconfiguration is a complex problem due to huge solution search *** search space mostly contains non-radial network *** these non-radial combinations during optimization process increases computational overhead and may end up at local optimal *** reduce the searching complexity,this paper considers the discretized network reconfiguration via dataset *** cycle algorithm(WCA)is used to obtain the near optimal solution of network reconfiguration,and sizing and sitting of *** addition,the power factor of DG is also optimized to reduce the power *** proposed method is tested on an IEEE 33-bus network and an IEEE 69-bus network considering different scenarios to show the effectiveness of simultaneous approach considering variable power *** results show that the discretization of reconfiguration search space avoids that WCA to get trapped in local *** proposed method outperforms other technique such as harmony search algorithm(HSA),fireworks algorithm(FWA),Cuckoo search algorithm(CSA)and uniform voltage distribution based constructive algorithm(UVDA)and improves the solution quality of IEEE 33-bus network and 69-bus network by 29.20%and 27.88%,respectively.

关键词： Reactive power injection sizing and siting of distributed generation(DG) dataset approach active distribution network power system

Research on Enhanced Contraband dataset ACXray Based on ETL

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第6期79卷 4551-4572页

作者： Xueping Song Jianming Yang Shuyu Zhang Jicun Zhang School of Mechanical Engineering Dalian Jiaotong UniversityDalian116028China Neusoft Reach Automotive Technology(Dalian)Co. Ltd.Dalian116085China

To address the shortage of public datasets for customs X-ray images of contraband and the difficulties in deploying trained models in engineering applications,a method has been proposed that employs the Extract-Transform-Load(ETL)approach to create an X-ray dataset of contraband ***,X-ray scatter image data is collected and *** Kafka message queues and the Elasticsearch(ES)distributed search engine,the data is transmitted in real-time to cloud ***,contraband data is annotated using a combination of neural networks and manual methods to improve annotation efficiency and implemented mean hash algorithm for quick image *** method of integrating targets with backgrounds has enhanced the X-ray contraband image data,increasing the number of positive ***,an Airport Customs X-ray dataset(ACXray)compatible with customs business scenarios has been constructed,featuring an increased number of positive contraband *** tests using three datasets to train the Mask Region-based Convolutional Neural Network(Mask R-CNN)algorithm and tested on 400 real customs images revealed that the recognition accuracy of algorithms trained with Security Inspection X-ray(SIXray)and Occluded Prohibited Items X-ray(OPIXray)decreased by 16.3%and 15.1%,respectively,while the ACXray dataset trained algorithm’s accuracy was almost *** indicates that the ACXray dataset-trained algorithm possesses strong generalization capabilities and is more suitable for customs detection scenarios.

关键词： X-ray contraband ETL data enhancement dataset

A Potential Risk Index dataset for Landfalling Tropical Cyclones over the Chinese Mainland(PRITC dataset V1.0)

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Advances in Atmospheric Sciences 2021年第10期38卷 1791-1802页

作者： Peiyan CHEN Hui YU Kevin K.W.CHEUNG Jiajie XIN Yi LU Shanghai Typhoon Institute China Meteorological AdministrationShanghai 200030China Key Laboratory of Numerical Modeling for Tropical Cyclone of China Meteorological Administration Shanghai 20030China The Joint Laboratory for Typhoon Forecasting Technique Applications between Shanghai Typhoon Institute and Wenzhou Meteorological Bureau Wenzhou 325027China Climate and Atmospheric Science NSW Department of Planning Industry and EnvironmentSydney 2000Australia

A dataset entitled“A potential risk index dataset for landfalling tropical cyclones over the Chinese mainland”(PRITC dataset V1.0)is described in this paper,as are some basic statistical *** the severity of the impacts of tropical cyclones(TCs)that make landfall on the Chinese mainland based on observations from 1401 meteorological stations was proposed in a previous study,including an index combining TC-induced precipitation and wind(IPWT)and further information,such as the corresponding category level(CAT_IPWT),an index of TC-induced wind(IWT),and an index of TC-induced precipitation(IPT).The current version of the dataset includes TCs that made landfall from 1949-2018;the dataset will be extended each ***-term trend analyses demonstrate that the severity of the TC impacts on the Chinese mainland have increased,as embodied by the annual mean IPWT values,and increases in TC-induced precipitation are the main contributor to this *** Winnie(1997)and TC Bilis(2006)were the two TCs with the highest IPWT and IPT values,*** PRITC V1.0 dataset was developed based on the China Meteorological Administration’s tropical cyclone database and can serve as a bridge between TC hazards and their social and economic impacts.

关键词： tropical cyclone risk dataset China

A 10-Yr Global Land Surface Reanalysis Interim dataset(CRA-Interim/Land):Implementation and Preliminary Evaluation

在线全文

学校读者我要写书评

暂无评论

Journal of Meteorological Research 2020年第1期34卷 101-116页

作者： Xiao LIANG Lipeng JIANG Yang PAN Chunxiang SHI Zhiquan LIU Zijiang ZHOU National Meteorological Information Center China Meteorological AdministrationBeijing 100081China National Center for Atmospheric Research BoulderCO 80301USA

A land surface reanalysis dataset covering the most recent decades is able to provide temporally consistent initial conditions for weather and climate models,and thus is crucial to verifying/improving numerical weather/climate forecasts/*** this paper,we report the development of a 10-yr China Meteorological Administration(CMA)global Land surface ReAnalysis Interim dataset(CRA-Interim/Land;2007–2016,6-h intervals,approximately 34-km horizontal resolution).The dataset was produced and evaluated by using the Global Land Data Assimilation System(GLDAS)and NCEP Climate Forecast System Reanalysis(CFSR)global land surface reanalysis datasets,as well as in situ observations in *** results show that the global spatial patterns and monthly variations of the CRA-Interim/Land,GLDAS,and CFSR climatology are highly consistent,while the soil moisture and temperature values of the CRA-Interim/Land dataset are in between those of the GLDAS and CFSR *** with ground observations in China,CRA-Interim/Land soil moisture is comparable to or better than that of GLDAS and CFSR datasets for the 0–10-cm soil layer and has higher correlations and slightly lower root mean square errors(RMSE)for the 10–40-cm soil ***,CRA-Interim/Land shows negative biases in 10–40-cm soil moisture in Northeast China and north of central *** ground temperature and the soil temperature in different layers,CRA-Interim/Land behaves better than the CFSR,especially in East and central ***-Interim/Land has added value over the land components of CRA-Interim due to the introduction of global precipitation observations and improved soil/vegetation ***,this dataset is potentially a critical supplement to the *** evaluation of the CRA-Interim/Land,assimilation of near-surface atmospheric forcing variables,and extension of the current dataset to 40 yr(1979–2018)are in progress.

关键词： global land surface reanalysis dataset CRA-Interim/Land ground temperature soil temperature soil moisture Global Land Data Assimilation System(GLDAS)

MBB-IoT:Construction and Evaluation of IoT DDoS Traffic dataset from a New Perspective

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2023年第8期76卷 2095-2119页

作者： Yi Qing Xiangyu Liu Yanhui Du Department of Artificial Intelligence People’s Public Security University of ChinaBeijing100038China Department of Cybersecurity People’s Public Security University of ChinaBeijing100038China

Distributed Denial of Service(DDoS)attacks have always been a major concern in the security *** the release of malware source codes such as BASHLITE and Mirai,Internet of Things(IoT)devices have become the new source of DDoS attacks against many Internet *** there are many datasets in the field of IoT intrusion detection,such as Bot-IoT,ConstrainedApplication Protocol–Denial of Service(CoAPDoS),and LATAM-DDoS-IoT(some of the names of DDoS datasets),which mainly focus on DDoS attacks,the datasets describing new IoT DDoS attack scenarios are extremely rare,and only N-BaIoT and IoT-23 datasets used IoT devices as DDoS attackers in the construction process,while they did not use Internet applications as victims *** supplement the description of the new trend of DDoS attacks in the dataset,we built an IoT environment with mainstream DDoS attack tools such as Mirai and BASHLITE being used to infect IoT devices and implement DDoS attacks against WEB ***,data aggregated into a dataset namedMBB-IoTwere captured atWEBservers and IoT *** the MBB-IoT dataset was split into a training set and a test set,it was applied to the training and testing of the Random Forests classification *** multi-class classification metrics were good and all above 90%.Secondly,in a cross-evaluation experiment based on Support Vector Machine(SVM),Light Gradient Boosting Machine(LightGBM),and Long Short Term Memory networks(LSTM)classification algorithms,the training set and test set were derived from different datasets(MBB-IoT or IoT-23),and the test performance is better when MBB-IoT is used as the training set.

关键词： Intrusion detection IoT malware botnet DDoS dataset

SciCN:A Scientific dataset for Chinese Named Entity Recognition

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第3期78卷 4303-4315页

作者： Jing Yang Bin Ji Shasha Li Jun Ma Jie Yu College of Computer National University of Defense TechnologyChangsha410073China

Named entity recognition(NER)is a fundamental task of information extraction(IE),and it has attracted considerable research attention in recent *** abundant annotated English NER datasets have significantly promoted the NER research in the English *** contrast,much fewer efforts are made to the Chinese NER research,especially in the scientific domain,due to the scarcity of Chinese NER *** alleviate this problem,we present aChinese scientificNER dataset–SciCN,which contains entity annotations of titles and abstracts derived from 3,500 scientific *** manually annotate a total of 62,059 entities,and these entities are classified into six *** to English scientific NER datasets,SciCN has a larger scale and is more diverse,for it not only contains more paper abstracts but these abstracts are derived from more research *** investigate the properties of SciCN and provide baselines for future research,we adapt a number of previous state-of-theart Chinese NER models to evaluate *** results show that SciCN is more challenging than other Chinese NER *** addition,previous studies have proven the effectiveness of using lexicons to enhance Chinese NER *** by this fact,we provide a scientific domain-specific *** results demonstrate that our lexicon delivers better performance gains than lexicons of other *** hope that the SciCN dataset and the lexicon will enable us to benchmark the NER task regarding the Chinese scientific domain and make progress for future *** dataset and lexicon are available at:https://***/yangjingla/***.

关键词： Named entity recognition dataset scientific information extraction lexicon