Informatics Publishing Limited

Parul D. Sindha ¹, Dipti M. Shah ², Atul Patel ³

Affiliations
1 CHARUSAT Changa, Gujarat - 388421, IN
2 G. H. Patel Post Graduate Department of Science & Technology, Sardar Patel University, V.V.Nagar - 388120, Gujarat, IN
3 CMPICA, CHARUSAT Changa - 388421, Gujarat, IN

Abstract

Background/Objectives: Traffic sign recognition system is a multimedia based real time which guides and assist the driver visually with audio to decrease the road accidents. Methods/Statistical analysis: Hundreds of small pictures are registered by the regional transportation authority of every country which guides the drivers as well as road users by providing information related to current status of the road. All most in all countries red and blue color with a particular shape like triangle, circle and rectangle are used to prepare small pictorial signs. In proposed research, features of color and shape are used to detect, track and recognize traffic signs. Templates are prepared using canny images and centroid of captured sign from the video. Such templates are matched with the knowledgebase by translating images according to the centroid point of a shape. Findings: Proposed research paper introduces the traffic sign recognition system with multiple output formats like video and audio with the textual information of the tracked signs. Analysis of the developed system is focuses in this paper. Each prohibitory, obligation, cautionary and informatory signs are tested to check the robustness of the system during number of experiments. According to the analysis report more the 100 signs are tested and the performance of the proposed research is nearly 93%. Knowledgebase has been prepared for all traffic signs. Application/ Improvements: The proposed system produces more than 93% which can be improved to get exactly 100% accuracy level.

Keywords

Centroid, DAS, Knowledgebase, Sign Recognition, Translation

Full Text

   

Nirali Honest ¹, Bankim Patel ², Atul Patel ¹

Affiliations
1 Smt. Chandaben Mohanbhai Patel Institute of Computer Applications, Charotar University of Science and Technology, (CHARUSAT), Changa - 388421, Gujarat, IN
2 Shrimad Rajchandra Institute of Management and Computer Application, Uka Tarsadia University, Tarsadi - 394350, Gujarat, IN

Abstract

Background/Objectives: The website is composed of permanent and temporary pages. Deriving a prediction model which considers the dynamic pages generated on the website requires to consider new aspects. Methods/Statistical Analysis: We adopt supervised learning models as they give better prediction results for new input data. After reading the log files and applying preprocessing, we build the user navigation patterns and then apply the prediction of pages. The main parameter on which we have modified is the time stamp. In earlier approaches the time stamp was divided into day, month and year and based on the timestamp granule selected the prediction model was formed. In our work we consider the same granule with introduction to new timestamp namely event. Findings: Markov model are very good in predicting the pages for n length, but the model doesn’t focus on temporal aspect for prediction. Temporal n-gram model covers the temporal aspect of prediction by forming the granules of time. This model gives good accuracy in predicting pages that are permanent for any given website, but doesn’t tend to be good for pages that are temporary in nature. Our model focuses on temporal aspect for both types of pages by creating an event based temporal n-gram model. Event means creating a special named interval for which the pages are made available on the website. This means that after the interval specified in the event the page will be no more visible on the website. The pages are predicted based on the nature of pages, we form broadly two types of nature of pages 1. Regular for permanent pages and 2. Event for temporary pages. By introducing this temporal aspect the prediction algorithm considers the specified interval only for the event specific pages, after the interval is over the pages are not considered for prediction. Application/Improvements: Specifying events help to derive better accuracy in prediction when we consider permanent and temporary pages, as we predict the pages based on the condition whether they are regular or event based pages.

Keywords

Classification Algorithms, Conditional Probability, Event Based Granule Model, Naive Baysian, Prediction Model

Full Text

   

Chirag Indravadanbhai Patel ¹, Dipti Shah ², Atul Patel ¹

Affiliations
1 Smt. Chandaben Mohanbhai Patel Institute of Computer Applications, CHARUSAT, Changa - 388421, Gujarat, IN
2 G. H. Patel Post Graduate Department of Science and Technology, Sardar Patel University, V. V. Nagar - 388120, Gujarat, IN

Abstract

Background/Objectives:In this paper a novel number plate segmentation algorithm is proposed, which can be further exploited to do character segmentation and character recognition. Methods/Statistical Analysis: The scope of this research is to segment vehicle number plate from the captured vehicle image. To accomplish this task image binarization is used as it translates any color image into black and white image. The image binarization process is carried based on the statistical formula mentioned in this paper. The binarized image is further processed to remove unwanted area and exactly segment vehicle number plate. Findings: The algorithm provides overall accuracy of ~97.67% with the average processing time of 134ms for 250 images captured during timings in day and night. The system works well in different illumination conditions. Conclusion/Improvements: The overall accuracy can be further improved by modifying the parameters of image binarization process.

Keywords

Column Trimming, Image Binarization, Image Segmentation, Number Plate, Row Trimming

Full Text

   

Pranav Vyas ¹, Bhushan Trivedi ², Atul Patel ¹

Affiliations
1 Smt. Chandaben Mohanbhai Patel Institute of Computer Applications, Charotar University of Science and Technology, Anand - 388421, Gujarat, IN
2 GLS Institute of Computer Technology, GLS University, Ahmedabad - 388421, Gujarat, IN

Abstract

Background/Objectives: Cryptographic protocols are used for securing information when transmitting it over insecure network such as Internet. This paper’s objective is to study recently proposed key exchange protocols for mobile environment. Methods/Statistical Analysis: In this paper we do a literature survey of recently proposed key exchange protocols for mobile environment. We analyze execution of protocol in three phases i.e. initialization, communication, renewal/ termination phase. In initialization protocol prepares for key exchange process. Next, protocol actually communicates with others to exchange secret key. Third protocol may terminate or renew connection for further communication. We also study activities done by protocols that define characteristics of protocol. Findings: In this paper we find that there are many parameters to consider when designing a key exchange protocols for mobile environment. However, significance of parameters is different, based on the security requirement of application for which protocol is being developed. Strength of a protocol is in the encryption technique that it uses. Hence, stronger encryption techniques results in better security of protocol. Speed of protocol is another important parameter. Length of steps in algorithm of protocol is directly proportional to its speed. A protocol must be able to withstand various attacks on it. A protocol should have high reliability if it is to be used in handling critical data. We found that modern key exchange protocols are not properly analyzed and tested before being proposed. Instead of working on already proposed protocols and solve their vulnerabilities and strengthening them researchers are proposing new protocols without testing them properly for vulnerabilities which are later exploited by malicious users. Applications/Improvements: This research paper will help researchers and protocol designers. It will give them idea about design parameters when designing key exchange protocol. It will enable them to take better decisions.

Keywords

Data Security, Key Agreement, Key Exchange, Key Management, Mobile Communication, Wireless Communication

Full Text

   

Jaimin N. Undavia ¹, Prashant Dolia ², Atul Patel ³

Affiliations
1 Faculty of Computer Science and Applications, Charotar University of Science And Technology (CHARUSAT), Changa - 388421, Gujarat, IN
2 Department of Computer Science and Applications, Maharaja Krishnakumarsinhji Bhavnagar University, Bhvnagar - 364001, Gujarat, IN
3 CharotarUniversity of Science and Technology (CHARUSAT), Changa - 388421, Gujarat, IN

Abstract

Background/Objectives: Excellence of Universities is based on students' success in their academic and it is possible if the students are instructed or counseled before getting admitted in their post graduation. So, we have developed a model for the post graduating students to utilize their intelligence in right direction. Methods/Statistical Analysis: If students are given admission in right course then their academic success is guaranteed by the university. To formulate the prediction, decision tree classifiers are best suitable as it has potential to generate comprehensible output. It is generating the tree and rules which will be used to formulate the predictions. Hence, this approach is of two steps approach known as training phase and testing phase. Findings: The model trains on the basis of the defined instances and from the defined instances the classified builds the rules. These rules are used to formulate prediction for unknown valued instances. This article depicts the customized classification model to predict the Post-Graduation degree of the students. The model is based on J48 decision tree algorithm for classification. The model is trained by the data collected through survey of different institutions with the purpose of differentiating and predicting students' choice and to generate unbiased result. We obtained certain patterns of the students preferences to select their post graduation course. On the basis of such rules which are derived from historical data, are used to predict post graduation course for unknown instance. We have used J48 classification algorithm for decision tree to predict the post graduation course based on their academic history and other identified parameters. We have identified total 14 parameters to predict the class label of 15thattribute. Applications/Improvements: We have customized a model using Weka which uses the J48 algorithm to predict students' post graduation degree. We have obtained 94.03% accuracy of prediction against 4 classes as final attribute.

Keywords

Classification, Customization in Weka, Post Graduation Course Selection, Prediction Model, Weka

Full Text

   

Jai Prakash Verma ¹, Atul Patel ²

Affiliations
1 Department of CE, Institute of Technology, Nirma University, Ahmedabad – 382481, Gujarat, IN
2 CMPICA, CHARUSAT University, Changa – 388421, Gujarat, IN

Abstract

Background/Objectives: Supervised techniques uses human generated summary to select features and parameter for summarization. The main problem in this approach is reliability of summary based on human generated parameters and features. Many researches have shown the conflicts in summary generated. Due to diversity of large scale datasets, supervised techniques based summarization also fails to meet the requirements. Big data analytics for text dataset also recommends unsupervised techniques than supervised techniques. Unsupervised techniques based summarization systems finds representative sentences from large amount of text dataset. Methods/Statistical Analysis: Co-selection based evaluation measure is applied for evaluating the proposed research work. The value of recall, precision, f-measure and similarity measure are determined for concluding the research outcome for the respective objective. Findings: The algorithms like KMeans, MiniBatchKMeans, and Graph based summarization techniques are discussed with all technical details. The results achieved by applying Graph Based Text Summarization techniques with large scale review and feedback data found improvement over previously published results based on sentence scoring using TF and TF-IDF. Graph based sentence scoring method is much efficient than other unsupervised learning techniques applied for extractive text summarization. Application/Improvements: The execution of graph based algorithm with Spark's Graph X programming environment will secure execution time for this types of large scale review and feedback dataset which is considered under Big Data Problem.

Keywords

Extractive Text Summarization, Sentence Scoring Methods, Unsupervised Learning.

Full Text

   

Sanskruti Patel ¹, Atul Patel ¹

Affiliations
1 Faculty of Computer Science and Applications, Charotar University of Science and Technology (CHARUSAT), Changa - 388421, Gujarat, IN

Abstract

Background/Objectives: With the growth in complexity and volume of medical data, an extensive set of information currently available in various forms related to diseases and its symptoms. Mechanisms are necessary to extract rules and patterns from these massive set of data. Identification and extraction of hidden patterns and rules in this massive data set certainly help us to understand about diseases progression facts. Methods: Machine learning provides an automatic way to uncover the patterns from data set and it will be helpful to health care professionals in order to provide precision medicine to their patients. Artificial Neural Network is a popular machine learning technique used for classification tasks in medical diagnosis for diseases detection. It is an eminent field of computer science which can be applied to the health care sector quite efficiently. In this study, Multi-Layer Feed Forward Neural Network has been applied to the dermatology dataset downloaded from UCI repository site to classify the dermatology diseases. Findings: Artificial Neural Network with back propagation algorithm produces the optimum results for classification and prediction problems. It also possesses the ability of generalization and applicable to real world problem. Applications: The experiment will be extended by applying on other types of diseases datasets and an automated diagnostic and advisory system with neural network integration definitely helps in diseases prediction problem.

Keywords

Artificial Neural Network, Classification, Disease Diagnosis

Full Text

   

Kalpit G. Soni ¹, Atul Patel ¹

Affiliations
1 Charotar University of Science and Technology, Changa - 388421, Gujarat, IN

Abstract

Background/Objectives: The emergence of Cloud computing in recent times as a globally accepted data and activity management platform is phenomenal, albeit with few clutches not only related with the data integrity and their privacy but also of their geographical location. The gravity of the issue arises with the plethora of guidelines and variations in legal procedures in storing the data and getting access to them in diverse geographical areas. The aim of present investigation is to figure-out a potentially effective method that can help detect a forged server with better location coordinates than the currently employed methods. Methods/Statistical Analysis: In this study we developed a made use of an established method to check the position of the files at a specified place as claimed by service provider. The main means employed during the course of this investigation is K-Medoid Algorithm for finding the server's geographical location putting emphasis on putative adverse outcome. Findings: The greatness of K-Medoid Algorithm is to detect a potentially forged location and also to identify the possible elements involved in forgery. This work involved the use of Tulip Dataset and Implementation in NS3 for validating algorithm modified by us to detect the forged location and possible location of involved adversaries. The investigation during this work culminated in proposing a modified K-Medoid algorithm for helping a user to find the Cloud file system which verifies the storage location with very high efficiency. Improvements/Applications: Through our research technique, user able to detect faked location and identify the dishonest service provider. Our work culminated in the best estimation of the location of server for a cloud service provider within a less distance than the previous work.

Keywords

Cloud Computing, Geolocation, K-Medoid, Trilateration.

Full Text