2019 IEEE Conference on Information and Communication Technology最新文献_第8页

PSNet: Parametric Sigmoid Norm Based CNN for Face Recognition 基于参数Sigmoid范数的CNN人脸识别

2019 IEEE Conference on Information and Communication Technology Pub Date : 2019-12-01 DOI: 10.1109/CICT48419.2019.9066169

Yash Srivastava, Vaishnav Murali, S. Dubey

{"title":"PSNet: Parametric Sigmoid Norm Based CNN for Face Recognition","authors":"Yash Srivastava, Vaishnav Murali, S. Dubey","doi":"10.1109/CICT48419.2019.9066169","DOIUrl":"https://doi.org/10.1109/CICT48419.2019.9066169","url":null,"abstract":"The Convolutional Neural Networks (CNN) have become very popular recently due to its outstanding performance in various computer vision applications. It is also used over widely studied face recognition problem. However, the existing layers of CNN are unable to cope with the problem of hard examples which generally produce lower class scores. Thus, the existing methods become biased towards the easy examples. In this paper, we resolve this problem by incorporating a Parametric Sigmoid Norm (PSN) layer just before the final fully-connected layer. We propose a PSNet CNN model by using the PSN layer. The PSN layer facilitates high gradient flow for harder examples as compared to easy examples. Thus, it forces the network to learn the visual features of difficult examples. We conduct the face recognition experiments to test the performance of PSN layer. The suitability of the PSN layer with different loss functions is also experimented. The widely used YouTube Faces (YTF) and Labeled Faces in the Wild (LFW) datasets are used in the experiments. The experimental results confirm the relevance of the proposed PSN laver.","PeriodicalId":234540,"journal":{"name":"2019 IEEE Conference on Information and Communication Technology","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128654066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 6

Secure and Privacy Focused Electronic Health Record Management System using Permissioned Blockchain 使用许可区块链的安全和隐私为重点的电子健康记录管理系统

2019 IEEE Conference on Information and Communication Technology Pub Date : 2019-12-01 DOI: 10.1109/CICT48419.2019.9066204

Vinay Mahore, Priyanshi Aggarwal, Nitish Andola, Raghav, S. Venkatesan

{"title":"Secure and Privacy Focused Electronic Health Record Management System using Permissioned Blockchain","authors":"Vinay Mahore, Priyanshi Aggarwal, Nitish Andola, Raghav, S. Venkatesan","doi":"10.1109/CICT48419.2019.9066204","DOIUrl":"https://doi.org/10.1109/CICT48419.2019.9066204","url":null,"abstract":"Blockchain has tremendous potential for use in various industries, including health care. It provides an immutable, shared and transparent history of all the transactions to build the applications with trust and accountability. It helps in simplifying pharmaceutical supply chains, aiding researchers in drug development by making patients' data more secure and accessible. Health care researchers today struggle with delayed communications, scattered data and delayed medical workflows caused by conflicting eHealth care systems which makes it difficult to provide personalized care. The main problem is a lack of trust between these independent eHealth care systems which makes it difficult to establish an end-to-end accessible network. Blockchain, due to its properties, can be a potential solution to provide this link. We propose a model which focuses on providing healthcare data to researchers for statistical analysis and providing privacy at the same time. The model exhibits high data security by aggregating the customized access control protocol and asymmetric cryptography. It uses proxy re-encryption technique for sensitive medical information sharing.","PeriodicalId":234540,"journal":{"name":"2019 IEEE Conference on Information and Communication Technology","volume":"30 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124384514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 18

Towards comparison of real time stream processing engines 实时流处理引擎的比较

2019 IEEE Conference on Information and Communication Technology Pub Date : 2019-12-01 DOI: 10.1109/CICT48419.2019.9066123

Devesh Kumar Lal, U. Suman

引用次数: 6

Breast Cancer Recurrence Prediction using Machine Learning 使用机器学习预测乳腺癌复发

2019 IEEE Conference on Information and Communication Technology Pub Date : 2019-12-01 DOI: 10.1109/CICT48419.2019.9066248

Kaustubh Chakradeo, Sanyog Vyawahare, Pranav Pawar

引用次数: 4

CSDG MOSFET Based Linear Voltage Regulator: A Mixed Signal Device Perspective 基于CSDG MOSFET的线性稳压器:混合信号器件的视角

2019 IEEE Conference on Information and Communication Technology Pub Date : 2019-12-01 DOI: 10.1109/CICT48419.2019.9066174

O. Oyedeji, V. Srivastava

引用次数: 0

Power Attack on VHDL Implementation of Continuously Running Block Ciphers 连续运行分组密码的VHDL实现中的功率攻击

2019 IEEE Conference on Information and Communication Technology Pub Date : 2019-12-01 DOI: 10.1109/CICT48419.2019.9066205

A. Singh, S. Mishra

{"title":"Power Attack on VHDL Implementation of Continuously Running Block Ciphers","authors":"A. Singh, S. Mishra","doi":"10.1109/CICT48419.2019.9066205","DOIUrl":"https://doi.org/10.1109/CICT48419.2019.9066205","url":null,"abstract":"The Correlation Power Analysis (CPA) is used to compromise the security of crypto systems by measuring and analyzing physical leakage that is the power consumption. Unlike classical cryptanalysis techniques, it requires very less computations to extract the secret information of the cipher systems. But, one of the main hindrances in mounting the CPA attack is the segregation of single power trace of multiple encryptions performed continuously (without pause) among number of traces corresponding to individual encryptions. To overcome this limitation, a new technique is proposed in this paper to split the power traces of AES and DES algorithms running continuously on FPGA. This (energy of samples based algorithm) finds the start of the encryption and computes number of samples in each clock of FPGA. It exploits the repetition of specific patterns available in the traces to determine the encryption length in terms of number of samples. With this information, it splits a power trace of several consecutive encryptions among number of traces corresponding to each encryption. The important thing is that it does not require information about clock frequency of FPGA board and sampling rate of the Oscilloscope. This algorithm was applied on the traces of 25, 50, 100, 125 and 250 samples per clock (when processing was performed at 1, 2 & 4 MHz clock frequencies). In all the cases, the key of AES and DES were retrieved by mounting the CPA attack on splitted traces resulting from the splitting technique.","PeriodicalId":234540,"journal":{"name":"2019 IEEE Conference on Information and Communication Technology","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114485961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Ultra Low Power High Gain High Speed OTA 超低功耗高增益高速OTA

2019 IEEE Conference on Information and Communication Technology Pub Date : 2019-12-01 DOI: 10.1109/CICT48419.2019.9066189

S. Chauhan, Akash Bahetra, Layak Singh Yadav, A. Chandan

引用次数: 0

Towards a Vehicle's behavior monitoring and Trust Computation for VANETs 基于VANETs的车辆行为监控与信任计算研究

2019 IEEE Conference on Information and Communication Technology Pub Date : 2019-12-01 DOI: 10.1109/CICT48419.2019.9066210

Anmol Tigga, P. Arun Raj Kumar

引用次数: 3

On Robust Precoder Design for Energy-Efficient MIMO-Cognitive Relay Networks 节能mimo -认知中继网络的鲁棒预编码器设计

2019 IEEE Conference on Information and Communication Technology Pub Date : 2019-12-01 DOI: 10.1109/CICT48419.2019.9066207

Shikha Maurya, M. Bansal

引用次数: 0

An Automatic Cell Nuclei Segmentation based on Deep Learning Strategies 基于深度学习策略的细胞核自动分割

2019 IEEE Conference on Information and Communication Technology Pub Date : 2019-12-01 DOI: 10.1109/CICT48419.2019.9066259

Ayush Mandloi, Ushnesha Daripa, Mukta Sharma, M. Bhattacharya

{"title":"An Automatic Cell Nuclei Segmentation based on Deep Learning Strategies","authors":"Ayush Mandloi, Ushnesha Daripa, Mukta Sharma, M. Bhattacharya","doi":"10.1109/CICT48419.2019.9066259","DOIUrl":"https://doi.org/10.1109/CICT48419.2019.9066259","url":null,"abstract":"Automatic analysis of histopathology specimens images can be utilized in early extraction and detection of diseases such brain tumor, breast malignancy, colon cancer etc. The early detection of cancer may allow patients to take proper treatment. In this paper, an automatic cell nuclei segmentation based on deep learning strategies using 2-$D$ histological images is proposed. In the proposed approach U-Net architecture is used and its hyper parameters are tuned to segment the cell nuclei. The proposed solution is built upon the highly adaptive nature of U - Net architecture. The task of nuclei segmentation in the proposed approach includes detection of nuclei in an image and extracting the foreground, while segmenting the connected foreground area into separated nuclei masks. In the experimental results the proposed approach is tested using the dataset having histopathological cell images of breast cancer. The results shows that the proposed deep learning based approach achieved the 86 % average accuracy in segmentation of cell nuclei and also outperforms the other deep learning architectures.","PeriodicalId":234540,"journal":{"name":"2019 IEEE Conference on Information and Communication Technology","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124165593","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 3