Pub Date : 2018-03-01DOI: 10.1109/INFOCT.2018.8356856
Wei-bin Chen, Ben Niu, Hongbin Gu, Xin Zhang
This paper presents a novel kernel-based method for small Target Tracking with multi-feature fusion against Varying Illumination. Firstly, the conventional tracker based on color histogram is unreliable or even failed under varying illumination. Therefore, a new fuzzy color histogram creation is proposed based on the HSV color space and utilizes the local background information around tracking target to dynamically correct its fuzzy color histogram model and eliminates the sensitive of conventional color histogram to illumination change and noise. Secondly, there is still not an effective method to cope with object occlusion, angle variation, scale change etc. The tracking algorithm utilizes feature points extracted by improved SIFT as the reference points of Mean-Shift and calculates the target area center, which combines the two methods together seamlessly. Lastly, the whole tracking algorithm utilizes fuzzy color histogram model and combination of improved SIFT as the reference points of Mean-Shift for small target tracking. Experiment results show that the proposed algorithm can keep tracking object of varying scales and various illumination even when the surrounding background being similar to the object's appearance.
{"title":"A novel strategy for kernel-based small target tracking against varying illumination with multiple features fusion","authors":"Wei-bin Chen, Ben Niu, Hongbin Gu, Xin Zhang","doi":"10.1109/INFOCT.2018.8356856","DOIUrl":"https://doi.org/10.1109/INFOCT.2018.8356856","url":null,"abstract":"This paper presents a novel kernel-based method for small Target Tracking with multi-feature fusion against Varying Illumination. Firstly, the conventional tracker based on color histogram is unreliable or even failed under varying illumination. Therefore, a new fuzzy color histogram creation is proposed based on the HSV color space and utilizes the local background information around tracking target to dynamically correct its fuzzy color histogram model and eliminates the sensitive of conventional color histogram to illumination change and noise. Secondly, there is still not an effective method to cope with object occlusion, angle variation, scale change etc. The tracking algorithm utilizes feature points extracted by improved SIFT as the reference points of Mean-Shift and calculates the target area center, which combines the two methods together seamlessly. Lastly, the whole tracking algorithm utilizes fuzzy color histogram model and combination of improved SIFT as the reference points of Mean-Shift for small target tracking. Experiment results show that the proposed algorithm can keep tracking object of varying scales and various illumination even when the surrounding background being similar to the object's appearance.","PeriodicalId":376443,"journal":{"name":"2018 International Conference on Information and Computer Technologies (ICICT)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132200687","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 : 2018-03-01DOI: 10.1109/INFOCT.2018.8356858
Azmi Alazzam
There has been always a high demand for process optimization in industry. Different mathematical and heuristic approaches have been used and developed for process optimization. Generally, the purpose of the process optimization is to maximize the throughput of the process under study, and also to minimize the cost associated with this process. In most cases, researchers study processes and analyze them in order to find the parameters that will optimize the system. In fact, some systems are harder to analyze and optimize compared to other systems. In this paper, a new optimization algorithm is introduced. The proposed algorithm is named Best Uniform Algorithm (BUA). This algorithm is a meta-heuristic approach that generates a set of random solutions randomly from the entire search space, and the best solution is then used to generate other solutions. The BUA algorithm is designed to optimize the process in the sequential type of pickup and place machines. There are multiple stationary feeders in the sequential machine, and each feeder is assumed to store the same type of components. The head usually moves to pick up components from feeders and place them into different places on the PCB based on the design. The head starts from a fixed place, and then it moves to pick up one component from one of the feeders and place it on the PCB. Afterwards, the head moves to pick up the next component from the feeder that stores it. The process will continue until all the components are placed. In order to optimize this process, that distance that the head must travel until all the components are placed need to be optimized. The optimal parameters that optimize the total distance will be also found; these parameters will include the position of the feeders and the sequence of the components that need to be placed. Throughout this paper, the new optimization algorithm is introduced and the methodology is discussed. Next, this algorithm is applied to the sequential pickup and place machine in order to optimize the distance that the placement head travels. Finally, the results of applying this algorithm to the pick and place optimization problem are elaborated and compared to the well-known Genetic Algorithm (GA).
{"title":"Using BUA algorithm to solve a sequential pick and place problem","authors":"Azmi Alazzam","doi":"10.1109/INFOCT.2018.8356858","DOIUrl":"https://doi.org/10.1109/INFOCT.2018.8356858","url":null,"abstract":"There has been always a high demand for process optimization in industry. Different mathematical and heuristic approaches have been used and developed for process optimization. Generally, the purpose of the process optimization is to maximize the throughput of the process under study, and also to minimize the cost associated with this process. In most cases, researchers study processes and analyze them in order to find the parameters that will optimize the system. In fact, some systems are harder to analyze and optimize compared to other systems. In this paper, a new optimization algorithm is introduced. The proposed algorithm is named Best Uniform Algorithm (BUA). This algorithm is a meta-heuristic approach that generates a set of random solutions randomly from the entire search space, and the best solution is then used to generate other solutions. The BUA algorithm is designed to optimize the process in the sequential type of pickup and place machines. There are multiple stationary feeders in the sequential machine, and each feeder is assumed to store the same type of components. The head usually moves to pick up components from feeders and place them into different places on the PCB based on the design. The head starts from a fixed place, and then it moves to pick up one component from one of the feeders and place it on the PCB. Afterwards, the head moves to pick up the next component from the feeder that stores it. The process will continue until all the components are placed. In order to optimize this process, that distance that the head must travel until all the components are placed need to be optimized. The optimal parameters that optimize the total distance will be also found; these parameters will include the position of the feeders and the sequence of the components that need to be placed. Throughout this paper, the new optimization algorithm is introduced and the methodology is discussed. Next, this algorithm is applied to the sequential pickup and place machine in order to optimize the distance that the placement head travels. Finally, the results of applying this algorithm to the pick and place optimization problem are elaborated and compared to the well-known Genetic Algorithm (GA).","PeriodicalId":376443,"journal":{"name":"2018 International Conference on Information and Computer Technologies (ICICT)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114571451","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 : 2018-03-01DOI: 10.1109/INFOCT.2018.8356863
R. Hashemian
Rapid growth in electronic technology makes it difficult to keep track of the device modeling for linear analysis and design. This is particularly hard for MOS transistors. The problem is even more crucial for training engineering students, where they need to be able to hand calculate, put it into circuit format and estimate the design parameters. With transistors model parameters that may run into pages for today's nanotechnology, manually identify the model components and come up with linear equivalent circuits is not easy although it may be a routine data crunching procedure for a modern circuit simulator. One way to get around the problem is to follow two paths. First simulate, or experiment with, the circuit in its original (nonlinear) form, and keep it as the “model”. In the second step, adopt a proper linearizing scheme for the transistors manually, and then modify the model parameters based on the responses received from the model circuit. In our proposed method we do exactly this by using FNPs to modify the parameters in the linearized circuit until the response from the model and that of the linearized circuit fit together.
{"title":"Nullors in finding/modifying transistor parameters, blindly","authors":"R. Hashemian","doi":"10.1109/INFOCT.2018.8356863","DOIUrl":"https://doi.org/10.1109/INFOCT.2018.8356863","url":null,"abstract":"Rapid growth in electronic technology makes it difficult to keep track of the device modeling for linear analysis and design. This is particularly hard for MOS transistors. The problem is even more crucial for training engineering students, where they need to be able to hand calculate, put it into circuit format and estimate the design parameters. With transistors model parameters that may run into pages for today's nanotechnology, manually identify the model components and come up with linear equivalent circuits is not easy although it may be a routine data crunching procedure for a modern circuit simulator. One way to get around the problem is to follow two paths. First simulate, or experiment with, the circuit in its original (nonlinear) form, and keep it as the “model”. In the second step, adopt a proper linearizing scheme for the transistors manually, and then modify the model parameters based on the responses received from the model circuit. In our proposed method we do exactly this by using FNPs to modify the parameters in the linearized circuit until the response from the model and that of the linearized circuit fit together.","PeriodicalId":376443,"journal":{"name":"2018 International Conference on Information and Computer Technologies (ICICT)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125747091","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 : 2018-03-01DOI: 10.1109/INFOCT.2018.8356847
Nooh Bany Muhammad
As technology advances, security breaches, hacks and intentional false positives increase. Therefore, security is a hot and imperative topic that every individual and organization must keep in mind as a high priority subject. This paper will discuss cell phone security, verification and authentication techniques, limitations of technologies of curbing the risks of these devices on us and the advances made in the provision of suitable solutions for these risks is the primary focus herein.
{"title":"A study on cell phone security: Authentication techniques","authors":"Nooh Bany Muhammad","doi":"10.1109/INFOCT.2018.8356847","DOIUrl":"https://doi.org/10.1109/INFOCT.2018.8356847","url":null,"abstract":"As technology advances, security breaches, hacks and intentional false positives increase. Therefore, security is a hot and imperative topic that every individual and organization must keep in mind as a high priority subject. This paper will discuss cell phone security, verification and authentication techniques, limitations of technologies of curbing the risks of these devices on us and the advances made in the provision of suitable solutions for these risks is the primary focus herein.","PeriodicalId":376443,"journal":{"name":"2018 International Conference on Information and Computer Technologies (ICICT)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114954179","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 : 2018-03-01DOI: 10.1109/INFOCT.2018.8356849
Nekiesha Edward, Jeffrey Elcock
Task scheduling in heterogeneous multiprocessing environments, continues to be one of the most important and also very challenging problems. Scheduling in such environments is NP-hard, so it is imperative that we tackle this critical issue with a desire of producing effective and efficient solutions. For several types of applications, the task scheduling problem is crucial, and throughout the literature, there are a number of algorithms to be found using several different techniques and approaches. Ant Colony Optimization (ACO) is one such technique to be found. This popular and robust optimization technique is based on the behavior of ants seeking to find the shortest path between their nest and food sources. In this paper, we propose an ACO-based algorithm, called rank-ACO, as an efficient solution to the task scheduling problem. Our algorithm allows for an initial random scheduled selection; utilizes pheromone and a priority-based heuristic, known as the upward rank value, as well as an insertion-based policy, along with a pheromone aging mechanism to guide the ants to good quality solutions. To evaluate the performance of our algorithm, we compared our algorithm with the ACS algorithm and the ACO-TMS algorithm using randomly generated directed acyclic graphs (DAGs). The simulation results indicated that our algorithm experienced comparable or even better performance, than the selected algorithms.
{"title":"An efficient task scheduling algorithm for heterogeneous multiprocessing environments","authors":"Nekiesha Edward, Jeffrey Elcock","doi":"10.1109/INFOCT.2018.8356849","DOIUrl":"https://doi.org/10.1109/INFOCT.2018.8356849","url":null,"abstract":"Task scheduling in heterogeneous multiprocessing environments, continues to be one of the most important and also very challenging problems. Scheduling in such environments is NP-hard, so it is imperative that we tackle this critical issue with a desire of producing effective and efficient solutions. For several types of applications, the task scheduling problem is crucial, and throughout the literature, there are a number of algorithms to be found using several different techniques and approaches. Ant Colony Optimization (ACO) is one such technique to be found. This popular and robust optimization technique is based on the behavior of ants seeking to find the shortest path between their nest and food sources. In this paper, we propose an ACO-based algorithm, called rank-ACO, as an efficient solution to the task scheduling problem. Our algorithm allows for an initial random scheduled selection; utilizes pheromone and a priority-based heuristic, known as the upward rank value, as well as an insertion-based policy, along with a pheromone aging mechanism to guide the ants to good quality solutions. To evaluate the performance of our algorithm, we compared our algorithm with the ACS algorithm and the ACO-TMS algorithm using randomly generated directed acyclic graphs (DAGs). The simulation results indicated that our algorithm experienced comparable or even better performance, than the selected algorithms.","PeriodicalId":376443,"journal":{"name":"2018 International Conference on Information and Computer Technologies (ICICT)","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127121377","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 : 2018-03-01DOI: 10.1109/INFOCT.2018.8356862
Laasya Vadlamani, K. Agyemang, Benedito J. B. Fonseca
This paper reports on the initial experiments performed using a low-cost, commercial grade, infrared camera to remotely monitor athlete runners and detect temperature variations during the exercise. The initial findings highlight the challenges in using commercial grade instruments to extract scientific measurements from subjects. The paper also describes the techniques used to circumvent some of these challenges and enable the extraction of useful information from the collected images. Results obtained are compared against existing results using scientific grade infrared cameras.
{"title":"Experiments with a low-cost infrared camera to remotely monitor athlete runners","authors":"Laasya Vadlamani, K. Agyemang, Benedito J. B. Fonseca","doi":"10.1109/INFOCT.2018.8356862","DOIUrl":"https://doi.org/10.1109/INFOCT.2018.8356862","url":null,"abstract":"This paper reports on the initial experiments performed using a low-cost, commercial grade, infrared camera to remotely monitor athlete runners and detect temperature variations during the exercise. The initial findings highlight the challenges in using commercial grade instruments to extract scientific measurements from subjects. The paper also describes the techniques used to circumvent some of these challenges and enable the extraction of useful information from the collected images. Results obtained are compared against existing results using scientific grade infrared cameras.","PeriodicalId":376443,"journal":{"name":"2018 International Conference on Information and Computer Technologies (ICICT)","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117127686","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 : 2018-03-01DOI: 10.1109/INFOCT.2018.8356855
Khaled Alrawashdeh, C. Purdy
Real-time scenarios of deep learning algorithms are challenged by two less frequently addressed issues. The first is data inefficiency i.e., the model requires several epochs of trial and error to converge which makes it impractical to be applied to real-time applications. The Second is the high precision computation load of the deep learning algorithms to achieve high accuracy during training and inference. In this paper, we address these two issues and apply our model to the task of online anomaly detection using FPGA. To address the first issue, we propose a compressed training model for the contrastive divergence algorithm (CD) in the Deep Belief Network (DBN). The goal is to dynamically adjust the training vector according to the feedback from the free energy and the reconstruction error, which allows for better generalization. To address the second issue, we propose a Hybrid-Stochastic-Dynamic-Fixed-Point (HSDFP) method, which provides training environment with high reduction in calculation, area, and power in FPGA. Our framework enables the DBN structure to take actions and detect attacks online. Thus, the network can collect efficient number of training samples and avoid overfitting. We show that (1) our proposed method converges faster than the state-of-the-art deep learning methods, (2) FPGA implementation achieves accelerated inference speed of 0.008ms and a high power efficiency of 37 G-ops/s/W compared to CPU, GPU, and 16-bit fixed-point arithmetic (3) FPGA also achieves minimal degradation in accuracy of 95%, 95.4%, and 97.9% on the benchmark datasets: MNIST, NSL-KDD, and Kyoto datasets.
{"title":"Fast hardware assisted online learning using unsupervised deep learning structure for anomaly detection","authors":"Khaled Alrawashdeh, C. Purdy","doi":"10.1109/INFOCT.2018.8356855","DOIUrl":"https://doi.org/10.1109/INFOCT.2018.8356855","url":null,"abstract":"Real-time scenarios of deep learning algorithms are challenged by two less frequently addressed issues. The first is data inefficiency i.e., the model requires several epochs of trial and error to converge which makes it impractical to be applied to real-time applications. The Second is the high precision computation load of the deep learning algorithms to achieve high accuracy during training and inference. In this paper, we address these two issues and apply our model to the task of online anomaly detection using FPGA. To address the first issue, we propose a compressed training model for the contrastive divergence algorithm (CD) in the Deep Belief Network (DBN). The goal is to dynamically adjust the training vector according to the feedback from the free energy and the reconstruction error, which allows for better generalization. To address the second issue, we propose a Hybrid-Stochastic-Dynamic-Fixed-Point (HSDFP) method, which provides training environment with high reduction in calculation, area, and power in FPGA. Our framework enables the DBN structure to take actions and detect attacks online. Thus, the network can collect efficient number of training samples and avoid overfitting. We show that (1) our proposed method converges faster than the state-of-the-art deep learning methods, (2) FPGA implementation achieves accelerated inference speed of 0.008ms and a high power efficiency of 37 G-ops/s/W compared to CPU, GPU, and 16-bit fixed-point arithmetic (3) FPGA also achieves minimal degradation in accuracy of 95%, 95.4%, and 97.9% on the benchmark datasets: MNIST, NSL-KDD, and Kyoto datasets.","PeriodicalId":376443,"journal":{"name":"2018 International Conference on Information and Computer Technologies (ICICT)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117006283","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 : 2018-03-01DOI: 10.1109/INFOCT.2018.8356850
S. Dinç, R. Aygun, F. Fahimi
An accurate vision based trajectory tracking application requires accurate pose estimation of the camera (or the carrier vehicle) during the motion. Aside from the algorithmic and computational difficulties of pose estimation, challenges arise from the environmental changes or instantaneous imaging errors. Usually, in such cases, the true pose cannot be estimated. This causes the determination of incorrect motion directives to the robotic vehicle. To compensate this miscalculation a Pose Recovery algorithm is necessary to lead mobile vehicle to desired trajectory. This paper presents a new pose recovery algorithm that employs previously generated virtual poses of a reference object to correct miscalculated pose. In our experiments, we use the Mirage pose estimation to demonstrate the performance of the proposed recovery algorithm. We evaluated our algorithm on 3 different experiments using one reference object. According to the results, our pose recovery algorithm can be effective for majority of pose failure cases with very low positional errors.
{"title":"An efficient pose recovery algorithm for vision based trajectory tracking applications","authors":"S. Dinç, R. Aygun, F. Fahimi","doi":"10.1109/INFOCT.2018.8356850","DOIUrl":"https://doi.org/10.1109/INFOCT.2018.8356850","url":null,"abstract":"An accurate vision based trajectory tracking application requires accurate pose estimation of the camera (or the carrier vehicle) during the motion. Aside from the algorithmic and computational difficulties of pose estimation, challenges arise from the environmental changes or instantaneous imaging errors. Usually, in such cases, the true pose cannot be estimated. This causes the determination of incorrect motion directives to the robotic vehicle. To compensate this miscalculation a Pose Recovery algorithm is necessary to lead mobile vehicle to desired trajectory. This paper presents a new pose recovery algorithm that employs previously generated virtual poses of a reference object to correct miscalculated pose. In our experiments, we use the Mirage pose estimation to demonstrate the performance of the proposed recovery algorithm. We evaluated our algorithm on 3 different experiments using one reference object. According to the results, our pose recovery algorithm can be effective for majority of pose failure cases with very low positional errors.","PeriodicalId":376443,"journal":{"name":"2018 International Conference on Information and Computer Technologies (ICICT)","volume":"312 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122699283","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 : 2018-03-01DOI: 10.1109/INFOCT.2018.8356860
Da-Kai Chen
The accuracy limitation of analog-to-digital converters (ADCs) is mainly set by component mismatches and amplifier offsets. A new two-stage architecture ADC is presented and some of the issues of digital calibrate technique are discussed. The design of ADC uses a modified pipeline technique based on Successive-Approximation ADCs. The proposed architecture does not need an auxiliary DAC for reconstruction at each stage. Rather than achieving high precision by adjustment of analogy component values, we present a native digital technique to calibrate pipeline ADC, which calibrates linearity and gain with the same hardware. In digital calibration operations, the overlap condition will occur in edge of the full-scale voltage inevitably. Here we design a simple circuit which uses one adder to implement the over-or underrange protection. In order to get bipolar mode, the converter using dual-voltage suppliers (±5V) is required. The circuit was designed in the P-well HVCMOS process. The experiments results are presented to show the effectiveness of the design method and the precision of the analog-to-digital conversion is up to16-bit.
{"title":"A 16b 5MSPS two-stage pipeline ADC with self-calibrated technology","authors":"Da-Kai Chen","doi":"10.1109/INFOCT.2018.8356860","DOIUrl":"https://doi.org/10.1109/INFOCT.2018.8356860","url":null,"abstract":"The accuracy limitation of analog-to-digital converters (ADCs) is mainly set by component mismatches and amplifier offsets. A new two-stage architecture ADC is presented and some of the issues of digital calibrate technique are discussed. The design of ADC uses a modified pipeline technique based on Successive-Approximation ADCs. The proposed architecture does not need an auxiliary DAC for reconstruction at each stage. Rather than achieving high precision by adjustment of analogy component values, we present a native digital technique to calibrate pipeline ADC, which calibrates linearity and gain with the same hardware. In digital calibration operations, the overlap condition will occur in edge of the full-scale voltage inevitably. Here we design a simple circuit which uses one adder to implement the over-or underrange protection. In order to get bipolar mode, the converter using dual-voltage suppliers (±5V) is required. The circuit was designed in the P-well HVCMOS process. The experiments results are presented to show the effectiveness of the design method and the precision of the analog-to-digital conversion is up to16-bit.","PeriodicalId":376443,"journal":{"name":"2018 International Conference on Information and Computer Technologies (ICICT)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130436530","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 : 2018-03-01DOI: 10.1109/INFOCT.2018.8356854
Thein Thein, Kalyar Myo San
Lip reading system is supportive technology to human being especially for hearing impaired, or elderly people. Lip reading is a process where visual information is extracted by watching lip movements of the speaker with or without sound. So, reliable lip movements are required to extract visual information. To our knowledge, this is the first work for lip movement recognition for Myanmar consonants. So, the major challenge is to recognize lip movements because of many possible lip motions and lip shapes. The accuracy and reliability of speech recognition systems can be improved by using visual information from the movements of the lips, and the need for lip-reading systems continues to grow for every language. Therefore, this paper presents Myanmar consonant recognition based on lip movements towards lip reading by using CIELa∗b∗ color transformation, Moore Neighborhood Tracing Algorithm and linear SVM classifier. The purpose of this study was to develop a visual training technique to accurately identify the characteristics of the lips movement for hearing impairment.
{"title":"Lip localization technique towards an automatic lip reading approach for Myanmar consonants recognition","authors":"Thein Thein, Kalyar Myo San","doi":"10.1109/INFOCT.2018.8356854","DOIUrl":"https://doi.org/10.1109/INFOCT.2018.8356854","url":null,"abstract":"Lip reading system is supportive technology to human being especially for hearing impaired, or elderly people. Lip reading is a process where visual information is extracted by watching lip movements of the speaker with or without sound. So, reliable lip movements are required to extract visual information. To our knowledge, this is the first work for lip movement recognition for Myanmar consonants. So, the major challenge is to recognize lip movements because of many possible lip motions and lip shapes. The accuracy and reliability of speech recognition systems can be improved by using visual information from the movements of the lips, and the need for lip-reading systems continues to grow for every language. Therefore, this paper presents Myanmar consonant recognition based on lip movements towards lip reading by using CIELa∗b∗ color transformation, Moore Neighborhood Tracing Algorithm and linear SVM classifier. The purpose of this study was to develop a visual training technique to accurately identify the characteristics of the lips movement for hearing impairment.","PeriodicalId":376443,"journal":{"name":"2018 International Conference on Information and Computer Technologies (ICICT)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128896488","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：