Pub Date : 2018-08-01DOI: 10.1109/ICCUBEA.2018.8697792
Supriya Kute, S. Javheri
Cloud computing allows us to create, configure, and personalize the operations online. In the cloud computing environment, the owner could use encryption with attributes to encrypt the uploaded data which accomplishing access control and data security. Similarly data can be decrypted if authorized person want to access it. In the previous work, user can encrypt and decrypt the file according to the set of attributes. But there are some problems and questions related to that work. For example, during the delegation, the cloud servers could replace data or represent wrong data and respond a fake result with malicious intent. In addition, the cloud server may cheat the authorized user by saying that they are not authorized one for accessing data. Also, the access structure is not flexible during cryptography, Since access structure is applicable to the circuits for greater data security. System develop an attribute-based design with a time specified attribute scheme encryption. In this system, whenever the owner uploads a file, it is labelled with a set of attributes that includes: department, work profile, branch, experience which is called as access structure. After this time interval, date and location also added. The user can decrypt and download the file if the time interval, date location and attributes matches with the owner set attributes. Before this, authority will check whether user is authorized to access any of the file. To achieve more security file is split into multiple fragments according to file size and stored on multiple nodes instead of being stored on a single node. The system has created a confirmable calculation, an authorized user access and detailed approach. It also gives us the guarantee of the correctness of the delegated computer results.
{"title":"Implementation of Secure File Storage on Cloud with Owner-Defined Attributes for Encryption","authors":"Supriya Kute, S. Javheri","doi":"10.1109/ICCUBEA.2018.8697792","DOIUrl":"https://doi.org/10.1109/ICCUBEA.2018.8697792","url":null,"abstract":"Cloud computing allows us to create, configure, and personalize the operations online. In the cloud computing environment, the owner could use encryption with attributes to encrypt the uploaded data which accomplishing access control and data security. Similarly data can be decrypted if authorized person want to access it. In the previous work, user can encrypt and decrypt the file according to the set of attributes. But there are some problems and questions related to that work. For example, during the delegation, the cloud servers could replace data or represent wrong data and respond a fake result with malicious intent. In addition, the cloud server may cheat the authorized user by saying that they are not authorized one for accessing data. Also, the access structure is not flexible during cryptography, Since access structure is applicable to the circuits for greater data security. System develop an attribute-based design with a time specified attribute scheme encryption. In this system, whenever the owner uploads a file, it is labelled with a set of attributes that includes: department, work profile, branch, experience which is called as access structure. After this time interval, date and location also added. The user can decrypt and download the file if the time interval, date location and attributes matches with the owner set attributes. Before this, authority will check whether user is authorized to access any of the file. To achieve more security file is split into multiple fragments according to file size and stored on multiple nodes instead of being stored on a single node. The system has created a confirmable calculation, an authorized user access and detailed approach. It also gives us the guarantee of the correctness of the delegated computer results.","PeriodicalId":422920,"journal":{"name":"2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131095883","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}
Pub Date : 2018-08-01DOI: 10.1109/ICCUBEA.2018.8697672
Mithra Venkatesan, A. Kulkarni, Radhika Menon
Cognitive Radios are smart radio which can reconfigure and adapt according to the requirement of end user. Cognitive Engine greatly contributes to the intelligent radio. Learning is an essential phase in the Cognitive engine, enabling forecasting of different functional factors. This paper proposes stochastic time series based learning outline which can be used for Cognitive Radio towards forecast of key parameters of throughput and data rates. The learning outline is capable to prediction up to 99% with minimum Root Mean Square Error. These learning schemes can be valuable inputs for Dynamic Spectrum Allocation. Subsequently, these outlines will form part of Cognitive Engine and can be utilized to perform allocation of spectrum resulting in a futuristic wise radio
{"title":"Stochastic Time Series Learning Scheme for Throughput Prediction in Cognitive Radio System","authors":"Mithra Venkatesan, A. Kulkarni, Radhika Menon","doi":"10.1109/ICCUBEA.2018.8697672","DOIUrl":"https://doi.org/10.1109/ICCUBEA.2018.8697672","url":null,"abstract":"Cognitive Radios are smart radio which can reconfigure and adapt according to the requirement of end user. Cognitive Engine greatly contributes to the intelligent radio. Learning is an essential phase in the Cognitive engine, enabling forecasting of different functional factors. This paper proposes stochastic time series based learning outline which can be used for Cognitive Radio towards forecast of key parameters of throughput and data rates. The learning outline is capable to prediction up to 99% with minimum Root Mean Square Error. These learning schemes can be valuable inputs for Dynamic Spectrum Allocation. Subsequently, these outlines will form part of Cognitive Engine and can be utilized to perform allocation of spectrum resulting in a futuristic wise radio","PeriodicalId":422920,"journal":{"name":"2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128847125","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}
Pub Date : 2018-08-01DOI: 10.1109/ICCUBEA.2018.8697819
Ruturaj Kulkarni, Shruti Dhavalikar, S. Bangar
Self-driving cars has the potential to revolutionize urban mobility by providing sustainable, safe, convenient and congestion free transportability. This vehicle autonomy as an application of AI has several challenges like infallibly recognizing traffic lights, signs, unclear lane markings, pedestrians, etc. These problems can be overcome by using the technological development in the fields of Deep Learning, Computer Vision due to availability of Graphical Processing Units (GPU) and cloud platform. In this paper, we propose a deep neural network based model for reliable detection and recognition of traffic lights using transfer learning. The method incorporates use of faster region based convolutional network (R-CNN) Inception V2 model in TensorFlow for transfer learning. The model was trained on dataset containing different images of traffic signals in accordance with Indian Traffic Signals which are distinguished in five types of classes. The model accomplishes its objective by detecting the traffic light with its correct class type.
{"title":"Traffic Light Detection and Recognition for Self Driving Cars Using Deep Learning","authors":"Ruturaj Kulkarni, Shruti Dhavalikar, S. Bangar","doi":"10.1109/ICCUBEA.2018.8697819","DOIUrl":"https://doi.org/10.1109/ICCUBEA.2018.8697819","url":null,"abstract":"Self-driving cars has the potential to revolutionize urban mobility by providing sustainable, safe, convenient and congestion free transportability. This vehicle autonomy as an application of AI has several challenges like infallibly recognizing traffic lights, signs, unclear lane markings, pedestrians, etc. These problems can be overcome by using the technological development in the fields of Deep Learning, Computer Vision due to availability of Graphical Processing Units (GPU) and cloud platform. In this paper, we propose a deep neural network based model for reliable detection and recognition of traffic lights using transfer learning. The method incorporates use of faster region based convolutional network (R-CNN) Inception V2 model in TensorFlow for transfer learning. The model was trained on dataset containing different images of traffic signals in accordance with Indian Traffic Signals which are distinguished in five types of classes. The model accomplishes its objective by detecting the traffic light with its correct class type.","PeriodicalId":422920,"journal":{"name":"2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126741531","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}
Pub Date : 2018-08-01DOI: 10.1109/ICCUBEA.2018.8697562
M. Kokare, S. V. Tade
In this paper, first time ever artificial colony bee search is applied for thermal unit commitment. The main aim of doing thermal unit commitment scheduling is to optimize running while ensuring security and reliability. In recent era, lot of attention is given to the environmental conditions. All developed and few developing countries are worrying about emission of carbon dioxide from thermal generating stations. In first section of this paper unit commitment and Artificial Bee Colony (ABC) is discussed. In next section methodology for adopting ABC and general constraints of unit commitment are discussed in great detail. In last section application of ABC for unit commitment is implementation of proposed algorithm is shown and obtained results are presented. It is observed that ABC produces good results for unit commitment problem and easy to implement.
{"title":"Application of Artificial Bee Colony Method for Unit Commitment","authors":"M. Kokare, S. V. Tade","doi":"10.1109/ICCUBEA.2018.8697562","DOIUrl":"https://doi.org/10.1109/ICCUBEA.2018.8697562","url":null,"abstract":"In this paper, first time ever artificial colony bee search is applied for thermal unit commitment. The main aim of doing thermal unit commitment scheduling is to optimize running while ensuring security and reliability. In recent era, lot of attention is given to the environmental conditions. All developed and few developing countries are worrying about emission of carbon dioxide from thermal generating stations. In first section of this paper unit commitment and Artificial Bee Colony (ABC) is discussed. In next section methodology for adopting ABC and general constraints of unit commitment are discussed in great detail. In last section application of ABC for unit commitment is implementation of proposed algorithm is shown and obtained results are presented. It is observed that ABC produces good results for unit commitment problem and easy to implement.","PeriodicalId":422920,"journal":{"name":"2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA)","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114417873","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}
Pub Date : 2018-08-01DOI: 10.1109/ICCUBEA.2018.8697370
M. Prasad, Rizwan Japanwala, Harshil Vora, Lakshmi Kurup
The aim of the project is to create an AI based application that can analyze emotions in frames from a live video feed of a person learning and determine the relative understanding level of the person during and after the learning process. This is achieved by using a Convolutional Neural Network trained by the FER-2013 Dataset “Emotions In The Wild” for emotion recognition and an algorithm to determine relative level of understanding using that information. Output will finally show the relative variation in the person's level of understanding over time. The patterns in the variation can then be easily interpreted to determine whether the person is understanding and at what point of time during the learning process did it drop or increase. A web-based tool for analysis of understanding level of the person is used at the end of session.
{"title":"Human Understanding Analyzer","authors":"M. Prasad, Rizwan Japanwala, Harshil Vora, Lakshmi Kurup","doi":"10.1109/ICCUBEA.2018.8697370","DOIUrl":"https://doi.org/10.1109/ICCUBEA.2018.8697370","url":null,"abstract":"The aim of the project is to create an AI based application that can analyze emotions in frames from a live video feed of a person learning and determine the relative understanding level of the person during and after the learning process. This is achieved by using a Convolutional Neural Network trained by the FER-2013 Dataset “Emotions In The Wild” for emotion recognition and an algorithm to determine relative level of understanding using that information. Output will finally show the relative variation in the person's level of understanding over time. The patterns in the variation can then be easily interpreted to determine whether the person is understanding and at what point of time during the learning process did it drop or increase. A web-based tool for analysis of understanding level of the person is used at the end of session.","PeriodicalId":422920,"journal":{"name":"2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA)","volume":"52 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113987889","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}
Pub Date : 2018-08-01DOI: 10.1109/ICCUBEA.2018.8697871
Rani G. Utekar, J. Umale
Automated healthcare system is the need and future of healthcare in India. The challenges in implementation of healthcare systems in developing country like India are technology, infrastructure, trained doctors and connectivity among all stakeholders. Due the rapid growth in population providing the healthcare services is becoming difficult day by day specially in rural areas. The remotely located patients are the patients away from doctor but needs his constant monitoring and support. Such as patients in ICU, at home or may be at distant places. The problems also lies in updating doctors of the monitoring parameters and the history of patients time to time. This paper presents the implementation of automated IoT based helthcare system for remotely located patients which helps doctors and guide them accordingly. The system provides alerts by the means of E-mails in case of abnormal conditions observed in the monitoring parameters of the patient. It also takes care of supporting the decision making of severity of health conditions. An example of heart patient monitoring is taken for demonstration of the implemented system. The implemented system is successful to provide an interface among doctors, the nurses in hospitals and the relatives of the patient.
{"title":"Automated IoT Based Healthcare System for Monitoring of Remotely Located Patients","authors":"Rani G. Utekar, J. Umale","doi":"10.1109/ICCUBEA.2018.8697871","DOIUrl":"https://doi.org/10.1109/ICCUBEA.2018.8697871","url":null,"abstract":"Automated healthcare system is the need and future of healthcare in India. The challenges in implementation of healthcare systems in developing country like India are technology, infrastructure, trained doctors and connectivity among all stakeholders. Due the rapid growth in population providing the healthcare services is becoming difficult day by day specially in rural areas. The remotely located patients are the patients away from doctor but needs his constant monitoring and support. Such as patients in ICU, at home or may be at distant places. The problems also lies in updating doctors of the monitoring parameters and the history of patients time to time. This paper presents the implementation of automated IoT based helthcare system for remotely located patients which helps doctors and guide them accordingly. The system provides alerts by the means of E-mails in case of abnormal conditions observed in the monitoring parameters of the patient. It also takes care of supporting the decision making of severity of health conditions. An example of heart patient monitoring is taken for demonstration of the implemented system. The implemented system is successful to provide an interface among doctors, the nurses in hospitals and the relatives of the patient.","PeriodicalId":422920,"journal":{"name":"2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122546083","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}
Pub Date : 2018-08-01DOI: 10.1109/ICCUBEA.2018.8697796
Shivaji Manik Awchar, S. Diwan, Pratik Arlikar
This paper presents a unique methodology for sensor-less BLDC motors in electric vehicles. The state-of-art ZCD detection method enables high performance specially for high-speed range because the relationship between magnitude of Back-EMF and rotor speed is directly proportional. This research work proposes a new solution for determination of rotor position by implementing Back-EMF observer over wide speed range. In this case the observer is designed using motor basic equations which results in high performance at near zero speed as well as on full speed range. Also, the rotor position identified is found to be independent of rotor speed. Moreover, the solution does require any additional circuitry in comparison to ZCD detection method. Additionally, this technique is executed using MATLAB for three/four-wheeler based electric vehicles (48V).
{"title":"Advanced Technique for Speed Control Of Sensor-Less BLDC Motor","authors":"Shivaji Manik Awchar, S. Diwan, Pratik Arlikar","doi":"10.1109/ICCUBEA.2018.8697796","DOIUrl":"https://doi.org/10.1109/ICCUBEA.2018.8697796","url":null,"abstract":"This paper presents a unique methodology for sensor-less BLDC motors in electric vehicles. The state-of-art ZCD detection method enables high performance specially for high-speed range because the relationship between magnitude of Back-EMF and rotor speed is directly proportional. This research work proposes a new solution for determination of rotor position by implementing Back-EMF observer over wide speed range. In this case the observer is designed using motor basic equations which results in high performance at near zero speed as well as on full speed range. Also, the rotor position identified is found to be independent of rotor speed. Moreover, the solution does require any additional circuitry in comparison to ZCD detection method. Additionally, this technique is executed using MATLAB for three/four-wheeler based electric vehicles (48V).","PeriodicalId":422920,"journal":{"name":"2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122687839","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}
Pub Date : 2018-08-01DOI: 10.1109/ICCUBEA.2018.8697589
R. Bharati, V. Attar
Distributed Transactional systems need to be fast and scalable to increase the performance of the system. Many distributed databases achieve high throughput and scalability through data partitioning. The paper presents comprehensive survey of different techniques and parameters related to distributed transactions used in the Distributed databases. It specifies different ways of partitioning algorithms. The purpose of this review is to study the techniques of horizontal partitioning like range partitioning, schema level partitioning, graph level partitioning etc. giving the high performance and availability of data. The vital aim of these techniques is to reduce distributed transaction and increase scalability of distributed databases.
{"title":"A Comprehensive Survey on Distributed Transactions Based Data Partitioning","authors":"R. Bharati, V. Attar","doi":"10.1109/ICCUBEA.2018.8697589","DOIUrl":"https://doi.org/10.1109/ICCUBEA.2018.8697589","url":null,"abstract":"Distributed Transactional systems need to be fast and scalable to increase the performance of the system. Many distributed databases achieve high throughput and scalability through data partitioning. The paper presents comprehensive survey of different techniques and parameters related to distributed transactions used in the Distributed databases. It specifies different ways of partitioning algorithms. The purpose of this review is to study the techniques of horizontal partitioning like range partitioning, schema level partitioning, graph level partitioning etc. giving the high performance and availability of data. The vital aim of these techniques is to reduce distributed transaction and increase scalability of distributed databases.","PeriodicalId":422920,"journal":{"name":"2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121163436","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}
Pub Date : 2018-08-01DOI: 10.1109/ICCUBEA.2018.8697221
S. S. Bere
To explain and detect different features in images scale-invariant feature transform can be used effectively. From starting, a set of reference images SIFT important points of objects are extracted and stored in a database. An object in a new image can be recognized by individually balancing each feature from the new image to this database and then finding features for candidate matching. As a valuable local SIFT can be utilize as a solution point descriptor for its invariance to lighting, scale, and rotation changes in images. Since SIFT is not flip invariant, flip invariant SIFT is planned. These F-SIFT is established to identify large scale duplicate videos, object finding as well as recognition. It requires to take out all the frames from query video and videos in dataset for similarity matching, time complexity of f-SIFT is more, So to remove such limitation we have projected dual threshold technique. Our method will eliminate redundant video frames by applying auto dual threshold method. So there will be no necessity to execute the extraction of features and matching of sequence with all video frames. Unnecessary frames are detached by making segments of video. Only the key frames are extracted for matching purposes. Here we are using two thresholds. The first is for identifying direct changes of visual information of extracting frames and other second for detecting usual changes of visual information of extracting frames. Threshold values are decided as per the information of the video. This system, extracting total three frames like first frame, last frame and key frame from video segment. By using the average feature value of all the frames in the segment, key frames are decided. For similar propose a key frame is used and remaining two frames are used to detect the segment location.
{"title":"Duplicate Video and Object Detection by Video Key Frame Using F-SIFT","authors":"S. S. Bere","doi":"10.1109/ICCUBEA.2018.8697221","DOIUrl":"https://doi.org/10.1109/ICCUBEA.2018.8697221","url":null,"abstract":"To explain and detect different features in images scale-invariant feature transform can be used effectively. From starting, a set of reference images SIFT important points of objects are extracted and stored in a database. An object in a new image can be recognized by individually balancing each feature from the new image to this database and then finding features for candidate matching. As a valuable local SIFT can be utilize as a solution point descriptor for its invariance to lighting, scale, and rotation changes in images. Since SIFT is not flip invariant, flip invariant SIFT is planned. These F-SIFT is established to identify large scale duplicate videos, object finding as well as recognition. It requires to take out all the frames from query video and videos in dataset for similarity matching, time complexity of f-SIFT is more, So to remove such limitation we have projected dual threshold technique. Our method will eliminate redundant video frames by applying auto dual threshold method. So there will be no necessity to execute the extraction of features and matching of sequence with all video frames. Unnecessary frames are detached by making segments of video. Only the key frames are extracted for matching purposes. Here we are using two thresholds. The first is for identifying direct changes of visual information of extracting frames and other second for detecting usual changes of visual information of extracting frames. Threshold values are decided as per the information of the video. This system, extracting total three frames like first frame, last frame and key frame from video segment. By using the average feature value of all the frames in the segment, key frames are decided. For similar propose a key frame is used and remaining two frames are used to detect the segment location.","PeriodicalId":422920,"journal":{"name":"2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130062850","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}
Pub Date : 2018-08-01DOI: 10.1109/ICCUBEA.2018.8697740
Sudeep D. Thepade, Madhura Kalbhor
Social networking sites have given rise to tremendous volume of images, which implies need of proper organisation of image databases with efficient retrival and categorisation mechanisms. Images if stored in appropriate fashion may help to reteive them fastly and correctly as and when required. Image category prediction with the proposed machine learning based approach palys an important role in visual content based image category prediction for efficient handling of voluminous data. The system uses the content as transformed fractional coeficients to generate the feature vector for image class peridication using Sine and Hartley transforms. Machine learning algoritms alias Random Forest, SVM, Simple logicts are employed for proposed image category prediction method. The paper also proposes ensembling of these machine learning alogorithms with majority voting at decision level for improved image category prediction. The experimentation is carried out on the fraction of the standard image dataset. The result analysis show that the fractional transform coefficients gives the capability for better discrimination than that of consideration of all transformed coefficients considered to form feature vector; as indicated by higher image category prediction accuracy values. Also it has been observed that ensembling of machine learning algorithms (Random Forest, Simple Logistic and SVM) has given best classification accuracy of 72.91% with Hartley transformed fractional content as features.
{"title":"Ensemble of Machine Learning Classifiers for Improved Image Category Prediction Using Fractional Coefficients of Hartley and Sine Transforms","authors":"Sudeep D. Thepade, Madhura Kalbhor","doi":"10.1109/ICCUBEA.2018.8697740","DOIUrl":"https://doi.org/10.1109/ICCUBEA.2018.8697740","url":null,"abstract":"Social networking sites have given rise to tremendous volume of images, which implies need of proper organisation of image databases with efficient retrival and categorisation mechanisms. Images if stored in appropriate fashion may help to reteive them fastly and correctly as and when required. Image category prediction with the proposed machine learning based approach palys an important role in visual content based image category prediction for efficient handling of voluminous data. The system uses the content as transformed fractional coeficients to generate the feature vector for image class peridication using Sine and Hartley transforms. Machine learning algoritms alias Random Forest, SVM, Simple logicts are employed for proposed image category prediction method. The paper also proposes ensembling of these machine learning alogorithms with majority voting at decision level for improved image category prediction. The experimentation is carried out on the fraction of the standard image dataset. The result analysis show that the fractional transform coefficients gives the capability for better discrimination than that of consideration of all transformed coefficients considered to form feature vector; as indicated by higher image category prediction accuracy values. Also it has been observed that ensembling of machine learning algorithms (Random Forest, Simple Logistic and SVM) has given best classification accuracy of 72.91% with Hartley transformed fractional content as features.","PeriodicalId":422920,"journal":{"name":"2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127172595","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：