Pub Date : 2021-08-26DOI: 10.1109/SPIN52536.2021.9565954
Bibaswan Bose, V. K. Tayal, Bedatri Moulik
The increasing use of non-renewable sources has led to an increase in global warming and carbon footprint. In this scenario EV, BESS and renewable sources such as wind turbine generator and SPVA have proved to be highly beneficial in creating a sustainable future. This work makes one such attempt and creates an islanded microgrid with renewable energy sources, BESS and V2G. A novel rule-based energy management unit has been developed in this work to help supply of energy generated as per the load demand. The results indicate that the developed system can successfully meet load requirements and this will help to reduce the carbon emissions to a bare minimum. The complete system has been primarily simulated on MATLAB/ Simulink and then verified using Hardware-in-loop configuration on a Hardware Test Bench.
{"title":"Design and Control of Islanded Microgrid Architecture with Vehicle to Grid Integration","authors":"Bibaswan Bose, V. K. Tayal, Bedatri Moulik","doi":"10.1109/SPIN52536.2021.9565954","DOIUrl":"https://doi.org/10.1109/SPIN52536.2021.9565954","url":null,"abstract":"The increasing use of non-renewable sources has led to an increase in global warming and carbon footprint. In this scenario EV, BESS and renewable sources such as wind turbine generator and SPVA have proved to be highly beneficial in creating a sustainable future. This work makes one such attempt and creates an islanded microgrid with renewable energy sources, BESS and V2G. A novel rule-based energy management unit has been developed in this work to help supply of energy generated as per the load demand. The results indicate that the developed system can successfully meet load requirements and this will help to reduce the carbon emissions to a bare minimum. The complete system has been primarily simulated on MATLAB/ Simulink and then verified using Hardware-in-loop configuration on a Hardware Test Bench.","PeriodicalId":343177,"journal":{"name":"2021 8th International Conference on Signal Processing and Integrated Networks (SPIN)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134305845","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 : 2021-08-26DOI: 10.1109/SPIN52536.2021.9565996
Rajdeep Chatterjee, Soham Roy, SK Hafizul Islam, Debabrata Samanta
Artificial intelligence systems have permeated into all spheres of our life-impacting everything from our food habits to our sleep patterns. One untouched area where such intelligent systems are still in their infancy is sports. There has not been enough indulgence of AI techniques in sports, and most of the works are carried on manually by coaching staff and human appointees. We believe that intelligent systems can make coaching staff’s work easier and produce findings that the human eye can often overlook. Here, we have proposed an intelligent system to analyze the beautiful game of tennis. With the use of computer vision architecture Detectron2 and activity-based pose estimation and subsequent classification, it can identify an action from a tennis shot (activity). It can produce a performance score for the player based on pose and movement like forehand and backhand. It can also be used to understand and evaluate the strengths and weaknesses of the player. The proposed approach provides a piece of valuable information for a player’s performance and activity detection to be used for better coaching. The study achieves a classification accuracy of 98.60% and outperforms other SOTA CNN models.
{"title":"An AI Approach to Pose-based Sports Activity Classification","authors":"Rajdeep Chatterjee, Soham Roy, SK Hafizul Islam, Debabrata Samanta","doi":"10.1109/SPIN52536.2021.9565996","DOIUrl":"https://doi.org/10.1109/SPIN52536.2021.9565996","url":null,"abstract":"Artificial intelligence systems have permeated into all spheres of our life-impacting everything from our food habits to our sleep patterns. One untouched area where such intelligent systems are still in their infancy is sports. There has not been enough indulgence of AI techniques in sports, and most of the works are carried on manually by coaching staff and human appointees. We believe that intelligent systems can make coaching staff’s work easier and produce findings that the human eye can often overlook. Here, we have proposed an intelligent system to analyze the beautiful game of tennis. With the use of computer vision architecture Detectron2 and activity-based pose estimation and subsequent classification, it can identify an action from a tennis shot (activity). It can produce a performance score for the player based on pose and movement like forehand and backhand. It can also be used to understand and evaluate the strengths and weaknesses of the player. The proposed approach provides a piece of valuable information for a player’s performance and activity detection to be used for better coaching. The study achieves a classification accuracy of 98.60% and outperforms other SOTA CNN models.","PeriodicalId":343177,"journal":{"name":"2021 8th International Conference on Signal Processing and Integrated Networks (SPIN)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131829110","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 : 2021-08-26DOI: 10.1109/SPIN52536.2021.9566112
Manan Shah, Pankaj Kumar
Breakthroughs in the convolution neural network(CNN) have resolved and improved many challenges of pattern recognition in natural images. With the increased use of proximal sensing and low-cost cameras, monitoring and automation systems have gained popularity in the agriculture fields. Detection, segmentation, clustering, and counting are some fundamental problems associated with it. Here we are working on the detection of wine grapes, a crop with a variety of shapes, colors, sizes, and structures. Object detection is a challenging task especially when we are working on natural images. It is an even more difficult task when we are working on blurred images. Blur arises when images are taken via handheld camera, moving object in the automation system, or low rate video frame.Here we are trying to solve the motion blur problem in grape detection using three existing image deblurring algorithms. Performance of the deblurring algorithm is generally measured by peak-signal to noise ratio(PSNR) and structure similarity index(SSIM), but in addition to it, we have also considered blind/referenceless image spatial quality evaluator(BRISQUE). In this paper, we have comparatively analyzed: Scale recurrent network(SRN) for deep image deblurring, Multiscale convolution neural network for dynamic scale deblurring(Deep deblur), and DeblurGANv2: Deblurring(orders of magnitude) faster and better. Grape detection has experimented with yolov5x. Raw images from the standard dataset(GoPro and WGSID) were corrupted with various kinds of motion blurs. From the obtained result we can conclude that image deblurring significantly improves the performance of grape detection on the corrupted motion blur dataset.
{"title":"Improved handling of motion blur for grape detection after deblurring","authors":"Manan Shah, Pankaj Kumar","doi":"10.1109/SPIN52536.2021.9566112","DOIUrl":"https://doi.org/10.1109/SPIN52536.2021.9566112","url":null,"abstract":"Breakthroughs in the convolution neural network(CNN) have resolved and improved many challenges of pattern recognition in natural images. With the increased use of proximal sensing and low-cost cameras, monitoring and automation systems have gained popularity in the agriculture fields. Detection, segmentation, clustering, and counting are some fundamental problems associated with it. Here we are working on the detection of wine grapes, a crop with a variety of shapes, colors, sizes, and structures. Object detection is a challenging task especially when we are working on natural images. It is an even more difficult task when we are working on blurred images. Blur arises when images are taken via handheld camera, moving object in the automation system, or low rate video frame.Here we are trying to solve the motion blur problem in grape detection using three existing image deblurring algorithms. Performance of the deblurring algorithm is generally measured by peak-signal to noise ratio(PSNR) and structure similarity index(SSIM), but in addition to it, we have also considered blind/referenceless image spatial quality evaluator(BRISQUE). In this paper, we have comparatively analyzed: Scale recurrent network(SRN) for deep image deblurring, Multiscale convolution neural network for dynamic scale deblurring(Deep deblur), and DeblurGANv2: Deblurring(orders of magnitude) faster and better. Grape detection has experimented with yolov5x. Raw images from the standard dataset(GoPro and WGSID) were corrupted with various kinds of motion blurs. From the obtained result we can conclude that image deblurring significantly improves the performance of grape detection on the corrupted motion blur dataset.","PeriodicalId":343177,"journal":{"name":"2021 8th International Conference on Signal Processing and Integrated Networks (SPIN)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125221142","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 : 2021-08-26DOI: 10.1109/SPIN52536.2021.9566092
Atman Raj Sahu, Bedatri Moulik, Bibaswan Bose
In hybrid electric vehicles parameters, the battery management systems play an important part in state estimation techniques which has to be reliable and precise. A HEV when running checks several parameters constantly to make the control strategies on the upcoming driving conditions. A battery management system consists of various components such as sensors and actuators that helps in the safety of battery, improving the range for driving and helps in cost minimization. A battery management system installed in a HEV assures the vehicle efficiency by estimation of these several parameters and processes the HEV according to it. These parameters are a vital information to the hybrid electric vehicles as this information decoded from the sensors or algorithm provides the update on the vehicle’s component running data. Data such as State of charge, cell’s temperature, state of health of the cell, state of power, state of life, etc. are all taken into the consideration for the running of a hybrid electric vehicles. These parameters are the vital information to the hybrid electric vehicles as this information decoded from the sensors or algorithm provides the update on the vehicle’s component running data. Data such as State of charge, cell’s temperature, state of health of the cell, state of power, state of life, etc. are all taken into the consideration for the operation of a hybrid electric vehicles. In this framework a battery model is planned which is put under direct measurement techniques to successfully estimate State of Charge. These two methods used are OCV method and coulomb counting method technique. A basic thermal model is also projected in this paper to measure temperature of the battery, to verify all these operations, a simulation results has been briefed at last.
{"title":"Online Approximation of SOC and temperature of a electric vehicle by combined OCV-CC method","authors":"Atman Raj Sahu, Bedatri Moulik, Bibaswan Bose","doi":"10.1109/SPIN52536.2021.9566092","DOIUrl":"https://doi.org/10.1109/SPIN52536.2021.9566092","url":null,"abstract":"In hybrid electric vehicles parameters, the battery management systems play an important part in state estimation techniques which has to be reliable and precise. A HEV when running checks several parameters constantly to make the control strategies on the upcoming driving conditions. A battery management system consists of various components such as sensors and actuators that helps in the safety of battery, improving the range for driving and helps in cost minimization. A battery management system installed in a HEV assures the vehicle efficiency by estimation of these several parameters and processes the HEV according to it. These parameters are a vital information to the hybrid electric vehicles as this information decoded from the sensors or algorithm provides the update on the vehicle’s component running data. Data such as State of charge, cell’s temperature, state of health of the cell, state of power, state of life, etc. are all taken into the consideration for the running of a hybrid electric vehicles. These parameters are the vital information to the hybrid electric vehicles as this information decoded from the sensors or algorithm provides the update on the vehicle’s component running data. Data such as State of charge, cell’s temperature, state of health of the cell, state of power, state of life, etc. are all taken into the consideration for the operation of a hybrid electric vehicles. In this framework a battery model is planned which is put under direct measurement techniques to successfully estimate State of Charge. These two methods used are OCV method and coulomb counting method technique. A basic thermal model is also projected in this paper to measure temperature of the battery, to verify all these operations, a simulation results has been briefed at last.","PeriodicalId":343177,"journal":{"name":"2021 8th International Conference on Signal Processing and Integrated Networks (SPIN)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130966489","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 : 2021-08-26DOI: 10.1109/SPIN52536.2021.9566134
Sushil Kumar, R. Yadav
The primary aim of this work is to analyze the potential of artificial intelligence in the field of automatic facial emotion recognition (AFER). Therefore, convolutional neural network is considered for classifying the 6 universal facial expressions. The feed-forward artificial neural network is also designed for comparative analysis. The designed techniques are implemented on extended Cohn-Kanade (CK+) database. Rigorous experimentation is carried out in order to analyze the efficacy of the suggested AFER scheme using different performance measures. It is revealed from the analysis that convolutional neural network-based classification proves to be superior in terms of accuracy, precision, recall and F1 score, as compared to the feedforward neural network-based classification scheme.
{"title":"Automatic Facial Emotion Recognition using Convolutional Neural Networks","authors":"Sushil Kumar, R. Yadav","doi":"10.1109/SPIN52536.2021.9566134","DOIUrl":"https://doi.org/10.1109/SPIN52536.2021.9566134","url":null,"abstract":"The primary aim of this work is to analyze the potential of artificial intelligence in the field of automatic facial emotion recognition (AFER). Therefore, convolutional neural network is considered for classifying the 6 universal facial expressions. The feed-forward artificial neural network is also designed for comparative analysis. The designed techniques are implemented on extended Cohn-Kanade (CK+) database. Rigorous experimentation is carried out in order to analyze the efficacy of the suggested AFER scheme using different performance measures. It is revealed from the analysis that convolutional neural network-based classification proves to be superior in terms of accuracy, precision, recall and F1 score, as compared to the feedforward neural network-based classification scheme.","PeriodicalId":343177,"journal":{"name":"2021 8th International Conference on Signal Processing and Integrated Networks (SPIN)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134465749","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 : 2021-08-26DOI: 10.1109/SPIN52536.2021.9566117
Shilpa Pavithran, Sanoj Viswasom, S. S, Asha J
In this paper, a comparative study of decision tree and random forest regression models for the design of a multiband inverted E and U-shaped compact antenna for future 5G applications is presented. The results obtained from these regression models are compared with the simulation results of openEMS software. Found that Random Forest (RF) regression model results are in good agreement with the openEMS results when compared to the decision tree regression model. The advantage of the proposed method lies in the fact that the final RF model can be used for the designing of this multiband antenna between 2GHz to 10GHz range of frequencies.
{"title":"Designing of a 5G Multiband Antenna Using Decision Tree and Random Forest Regression Models","authors":"Shilpa Pavithran, Sanoj Viswasom, S. S, Asha J","doi":"10.1109/SPIN52536.2021.9566117","DOIUrl":"https://doi.org/10.1109/SPIN52536.2021.9566117","url":null,"abstract":"In this paper, a comparative study of decision tree and random forest regression models for the design of a multiband inverted E and U-shaped compact antenna for future 5G applications is presented. The results obtained from these regression models are compared with the simulation results of openEMS software. Found that Random Forest (RF) regression model results are in good agreement with the openEMS results when compared to the decision tree regression model. The advantage of the proposed method lies in the fact that the final RF model can be used for the designing of this multiband antenna between 2GHz to 10GHz range of frequencies.","PeriodicalId":343177,"journal":{"name":"2021 8th International Conference on Signal Processing and Integrated Networks (SPIN)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131298242","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 : 2021-08-26DOI: 10.1109/SPIN52536.2021.9565993
S. Mohanty, Suvendu Rup
Foreground detection is a prime task in the field of computer vision for targeting the emerging applications like video surveillance, object tracking, action recognition, scene analysis. For moving object detection, it is always desirable to accurately extract the foreground under complex background conditions with less computational overhead. In this work, we propose a multifeature-based moving object detection scheme, where the feature vector for each pixel constitutes gray level intensity value and extended scale-invariant local ternary pattern (E-SILTP) over a local region. Further, to improve the detection accuracy with minimum computational cost, extended Canberra distance is employed for similarity distance between model and current pixel instead of popular Mahalanobis distance and Forstner distance. The experimental results are validated using some standard data sets and shows superior performance than that of the benchmark schemes.
{"title":"On the Development of Foreground Detection under Complex Background","authors":"S. Mohanty, Suvendu Rup","doi":"10.1109/SPIN52536.2021.9565993","DOIUrl":"https://doi.org/10.1109/SPIN52536.2021.9565993","url":null,"abstract":"Foreground detection is a prime task in the field of computer vision for targeting the emerging applications like video surveillance, object tracking, action recognition, scene analysis. For moving object detection, it is always desirable to accurately extract the foreground under complex background conditions with less computational overhead. In this work, we propose a multifeature-based moving object detection scheme, where the feature vector for each pixel constitutes gray level intensity value and extended scale-invariant local ternary pattern (E-SILTP) over a local region. Further, to improve the detection accuracy with minimum computational cost, extended Canberra distance is employed for similarity distance between model and current pixel instead of popular Mahalanobis distance and Forstner distance. The experimental results are validated using some standard data sets and shows superior performance than that of the benchmark schemes.","PeriodicalId":343177,"journal":{"name":"2021 8th International Conference on Signal Processing and Integrated Networks (SPIN)","volume":"139 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114655343","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 : 2021-08-26DOI: 10.1109/SPIN52536.2021.9565970
P. Ashish, M. Tripathy
In this paper, the graphene-based patch antenna with silicon dioxide substrate is designed and analyzed for WBAN applications operating at the THz band. The dimension of the antenna is 200pm x 180pm x 25.54 pm. Three resonant frequencies are obtained at 6.05 THz, 12.27 THz, and 15.74 THz, with the Sn value of -52.537 dB, -44.568 dB, and — 59.300 dB, respectively. The gains obtained at these frequencies are 8.807 dB, 12.75 dB, 14.17 dB, respectively. This compact and high gain terahertz antenna can be used for WBAN systems, notably for human health tracking applications.
本文设计并分析了基于二氧化硅衬底的石墨烯贴片天线在太赫兹波段的WBAN应用。天线的尺寸为200pm x 180pm x 25.54 pm。在6.05 THz、12.27 THz和15.74 THz处得到三个谐振频率，Sn值分别为-52.537 dB、-44.568 dB和- 59.300 dB。在这些频率下获得的增益分别为8.807 dB、12.75 dB、14.17 dB。这种紧凑的高增益太赫兹天线可用于WBAN系统，特别是用于人体健康跟踪应用。
{"title":"Design and Analysis of Graphene-Based Patch Antenna for WBAN at Terahertz Frequency","authors":"P. Ashish, M. Tripathy","doi":"10.1109/SPIN52536.2021.9565970","DOIUrl":"https://doi.org/10.1109/SPIN52536.2021.9565970","url":null,"abstract":"In this paper, the graphene-based patch antenna with silicon dioxide substrate is designed and analyzed for WBAN applications operating at the THz band. The dimension of the antenna is 200pm x 180pm x 25.54 pm. Three resonant frequencies are obtained at 6.05 THz, 12.27 THz, and 15.74 THz, with the Sn value of -52.537 dB, -44.568 dB, and — 59.300 dB, respectively. The gains obtained at these frequencies are 8.807 dB, 12.75 dB, 14.17 dB, respectively. This compact and high gain terahertz antenna can be used for WBAN systems, notably for human health tracking applications.","PeriodicalId":343177,"journal":{"name":"2021 8th International Conference on Signal Processing and Integrated Networks (SPIN)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123538478","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 : 2021-08-26DOI: 10.1109/SPIN52536.2021.9565942
N. Sharma, Anubhav Kumar, A. De, R. K. Jain
A compact and circularly polarized CPW-fed antenna for ISM, Biomedical and satellite applications is presented. The L-shaped stub in the ground is used to improve the impedance matching and perturb the current which is responsible for circular polarization(CP). The |S11| in dB of the antenna varies from 5.35 GHz to 8.88 GHz. The 3dB Axial Ratio of antenna varies from 7.83 GHz to 8.77 GHz. The antenna is analyzed for wearable applications on a three-layer skin phantom model and the SAR value obtained is 0.38 W/Kg, which is below the maximum permissible level.
{"title":"Circularly Polarized CPW-Fed Antenna for ISM (5.8 GHz) and Satellite Communication Applications","authors":"N. Sharma, Anubhav Kumar, A. De, R. K. Jain","doi":"10.1109/SPIN52536.2021.9565942","DOIUrl":"https://doi.org/10.1109/SPIN52536.2021.9565942","url":null,"abstract":"A compact and circularly polarized CPW-fed antenna for ISM, Biomedical and satellite applications is presented. The L-shaped stub in the ground is used to improve the impedance matching and perturb the current which is responsible for circular polarization(CP). The |S11| in dB of the antenna varies from 5.35 GHz to 8.88 GHz. The 3dB Axial Ratio of antenna varies from 7.83 GHz to 8.77 GHz. The antenna is analyzed for wearable applications on a three-layer skin phantom model and the SAR value obtained is 0.38 W/Kg, which is below the maximum permissible level.","PeriodicalId":343177,"journal":{"name":"2021 8th International Conference on Signal Processing and Integrated Networks (SPIN)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125858856","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 : 2021-08-26DOI: 10.1109/SPIN52536.2021.9566016
Jianxin Cao, Shujun Liu, Kui Zhang
Compressed sensing (CS) allows accelerated magnetic resonance imaging (MRI) by highly undersampling k-space data. The key to high quality CS-MRI reconstruction is rational utilization of the sparsity of image in a certain transform domain. Existing CS-MRI methods commonly uses l0 norm or l1 norm to enforce the sparsity of image coefficients but lack parameter adaptation. In this work, a patch level sparse representation is derived from the joint maximum a posteriori (MAP) estimation under a probabilistic model, which adopts a hierarchical prior to characterize sparse image coefficients. The corresponding image reconstruction model is efficiently optimized by alternating direction method of multipliers (ADMM). Simulation results reveal that the proposed approach achieves higher reconstruction performance than competing CS-MRI methods, and is proven to be superior to general Ip norm based methods.
{"title":"Hierarchical prior based sparse representation for compressed sensing MRI","authors":"Jianxin Cao, Shujun Liu, Kui Zhang","doi":"10.1109/SPIN52536.2021.9566016","DOIUrl":"https://doi.org/10.1109/SPIN52536.2021.9566016","url":null,"abstract":"Compressed sensing (CS) allows accelerated magnetic resonance imaging (MRI) by highly undersampling k-space data. The key to high quality CS-MRI reconstruction is rational utilization of the sparsity of image in a certain transform domain. Existing CS-MRI methods commonly uses l0 norm or l1 norm to enforce the sparsity of image coefficients but lack parameter adaptation. In this work, a patch level sparse representation is derived from the joint maximum a posteriori (MAP) estimation under a probabilistic model, which adopts a hierarchical prior to characterize sparse image coefficients. The corresponding image reconstruction model is efficiently optimized by alternating direction method of multipliers (ADMM). Simulation results reveal that the proposed approach achieves higher reconstruction performance than competing CS-MRI methods, and is proven to be superior to general Ip norm based methods.","PeriodicalId":343177,"journal":{"name":"2021 8th International Conference on Signal Processing and Integrated Networks (SPIN)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128309171","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：