2019 2nd International Conference on Advancements in Computational Sciences (ICACS)最新文献

{"title":"Recognizing Actions of Distracted Drivers using Inception v3 and Xception Convolutional Neural Networks","authors":"Zaeem Ahmad Varaich, S. Khalid","doi":"10.23919/ICACS.2019.8689131","DOIUrl":"https://doi.org/10.23919/ICACS.2019.8689131","url":null,"abstract":"In recent years, Deep Convolutional Neural Networks have shown remarkable success in image classification tasks. In our research, we compare the performance of two competing DCNN architectures, viz. Inception v3 and Xception, and use them to recognize ten unique actions of the drivers in the Kaggle’s State Farm Distracted Driver Detection challenge. We discuss the performance of both architectures in detail (in terms of loss, accuracy and Kaggle test set scores) under two different weight initialization schemes, i.e. random initialization and transfer learning using ImageNet weights, with the training set split on drivers. Additionally, for comparison, we also used a randomly split training set and trained the models using ImageNet weights. By splitting training dataset on drivers, we find that high top-1 validation accuracy of 85.4% is achieved by training the Xception architecture using transfer learning with ImageNet initialized weights. This accuracy is further increased to 99.3% for the same Xception architecture scheme, when we split training data randomly instead of splitting it on subjects. Our best trained model utilizing the Xception architecture with ImageNet initialized weights ranks at 325th position (out of 1440 entries) on Kaggle’s Private Leaderboard with a remarkable test loss of 0.51285.","PeriodicalId":290819,"journal":{"name":"2019 2nd International Conference on Advancements in Computational Sciences (ICACS)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130641119","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}

{"title":"Modeling Task Capability in Full Velocity Differential Model","authors":"Wajiha Batool, Mian Muhammad Mubasher, Syed Waqar ul Qounian","doi":"10.23919/ICACS.2019.8689141","DOIUrl":"https://doi.org/10.23919/ICACS.2019.8689141","url":null,"abstract":"Car following (CF) models formally explain acceleration behavior of drivers. Historically, human factors are not considered in CF models. Attention is a very critical human factor. Drug use, panic, fear, or anger may negatively affect attention and consequently driving behavior. In the recent years, researchers have focused on modeling of CF behavior considering human factors as an outcome of research by traffic psychologists and engineers. These observations make clear that integration of human factors into car following models is necessary to develop a more realistic depiction of CF maneuvers under intricate driving situations. In complex driving situations, it is important to measure the dynamic interaction of driving task demand and ability of driver to handle the task at hand. The basic idea of Task Capability Interface (TCI) model is to incorporate task difficulty and task demand within a framework which gives the detailed account of their influence on one another. Task demand and capability plays a key role in decision making. TCI model has earlier been used to improve two traditional CF models namely Gipps’ model and Intelligent Driver Model (IDM). The enhanced models are referred as TD-Gipps model and TD-IDM. There is another model namely Full Velocity Differential Model (FVDM). Unlike its predecessors, FVDM doesn’t suffer from unrealistic acceleration and deacceleration. But FVDM has not been enhanced using TCI model. In this work, FVDM has been enhanced to incorporate TCI model. The enhanced model namely TD-FVDM has been verified by comparing it with TD-Gipps using simulation-based experiments. The enhanced proposed model reproduces acceleration behavior as intended.","PeriodicalId":290819,"journal":{"name":"2019 2nd International Conference on Advancements in Computational Sciences (ICACS)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128083385","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}

{"title":"Categorization of Crowd Varieties using Deep Concurrent Convolution Neural Network","authors":"Gulraiz Khan, Muhammad Ali Farooq, Junaid Hussain, Zeeshan Tariq, Muhammad Usman Ghani Khan","doi":"10.23919/ICACS.2019.8689129","DOIUrl":"https://doi.org/10.23919/ICACS.2019.8689129","url":null,"abstract":"Visual understanding of crowd scenes is a challenging and important issue in computer vision domain. Identification of crowd type is a basic requirement for analyzing crowd scenarios. With the advancement of deep convolution neural networks image recognition problems have become easy. In this paper, we propose a novel architecture (DeepCrowd) inspired by Resnet to incorporate spatial features comprehensively. To train and evaluate proposed system, a robust and unique dataset of nearly six thousand images is generated. Evaluating the system extensively highlighted accuracy of 83.11% that is comparable with others state-of-the-art methods.","PeriodicalId":290819,"journal":{"name":"2019 2nd International Conference on Advancements in Computational Sciences (ICACS)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133926025","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}

{"title":"Early Student Grade Prediction: An Empirical Study","authors":"Z. Iqbal, A. Qayyum, S. Latif, Junaid Qadir","doi":"10.23919/ICACS.2019.8689136","DOIUrl":"https://doi.org/10.23919/ICACS.2019.8689136","url":null,"abstract":"In higher educational institutes, early grade prediction is an important area of interest as it allows instructors to improve students’ performance in their courses by providing special attention at the early stages. Machine learning techniques can be utilized for students’ grades prediction in different courses. However, the performance of these techniques is highly dependent on the quality of data that made the selection of model a challenging task. Therefore, in this paper, we evaluate different state-of-the-art machine learning techniques for university students grade prediction. Ultimately we find that Restricted Boltzmann Machines (RBM) can more accurately predict students’ grades. The predicted grades by these techniques visualize uncertainty on student learning and can be used for confidence gains, student degree planning, personalized advising, and to enable instructors to identify potential students who might need assistance in relevant courses.","PeriodicalId":290819,"journal":{"name":"2019 2nd International Conference on Advancements in Computational Sciences (ICACS)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124392086","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}

{"title":"ICACS 2019 Program Committee","authors":"Mubasher Mian, Imran Farid Khan","doi":"10.23919/icacs.2019.8689142","DOIUrl":"https://doi.org/10.23919/icacs.2019.8689142","url":null,"abstract":"Mubasher Mian Punjab University College of Information Technology, University of the Punjab, Lahore, Pakistan Zara Nasar Punjab University College of Information Technology, University of the Punjab, Lahore, Pakistan Toqir Rana The University of Lahore, Lahore, Pakistan Awais Athar European Bioinformatics Institute, UK Zeeshan Rana FAST-National University of Computer and Emerging Sciences, Pakistan Zulfiqar Ali The University Of Lahore, Lahore, Pakistan Tayyab Naseer University of Freiburg, Germany Tayyaba Anees University of Management and Technology, Lahore, Pakistan Khurram Shahzad Punjab University College of Information Technology, University of the Punjab, Lahore, Pakistan Awais Mahmood King Saud University, Riyadh, Saudi Arabia Ahmad Salman Khan The University of Lahore, Lahore, Pakistan Muhammad Shahzad Cheema Bonn-Aachen International Center for Information Technology, University of Bonn, Germany Nadeem Ahmad The Superior College (University Campus), Lahore, Pakistan Abbas Khalid The University of Lahore, Lahore, Pakistan Mohammad Ahmed Shah Istanbul Arel University, Turkey Muhammad Asif National Textile University, Faisalabad, Pakistan Eraj Khan The University of Lahore, Lahore, Pakistan Shahzad Sarwar Punjab University College of Information Technology, University of the Punjab, Lahore, Pakistan Ahmed Bendahmane Computer Science and Systems Engineering Laboratory, Abdelmalek Essaadi University, Tetuan, Morocco Alexander Köpper Robotics Research Lab, University of Kaiserslautern, Germany Xiaojiang Peng Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, China Hafiz Mahfooz Ul Haque The University of Lahore, Lahore, Pakistan Ghulam Rasool COMSATS University, Islamabad, Pakistan Haroon Mahmood Politecnico di Torino, Italy Syed Waqar Ul Qounain Jaffry Punjab University College of Information Technology, University of the Punjab, Lahore, Pakistan Sheheryar Malik Superior University, Lahore, Pakistan Muhammad Humayoun University of Central Punjab, Lahore, Pakistan Khawaja Muhammad Umar Suleman FAST-National University of Computer and Emerging Sciences, Pakistan Imran Ali National Engineering and Scientific Commission (NESCOM), Pakistan Tufail Muhammad The University of Lahore, Lahore, Pakistan Mehtab Afzal The University of Lahore, Lahore, Pakistan Arshad Ali The University of Lahore, Lahore, Pakistan Abdur Rakib The University of the West of England, Bristol, UK","PeriodicalId":290819,"journal":{"name":"2019 2nd International Conference on Advancements in Computational Sciences (ICACS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130180835","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}

{"title":"Scene Recognition of Surveillance Data using Deep Features and Supervised Classifiers","authors":"Saira Jabeen, Summra Saleem, Abdulrehman Azam, Muhammad Usman Ghani Khan","doi":"10.23919/ICACS.2019.8689001","DOIUrl":"https://doi.org/10.23919/ICACS.2019.8689001","url":null,"abstract":"Precise labeling of an image based on its semantic description is quite challenging task and has its significant applications in surveillance area. Majority of scene classification techniques during past few decades have targeted low level feature by handcraft engineering or unsupervised feature extraction techniques. In this paper, we aim to categorize scene classes for surveillance systems by exploiting deep convolutional features to manifold projection along with supervised classification algorithms. A topology is constructed to depict high dimensions of convolution heat-maps to 128D salient features. Parameters of pre-trained network are tuned to precisely fit with the output of our problem. Experimental results depict that our methodology is more robust and competitive as compared to state of the art methods.","PeriodicalId":290819,"journal":{"name":"2019 2nd International Conference on Advancements in Computational Sciences (ICACS)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127911420","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}

{"title":"[Title page]","authors":"","doi":"10.23919/icacs.2019.8689126","DOIUrl":"https://doi.org/10.23919/icacs.2019.8689126","url":null,"abstract":"","PeriodicalId":290819,"journal":{"name":"2019 2nd International Conference on Advancements in Computational Sciences (ICACS)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130187494","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}

{"title":"Smart Contracts Integration between Blockchain and Internet of Things: Opportunities and Challenges","authors":"A. Rashid, M. Siddique","doi":"10.23919/ICACS.2019.8689132","DOIUrl":"https://doi.org/10.23919/ICACS.2019.8689132","url":null,"abstract":"A Smart Contract is self-executable and self-enforceable program code that runs on the top of blockchain to manage complex business logic. It eliminates the need of extrinsic enforcement of legal agreements. Furthermore, it enforces the terms and conditions of an agreement that lies between untrustworthy parties in which the trusted third parties cannot interfere. The cryptography logic used in smart contract enables the blockchain network to provide trust and authority to all parties in transaction. Network decentralization, data immutability and transparency, resiliency and security make blockchain technology more versatile. Recently, it has become a potential quality and capability of IoT to connect uncountable electronic objects or devices at the same time. The most prominent feature of blockchain-based IoT applications is the integration of smart contracts between blockchain and IoT.A brief comparison has been given in the paper that how the smart contracts react on multiple blockchain platforms with respect to scalability, system complexity and consensus protocol factors. Furthermore, the context of Smart contract integration between blockchain and IoT with highlighting the integration opportunities and challenges along with future research directions. Therefore, we have concluded in the current paper that amalgamation of Blockchain with IoT through Smart Contract can provide a strong framework for distributed application and the newly introduced business communities.","PeriodicalId":290819,"journal":{"name":"2019 2nd International Conference on Advancements in Computational Sciences (ICACS)","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115551504","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}

{"title":"Use of Directional UWB Antenna for Lung Tumour Detection","authors":"Wajeeha Ameer, D. Awan, S. Bashir, A. Waheed","doi":"10.23919/ICACS.2019.8689137","DOIUrl":"https://doi.org/10.23919/ICACS.2019.8689137","url":null,"abstract":"In this research, the use of directional UWB antennas for the purpose of lung tumour detection is investigated. A comparison of the performance of directional and non-directional UWB antennas in the detection of lung tumour is presented through simulations and it is shown that using a directional UWB antenna results in clear detection of tumor and also allows to increase the distance between the patient and the antenna to perform the test. The difference in the plots of S parameters of antenna for normal lung and cancerous lung provides the basis for the detection of tumour. These findings can further pave the way for manufacturing patient friendly microwave medical imaging equipment in future.","PeriodicalId":290819,"journal":{"name":"2019 2nd International Conference on Advancements in Computational Sciences (ICACS)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128819843","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}

{"title":"ICACS 2019 Author Index","authors":"","doi":"10.23919/icacs.2019.8688998","DOIUrl":"https://doi.org/10.23919/icacs.2019.8688998","url":null,"abstract":"","PeriodicalId":290819,"journal":{"name":"2019 2nd International Conference on Advancements in Computational Sciences (ICACS)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130388114","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}