Pub Date : 2016-05-20DOI: 10.1109/RTEICT.2016.7807830
Prajakta Pawaskar, P. Chawande, Seema Jadhav
A three phase digital energy meter developed using Digital Signal Processor (DSP), for measurement of energy consumed by three phase load, ensures energy conservation and maintains high accuracy. This paper presents a simple and cost effective design of three phase digital energy meter, without compromising the power measurement quality. It uses a voltage and current signal conditioning circuit and antialising filter, to sense system voltage and current. The power measurement is done by multiplication of instantaneous samples/values of current i(t) and voltage v(t) without need of a separate power factor measurement circuit. For power measurement, definitions in IEEE standard considered. The analog circuit with antialiasing filter is tested and DSP interface is developed, in a laboratory setup.
{"title":"Design and implementation of low cost three phase energy meter","authors":"Prajakta Pawaskar, P. Chawande, Seema Jadhav","doi":"10.1109/RTEICT.2016.7807830","DOIUrl":"https://doi.org/10.1109/RTEICT.2016.7807830","url":null,"abstract":"A three phase digital energy meter developed using Digital Signal Processor (DSP), for measurement of energy consumed by three phase load, ensures energy conservation and maintains high accuracy. This paper presents a simple and cost effective design of three phase digital energy meter, without compromising the power measurement quality. It uses a voltage and current signal conditioning circuit and antialising filter, to sense system voltage and current. The power measurement is done by multiplication of instantaneous samples/values of current i(t) and voltage v(t) without need of a separate power factor measurement circuit. For power measurement, definitions in IEEE standard considered. The analog circuit with antialiasing filter is tested and DSP interface is developed, in a laboratory setup.","PeriodicalId":6527,"journal":{"name":"2016 IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT)","volume":"34 2 1","pages":"296-300"},"PeriodicalIF":0.0,"publicationDate":"2016-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83545613","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 : 2016-05-20DOI: 10.1109/RTEICT.2016.7808177
Sangeeta D. Palekar, N. Narkhede
Many fields of science, engineering, finance, mathematical optimization methods, Artificial Neural Networks, signal and image processing algorithms requires the operations and manipulations of real numbers. Floating-point operations are most extensively adopted approach for exploiting real numbers. The speed of Floating-point arithmetic unit is very crucial performance parameter which impinges the operation of the system. On that account a 32 bit floating point arithmetic unit is designed for different applications which insists for eminent speed. The intent of this design is to reduce the area and combinational path delay to enhance the speed of operation which is attained by parallelism in multiplier which is used for mantissa multiplication. For Floating-point multiplier Booth recoded multiplier is used where the number of partial product are reduced which in turns boost the speed of multiplication. The proposed module is implemented on Spartan 6 FPGA. Performance of the floating point arithmetic unit is compared with latest research papers regarding delay and it is ascertained that there is 59% of optimization in critical path delay of floating point multiplier and 50 % of optimization of floating point adder. The result illustrates that proposed arithmetic unit has a great impact on convalescent the speed and area of the design.
{"title":"High speed and area efficient single precision floating point arithmetic unit","authors":"Sangeeta D. Palekar, N. Narkhede","doi":"10.1109/RTEICT.2016.7808177","DOIUrl":"https://doi.org/10.1109/RTEICT.2016.7808177","url":null,"abstract":"Many fields of science, engineering, finance, mathematical optimization methods, Artificial Neural Networks, signal and image processing algorithms requires the operations and manipulations of real numbers. Floating-point operations are most extensively adopted approach for exploiting real numbers. The speed of Floating-point arithmetic unit is very crucial performance parameter which impinges the operation of the system. On that account a 32 bit floating point arithmetic unit is designed for different applications which insists for eminent speed. The intent of this design is to reduce the area and combinational path delay to enhance the speed of operation which is attained by parallelism in multiplier which is used for mantissa multiplication. For Floating-point multiplier Booth recoded multiplier is used where the number of partial product are reduced which in turns boost the speed of multiplication. The proposed module is implemented on Spartan 6 FPGA. Performance of the floating point arithmetic unit is compared with latest research papers regarding delay and it is ascertained that there is 59% of optimization in critical path delay of floating point multiplier and 50 % of optimization of floating point adder. The result illustrates that proposed arithmetic unit has a great impact on convalescent the speed and area of the design.","PeriodicalId":6527,"journal":{"name":"2016 IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT)","volume":"55 1","pages":"1950-1954"},"PeriodicalIF":0.0,"publicationDate":"2016-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83901579","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 : 2016-05-20DOI: 10.1109/RTEICT.2016.7808140
Kuntal Kumar Pal, K. S. Sudeep
In recent times, the Convolutional Neural Networks have become the most powerful method for image classification. Various researchers have shown the importance of network architecture in achieving better performances by making changes in different layers of the network. Some have shown the importance of the neuron's activation by using various types of activation functions. But here we have shown the importance of preprocessing techniques for image classification using the CIFAR10 dataset and three variations of the Convolutional Neural Network. The results that we have achieved, clearly shows that the Zero Component Analysis(ZCA) outperforms both the Mean Normalization and Standardization techniques for all the three networks and thus it is the most important preprocessing technique for image classification with Convolutional Neural Networks.
{"title":"Preprocessing for image classification by convolutional neural networks","authors":"Kuntal Kumar Pal, K. S. Sudeep","doi":"10.1109/RTEICT.2016.7808140","DOIUrl":"https://doi.org/10.1109/RTEICT.2016.7808140","url":null,"abstract":"In recent times, the Convolutional Neural Networks have become the most powerful method for image classification. Various researchers have shown the importance of network architecture in achieving better performances by making changes in different layers of the network. Some have shown the importance of the neuron's activation by using various types of activation functions. But here we have shown the importance of preprocessing techniques for image classification using the CIFAR10 dataset and three variations of the Convolutional Neural Network. The results that we have achieved, clearly shows that the Zero Component Analysis(ZCA) outperforms both the Mean Normalization and Standardization techniques for all the three networks and thus it is the most important preprocessing technique for image classification with Convolutional Neural Networks.","PeriodicalId":6527,"journal":{"name":"2016 IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT)","volume":"2014 1","pages":"1778-1781"},"PeriodicalIF":0.0,"publicationDate":"2016-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87941971","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 : 2016-05-20DOI: 10.1109/RTEICT.2016.7808199
Shruti Gupta, B. Prince
Wireless sensor network (WSN) consists of numerous sensor nodes which are deployed to inspect a certain subject. As a stream of packets from the source node is sent to the sink, the adversary may be able to trace the messages and try to find out the source location. While monitoring the target, the revealed information about the subject can be misused by the adversary. Hence, our aim is to hide the source location from the adversary. It is quite difficult to efficiently achieve the source location privacy although the confidentiality of the messages can be well assured through data encryption. In WSN, the source location privacy is more complex due to the fact that the nodes comprise low-power and low-cost devices. The protocols based on large-scale communications are not suitable for WSN. In this paper, we provide a survey of the solutions which achieves source location privacy. We discuss the concept of source location privacy with different metrics such as message latency, safety period and capture likelihood. Then, a summarized view of the solutions with their limitations as found in the literature is presented. We mention an overview of each solution in relation with their assumptions for adversary model.
{"title":"Preserving privacy of source location using random walk: A survey","authors":"Shruti Gupta, B. Prince","doi":"10.1109/RTEICT.2016.7808199","DOIUrl":"https://doi.org/10.1109/RTEICT.2016.7808199","url":null,"abstract":"Wireless sensor network (WSN) consists of numerous sensor nodes which are deployed to inspect a certain subject. As a stream of packets from the source node is sent to the sink, the adversary may be able to trace the messages and try to find out the source location. While monitoring the target, the revealed information about the subject can be misused by the adversary. Hence, our aim is to hide the source location from the adversary. It is quite difficult to efficiently achieve the source location privacy although the confidentiality of the messages can be well assured through data encryption. In WSN, the source location privacy is more complex due to the fact that the nodes comprise low-power and low-cost devices. The protocols based on large-scale communications are not suitable for WSN. In this paper, we provide a survey of the solutions which achieves source location privacy. We discuss the concept of source location privacy with different metrics such as message latency, safety period and capture likelihood. Then, a summarized view of the solutions with their limitations as found in the literature is presented. We mention an overview of each solution in relation with their assumptions for adversary model.","PeriodicalId":6527,"journal":{"name":"2016 IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT)","volume":"30 1","pages":"2047-2051"},"PeriodicalIF":0.0,"publicationDate":"2016-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80410792","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 : 2016-05-20DOI: 10.1109/RTEICT.2016.7808214
C. Satish, B. Girija, K. Zumair, S. A. Kumar, K. Ramesh, K. A. Vishnumurthy
Milk is considered to be a complete food an disconsumed by individuals of all age groups. Urea is added to milk as an adulterant to increase its shelf life and Solid Not Fat (SNF) value. Excess urea consumption is harmful to human beings which results in serious health issues. Hence, there is an utmost necessity of an instrument to detect and quantify the presence of urea content in milk. ZnO nano powder was synthesized using Sol-Gel process and it was coated on a glass substrate in the form of thick film using the Doctor Blade method. The ZnO immobilized with urease enzyme was used as a sensing layer for detection of urea in milk. The fabricated sensor has a sensitivity of 0.6mV/10mg and the designed instrument indicates the presence of urea within the permissible range by corresponding LED.
{"title":"Design of portable electronic instrument for detection of urea in milk by fabrication of urea sensor using ZnO nanoparticles","authors":"C. Satish, B. Girija, K. Zumair, S. A. Kumar, K. Ramesh, K. A. Vishnumurthy","doi":"10.1109/RTEICT.2016.7808214","DOIUrl":"https://doi.org/10.1109/RTEICT.2016.7808214","url":null,"abstract":"Milk is considered to be a complete food an disconsumed by individuals of all age groups. Urea is added to milk as an adulterant to increase its shelf life and Solid Not Fat (SNF) value. Excess urea consumption is harmful to human beings which results in serious health issues. Hence, there is an utmost necessity of an instrument to detect and quantify the presence of urea content in milk. ZnO nano powder was synthesized using Sol-Gel process and it was coated on a glass substrate in the form of thick film using the Doctor Blade method. The ZnO immobilized with urease enzyme was used as a sensing layer for detection of urea in milk. The fabricated sensor has a sensitivity of 0.6mV/10mg and the designed instrument indicates the presence of urea within the permissible range by corresponding LED.","PeriodicalId":6527,"journal":{"name":"2016 IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT)","volume":"27 1","pages":"2120-2124"},"PeriodicalIF":0.0,"publicationDate":"2016-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78137682","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 : 2016-05-20DOI: 10.1109/RTEICT.2016.7808087
Krishnat B. Pawar, S. Nalbalwar
Morphological Filter based Distributed Canny edge detection algorithm for Raspberry Pi platform using Simulink model is presented in this paper. Traditional canny edge detection algorithm uses frame based statistics which gives high accuracy but computationally more complex. Also canny algorithm is more sensitive to noise. In this experiment, an attempt is made to make canny algorithm more robust to noise using morphological filtering. Here canny algorithm is implemented at block level without any compromise in edge detection performance. If frame level statistics are used for threshold selection, it would result in either loss of edges or surplus edge detection. To solve this problem threshold selection is made based on type of block. Smooth and texture pixel counts are calculated for image block. Instead of using probability, actual pixel counts are used to calculate threshold. This makes threshold selection block more adaptive. Finally, objective analysis is carried out which shows proposed block based distributed algorithm is better than traditional frame based algorithm, especially in presence of impulse noise.
{"title":"Distributed canny edge detection algorithm using morphological filter","authors":"Krishnat B. Pawar, S. Nalbalwar","doi":"10.1109/RTEICT.2016.7808087","DOIUrl":"https://doi.org/10.1109/RTEICT.2016.7808087","url":null,"abstract":"Morphological Filter based Distributed Canny edge detection algorithm for Raspberry Pi platform using Simulink model is presented in this paper. Traditional canny edge detection algorithm uses frame based statistics which gives high accuracy but computationally more complex. Also canny algorithm is more sensitive to noise. In this experiment, an attempt is made to make canny algorithm more robust to noise using morphological filtering. Here canny algorithm is implemented at block level without any compromise in edge detection performance. If frame level statistics are used for threshold selection, it would result in either loss of edges or surplus edge detection. To solve this problem threshold selection is made based on type of block. Smooth and texture pixel counts are calculated for image block. Instead of using probability, actual pixel counts are used to calculate threshold. This makes threshold selection block more adaptive. Finally, objective analysis is carried out which shows proposed block based distributed algorithm is better than traditional frame based algorithm, especially in presence of impulse noise.","PeriodicalId":6527,"journal":{"name":"2016 IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT)","volume":"1 1","pages":"1523-1527"},"PeriodicalIF":0.0,"publicationDate":"2016-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82632594","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 : 2016-05-20DOI: 10.1109/RTEICT.2016.7808070
S. Srisruthi, N. Swarna, G. M. Susmitha Ros, E. Elizabeth
Agriculture requires the dedication of many natural resources, including land, water, and energy. The quality and quantity of these natural resources has degraded over the years due to various economic problems associated with increased cost of inputs, decreased farm incomes, ever declining land, labour, energy resources and also ecological problems such as soil, water pollution and soil erosion, putting the viability of future agricultural operations at risk. The remedy to this is to adopt sustainable agriculture which supports careful management and cultivation of crops involving less use of fertilizer, pesticides, calculated use of precious natural resources like energy, water through controlled irrigation and fertigation practices with the help of green sensor technology and electronic control systems. The paper provides efficient automated farm monitoring and irrigation techniques which incorporate wide range of sensors to remotely sense and monitor various parameters of the soil like temperature, moisture, fertility and regulate the supply of water and fertilizer to the land based on the requirement. An algorithm formulated with the threshold values of sensor outputs is used to code the microcontroller which performs the required actions by employing relays until the strayed-out parameter has been brought back to its optimum level. The cloud based user friendly interface facilitates real-time data logging of environmental parameters while also supporting analysis of past statistics for future growth by means of a web-based customizable application. Furthermore, the project aims to optimize the use of land and labour, conserve water, increase crop yield, avoid wastage of energy and provide maximum automation and benefit the society by adopting smart environment friendly technology to implement newer and sustainable ways of agriculture.
{"title":"Sustainable agriculture using eco-friendly and energy efficient sensor technology","authors":"S. Srisruthi, N. Swarna, G. M. Susmitha Ros, E. Elizabeth","doi":"10.1109/RTEICT.2016.7808070","DOIUrl":"https://doi.org/10.1109/RTEICT.2016.7808070","url":null,"abstract":"Agriculture requires the dedication of many natural resources, including land, water, and energy. The quality and quantity of these natural resources has degraded over the years due to various economic problems associated with increased cost of inputs, decreased farm incomes, ever declining land, labour, energy resources and also ecological problems such as soil, water pollution and soil erosion, putting the viability of future agricultural operations at risk. The remedy to this is to adopt sustainable agriculture which supports careful management and cultivation of crops involving less use of fertilizer, pesticides, calculated use of precious natural resources like energy, water through controlled irrigation and fertigation practices with the help of green sensor technology and electronic control systems. The paper provides efficient automated farm monitoring and irrigation techniques which incorporate wide range of sensors to remotely sense and monitor various parameters of the soil like temperature, moisture, fertility and regulate the supply of water and fertilizer to the land based on the requirement. An algorithm formulated with the threshold values of sensor outputs is used to code the microcontroller which performs the required actions by employing relays until the strayed-out parameter has been brought back to its optimum level. The cloud based user friendly interface facilitates real-time data logging of environmental parameters while also supporting analysis of past statistics for future growth by means of a web-based customizable application. Furthermore, the project aims to optimize the use of land and labour, conserve water, increase crop yield, avoid wastage of energy and provide maximum automation and benefit the society by adopting smart environment friendly technology to implement newer and sustainable ways of agriculture.","PeriodicalId":6527,"journal":{"name":"2016 IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT)","volume":"8 1","pages":"1442-1446"},"PeriodicalIF":0.0,"publicationDate":"2016-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88894488","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 : 2016-05-20DOI: 10.1109/RTEICT.2016.7807919
Satish D. Pore, B. Momin
In video surveillance to count the number of people walking through a corridor or door people counting is used. Counting persons passing through supervised area is important issue of this domain. System of people counting is important for security application (such as in the situation of an evacuation, it is vital to know how many people are present inside the surveillance area at any given time), pedestrian traffic management, tourists flow estimation. The counter requires two steps: detection of people and tracking to count people directionally. Detection is based on finding people through application of HOG descriptor and trajectories of people are generated through application of Kalman channel. Finally counting system modifies in and out count based on direction of trajectories generated by kalman filter. Different indoor video sequences are taken from surveillance camera and results obtained with the accuracy ranging from 91% to 100%. Accuracies of result depends on number of people crossing the counting zone simultaneously, occlusions, intensity variation in video sequence etc.
{"title":"Bidirectional people counting system in video surveillance","authors":"Satish D. Pore, B. Momin","doi":"10.1109/RTEICT.2016.7807919","DOIUrl":"https://doi.org/10.1109/RTEICT.2016.7807919","url":null,"abstract":"In video surveillance to count the number of people walking through a corridor or door people counting is used. Counting persons passing through supervised area is important issue of this domain. System of people counting is important for security application (such as in the situation of an evacuation, it is vital to know how many people are present inside the surveillance area at any given time), pedestrian traffic management, tourists flow estimation. The counter requires two steps: detection of people and tracking to count people directionally. Detection is based on finding people through application of HOG descriptor and trajectories of people are generated through application of Kalman channel. Finally counting system modifies in and out count based on direction of trajectories generated by kalman filter. Different indoor video sequences are taken from surveillance camera and results obtained with the accuracy ranging from 91% to 100%. Accuracies of result depends on number of people crossing the counting zone simultaneously, occlusions, intensity variation in video sequence etc.","PeriodicalId":6527,"journal":{"name":"2016 IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT)","volume":"53 1","pages":"724-727"},"PeriodicalIF":0.0,"publicationDate":"2016-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73037288","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 : 2016-05-20DOI: 10.1109/RTEICT.2016.7807824
Anup Rokade, Akshay Malhotra, Ankita S. Wanchoo
The Indian Stock market, with over 5000 listed companies, where sorting of companies requires analysis of numerous financial parameters and ratios, is an example of a database from which single sub set or group of companies need to be found out for investing. With growing popularity of artificial intelligence in stock trading, it has been applied heavily into stock market for picking stocks. We apply K-means clustering algorithm, which uses unsupervised learning for clustering a dataset, to the listed Indian companies in an attempt to isolate businesses showing exceptionally high growth. The results show that, it is possible to achieve such grouping in an efficient and timely manner by K-means clustering. Testing of results is done by comparing absolute price returns and risk-adjusted returns of the obtained cluster and that of the popular and relevant market indices and top performing funds. The returns obtained from the cluster obtained using K-means clustering algorithm are found to outperform the indices and mutual funds by a comprehensive margin. These results emphasizes the role played by growth in stock performance in emerging markets along with subsequent conclusions arrived at therein, have important implications in the field of stock selection and portfolio management.
{"title":"Enhancing portfolio returns by identifying high growth companies in Indian stock market using artificial intelligence","authors":"Anup Rokade, Akshay Malhotra, Ankita S. Wanchoo","doi":"10.1109/RTEICT.2016.7807824","DOIUrl":"https://doi.org/10.1109/RTEICT.2016.7807824","url":null,"abstract":"The Indian Stock market, with over 5000 listed companies, where sorting of companies requires analysis of numerous financial parameters and ratios, is an example of a database from which single sub set or group of companies need to be found out for investing. With growing popularity of artificial intelligence in stock trading, it has been applied heavily into stock market for picking stocks. We apply K-means clustering algorithm, which uses unsupervised learning for clustering a dataset, to the listed Indian companies in an attempt to isolate businesses showing exceptionally high growth. The results show that, it is possible to achieve such grouping in an efficient and timely manner by K-means clustering. Testing of results is done by comparing absolute price returns and risk-adjusted returns of the obtained cluster and that of the popular and relevant market indices and top performing funds. The returns obtained from the cluster obtained using K-means clustering algorithm are found to outperform the indices and mutual funds by a comprehensive margin. These results emphasizes the role played by growth in stock performance in emerging markets along with subsequent conclusions arrived at therein, have important implications in the field of stock selection and portfolio management.","PeriodicalId":6527,"journal":{"name":"2016 IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT)","volume":"59 3 1","pages":"262-266"},"PeriodicalIF":0.0,"publicationDate":"2016-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78270426","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 : 2016-05-20DOI: 10.1109/RTEICT.2016.7807773
Rikita Chokshi, Dippal Israni, Nishidh Chavda
Distortion in images is a biggest challenge now-a-days. This affects in many areas ranging from photography to medical imaging, astronomy, remote sensing and microscopy. Images get obscured due to many reasons like vibration due to hand movement as well as launch of vehicle (Satellite), Noise in image, Adverse Image/Environment condition, and Quick movement of objects. A technique is required which can solve the above mentioned problems and make possible steps to keep image obscureness as minimum as possible. Out of several steps of restoration, blur detection is a primary step required for any blind image restoration. In this paper comparison of various techniques are proposed which finds out type of blur from the corrupted/degraded image using features like Moment Invariants, Histogram of Oriented Gradients, ZernikeMoment. This paper also describes comparison of different linear and nonlinear restoration techniques. The analysis and comparison was yielded out based on types of blur, estimation of blur, Structural Similarity Index (SSIM), Peak Signal-to-Noise Ratio (PSNR).
{"title":"An efficient deconvolution technique by identification and estimation of blur","authors":"Rikita Chokshi, Dippal Israni, Nishidh Chavda","doi":"10.1109/RTEICT.2016.7807773","DOIUrl":"https://doi.org/10.1109/RTEICT.2016.7807773","url":null,"abstract":"Distortion in images is a biggest challenge now-a-days. This affects in many areas ranging from photography to medical imaging, astronomy, remote sensing and microscopy. Images get obscured due to many reasons like vibration due to hand movement as well as launch of vehicle (Satellite), Noise in image, Adverse Image/Environment condition, and Quick movement of objects. A technique is required which can solve the above mentioned problems and make possible steps to keep image obscureness as minimum as possible. Out of several steps of restoration, blur detection is a primary step required for any blind image restoration. In this paper comparison of various techniques are proposed which finds out type of blur from the corrupted/degraded image using features like Moment Invariants, Histogram of Oriented Gradients, ZernikeMoment. This paper also describes comparison of different linear and nonlinear restoration techniques. The analysis and comparison was yielded out based on types of blur, estimation of blur, Structural Similarity Index (SSIM), Peak Signal-to-Noise Ratio (PSNR).","PeriodicalId":6527,"journal":{"name":"2016 IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT)","volume":"9 1","pages":"17-23"},"PeriodicalIF":0.0,"publicationDate":"2016-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78642619","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：