Pub Date : 2018-12-01DOI: 10.1109/ICMLA.2018.00068
Manal Almutairi, Frederic T. Stahl, M. Bramer
Rule-based classifiers are considered more expressive, human readable and less prone to over-fitting compared with decision trees, especially when there is noise in the data. Furthermore, rule-based classifiers do not suffer from the replicated subtree problem as classifiers induced by top down induction of decision trees (also known as 'Divide and Conquer'). This research explores some recent developments of a family of rulebased classifiers, the Prism family and more particular G-Prism-FB and G-Prism-DB algorithms, in terms of local discretisation methods used to induce rule terms for continuous data. The paper then proposes a new algorithm of the Prism family based on a combination of Gauss Probability Density Distribution (GPDD), InterQuartile Range (IQR) and data transformation methods. This new rule-based algorithm, termed G-Rules-IQR, is evaluated empirically and outperforms other members of the Prism family in execution time, accuracy and tentative accuracy.
{"title":"A Rule-Based Classifier with Accurate and Fast Rule Term Induction for Continuous Attributes","authors":"Manal Almutairi, Frederic T. Stahl, M. Bramer","doi":"10.1109/ICMLA.2018.00068","DOIUrl":"https://doi.org/10.1109/ICMLA.2018.00068","url":null,"abstract":"Rule-based classifiers are considered more expressive, human readable and less prone to over-fitting compared with decision trees, especially when there is noise in the data. Furthermore, rule-based classifiers do not suffer from the replicated subtree problem as classifiers induced by top down induction of decision trees (also known as 'Divide and Conquer'). This research explores some recent developments of a family of rulebased classifiers, the Prism family and more particular G-Prism-FB and G-Prism-DB algorithms, in terms of local discretisation methods used to induce rule terms for continuous data. The paper then proposes a new algorithm of the Prism family based on a combination of Gauss Probability Density Distribution (GPDD), InterQuartile Range (IQR) and data transformation methods. This new rule-based algorithm, termed G-Rules-IQR, is evaluated empirically and outperforms other members of the Prism family in execution time, accuracy and tentative accuracy.","PeriodicalId":6533,"journal":{"name":"2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"186 1","pages":"413-420"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74949588","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 : 2018-12-01DOI: 10.1109/ICMLA.2018.00032
R. Valisetty, R. Haynes, R. Namburu, Michael Lee
US Army unmanned aerial vehicles (UAVs) in the future will be sustained for longer durations if damage in structural parts is continuously monitored from the damage-inception stage and continuously through vehicle life. Neural networks based machine learning (ML) are proposed, demonstrating that the length of a developing fatigue crack can be estimated continuously using the ultrasound signals. Using a 0.5-TB data set that was obtained from a carefully selected set of experiments, the ML was developed in three stages: 1) feature development, 2) outlier elimination and 3) role of the excitation frequency and exciter-receiver path in the ML of the crack length. In the first stage, the recorded 8000-point ultrasound signals were reduced, first, to 63 features comprising the major statistical features of the returned signal and the seven scales of a seven scale wavelet decomposition of the returned signal. Using an autoencoder algorithm, outliers in the input were identified and removed. A four-layer, 63-32-16-1 neural network based linear regression algorithm was used to predict the crack length from the input features. The results indicated that the ML algorithm gave correlation in the range of 99.43-99.97% when both the exciter-frequency and the exciter-receiver paths are fixed. For investigating the effects of the excitation frequency and the exciter-receiver path on the crack-length information in the returned signal, a similar neural network algorithm was used. One or two additional variables were added to the incoming samples' feature space depending on whether the excitation frequency or the exciter-receiver path or both were variables. ML for crack-length estimation showed promise for these situations, too. In the more practical first situation, where the exciter frequency is fixed and the exciter-receiver path is uncertain, the algorithm showed an accuracy in the range of 96.97-98.92%. This algorithm still gave a correlation above 85% when there was uncertainty in the excitation frequency and exciter-receiver paths, as well. This work thus demonstrates the potential for monitoring fatigue crack length growth throughout the life of a vehicle for an increased sustainment of the US Army UAVs.
{"title":"Machine Learning for US Army UAVs Sustainment: Assessing Effect of Sensor Frequency and Placement on Damage Information in the Ultrasound Signals","authors":"R. Valisetty, R. Haynes, R. Namburu, Michael Lee","doi":"10.1109/ICMLA.2018.00032","DOIUrl":"https://doi.org/10.1109/ICMLA.2018.00032","url":null,"abstract":"US Army unmanned aerial vehicles (UAVs) in the future will be sustained for longer durations if damage in structural parts is continuously monitored from the damage-inception stage and continuously through vehicle life. Neural networks based machine learning (ML) are proposed, demonstrating that the length of a developing fatigue crack can be estimated continuously using the ultrasound signals. Using a 0.5-TB data set that was obtained from a carefully selected set of experiments, the ML was developed in three stages: 1) feature development, 2) outlier elimination and 3) role of the excitation frequency and exciter-receiver path in the ML of the crack length. In the first stage, the recorded 8000-point ultrasound signals were reduced, first, to 63 features comprising the major statistical features of the returned signal and the seven scales of a seven scale wavelet decomposition of the returned signal. Using an autoencoder algorithm, outliers in the input were identified and removed. A four-layer, 63-32-16-1 neural network based linear regression algorithm was used to predict the crack length from the input features. The results indicated that the ML algorithm gave correlation in the range of 99.43-99.97% when both the exciter-frequency and the exciter-receiver paths are fixed. For investigating the effects of the excitation frequency and the exciter-receiver path on the crack-length information in the returned signal, a similar neural network algorithm was used. One or two additional variables were added to the incoming samples' feature space depending on whether the excitation frequency or the exciter-receiver path or both were variables. ML for crack-length estimation showed promise for these situations, too. In the more practical first situation, where the exciter frequency is fixed and the exciter-receiver path is uncertain, the algorithm showed an accuracy in the range of 96.97-98.92%. This algorithm still gave a correlation above 85% when there was uncertainty in the excitation frequency and exciter-receiver paths, as well. This work thus demonstrates the potential for monitoring fatigue crack length growth throughout the life of a vehicle for an increased sustainment of the US Army UAVs.","PeriodicalId":6533,"journal":{"name":"2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"60 1","pages":"165-172"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73795856","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 : 2018-12-01DOI: 10.1109/ICMLA.2018.00061
Farzana Anowar, S. Sadaoui, Malek Mouhoub
Online auctions created a very attractive environment for dishonest moneymakers who can commit different types of fraud. Shill Bidding (SB) is the most predominant auction fraud and also the most difficult to detect because of its similarity to usual bidding behavior. Based on a newly produced SB dataset, in this study, we devise a fraud classification model that is able to efficiently differentiate between honest and malicious bidders. First, we label the SB data by combining a hierarchical clustering technique and a semi-automated labeling approach. To solve the imbalanced learning problem, we apply several advanced data sampling methods and compare their performance using the SVM model. As a result, we develop an optimal SB classifier that exhibits very satisfactory detection and low misclassification rates.
{"title":"Auction Fraud Classification Based on Clustering and Sampling Techniques","authors":"Farzana Anowar, S. Sadaoui, Malek Mouhoub","doi":"10.1109/ICMLA.2018.00061","DOIUrl":"https://doi.org/10.1109/ICMLA.2018.00061","url":null,"abstract":"Online auctions created a very attractive environment for dishonest moneymakers who can commit different types of fraud. Shill Bidding (SB) is the most predominant auction fraud and also the most difficult to detect because of its similarity to usual bidding behavior. Based on a newly produced SB dataset, in this study, we devise a fraud classification model that is able to efficiently differentiate between honest and malicious bidders. First, we label the SB data by combining a hierarchical clustering technique and a semi-automated labeling approach. To solve the imbalanced learning problem, we apply several advanced data sampling methods and compare their performance using the SVM model. As a result, we develop an optimal SB classifier that exhibits very satisfactory detection and low misclassification rates.","PeriodicalId":6533,"journal":{"name":"2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"25 1","pages":"366-371"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84780762","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 : 2018-12-01DOI: 10.1109/ICMLA.2018.00153
R. LeMoyne, Timothy Mastroianni
The smartphone represents a wearable and wireless system with the potential to have transformative influence on the biomedical and healthcare industry. An intrinsic feature of the smartphone is a gyroscope sensor, for which with a software application the smartphone functions as a wearable and wireless gyroscope platform. The resultant gyroscope data recording presents a clinical recognizable signal, which has been successful demonstrated to quantify aspects of human movement characteristics, such as the patellar tendon reflex. Gait another associated feature of human movement can be readily quantified by a smartphone functioning as a wearable and wireless gyroscope platform. The research objective is to distinguish between an affected leg and unaffected leg during hemiplegic gait based on a smartphone functioning as a wearable and wireless gyroscope platform though machine learning classification. A single smartphone is applied to quantify hemiplegic gait. The smartphone is first mounted to the affected leg and then the unaffected leg with velocity constrained to a constant velocity by a treadmill. Through wireless connectivity to the Internet the gyroscope signal data is conveyed as an email attachment for post-processing at a remote location. Software automation consolidates the gyroscope signal data of hemiplegic gait to a feature set for machine learning classification. With the application of a multilayer perceptron neural network considerable classification accuracy is attained for distinguishing between the affected leg and unaffected leg of hemiplegic gait. Future implications of the successful implementation of a smartphone as a wearable and wireless gyroscope for machine learning classification of hemiplegic gait through a multilayer perceptron neural network elucidate pathways to highly optimized therapy through machine learning with the potential for patients to reside remote from their therapist.
{"title":"Implementation of a Smartphone as a Wearable and Wireless Gyroscope Platform for Machine Learning Classification of Hemiplegic Gait Through a Multilayer Perceptron Neural Network","authors":"R. LeMoyne, Timothy Mastroianni","doi":"10.1109/ICMLA.2018.00153","DOIUrl":"https://doi.org/10.1109/ICMLA.2018.00153","url":null,"abstract":"The smartphone represents a wearable and wireless system with the potential to have transformative influence on the biomedical and healthcare industry. An intrinsic feature of the smartphone is a gyroscope sensor, for which with a software application the smartphone functions as a wearable and wireless gyroscope platform. The resultant gyroscope data recording presents a clinical recognizable signal, which has been successful demonstrated to quantify aspects of human movement characteristics, such as the patellar tendon reflex. Gait another associated feature of human movement can be readily quantified by a smartphone functioning as a wearable and wireless gyroscope platform. The research objective is to distinguish between an affected leg and unaffected leg during hemiplegic gait based on a smartphone functioning as a wearable and wireless gyroscope platform though machine learning classification. A single smartphone is applied to quantify hemiplegic gait. The smartphone is first mounted to the affected leg and then the unaffected leg with velocity constrained to a constant velocity by a treadmill. Through wireless connectivity to the Internet the gyroscope signal data is conveyed as an email attachment for post-processing at a remote location. Software automation consolidates the gyroscope signal data of hemiplegic gait to a feature set for machine learning classification. With the application of a multilayer perceptron neural network considerable classification accuracy is attained for distinguishing between the affected leg and unaffected leg of hemiplegic gait. Future implications of the successful implementation of a smartphone as a wearable and wireless gyroscope for machine learning classification of hemiplegic gait through a multilayer perceptron neural network elucidate pathways to highly optimized therapy through machine learning with the potential for patients to reside remote from their therapist.","PeriodicalId":6533,"journal":{"name":"2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"62 1","pages":"946-950"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85197097","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 : 2018-12-01DOI: 10.1109/ICMLA.2018.00049
Trevor Morris, Tiffany Chien, Eric L. Goodman
In this paper we apply Convolutional Neural Networks (CNNs) to the task of automatic threat detection, specifically conventional explosives, in security X-ray scans of passenger baggage. We present the first results of utilizing CNNs for explosives detection, and introduce a dataset, the Passenger Baggage Object Database (PBOD), which can be used by researchers to develop new threat detection algorithms. Using state-of-the-art CNN models and taking advantage of the properties of the Xray scanner, we achieve reliable detection of threats, with the best model achieving an AUC of the ROC of 0.95. We also explore heatmaps as a visualization of the location of the threat.
{"title":"Convolutional Neural Networks for Automatic Threat Detection in Security X-Ray Images","authors":"Trevor Morris, Tiffany Chien, Eric L. Goodman","doi":"10.1109/ICMLA.2018.00049","DOIUrl":"https://doi.org/10.1109/ICMLA.2018.00049","url":null,"abstract":"In this paper we apply Convolutional Neural Networks (CNNs) to the task of automatic threat detection, specifically conventional explosives, in security X-ray scans of passenger baggage. We present the first results of utilizing CNNs for explosives detection, and introduce a dataset, the Passenger Baggage Object Database (PBOD), which can be used by researchers to develop new threat detection algorithms. Using state-of-the-art CNN models and taking advantage of the properties of the Xray scanner, we achieve reliable detection of threats, with the best model achieving an AUC of the ROC of 0.95. We also explore heatmaps as a visualization of the location of the threat.","PeriodicalId":6533,"journal":{"name":"2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"8 1","pages":"285-292"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81898814","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 : 2018-12-01DOI: 10.1109/ICMLA.2018.00221
Lei Zhang
The goal of this research is to develop an efficient artificial neural network (ANN) architecture to predict three chaotic time series outputs for Lorenz system using single time series input. The training performances are evaluated and compared for different ANN architectures with multiple hidden layers, as well as for input data with different combination of time series, including the first and second order differences of the time series. It is found that given the same ANN architecture, the training results of multiple time series outputs using single time series (x) input are much worse than those using multiple time series inputs. However, the training results can be improved significantly by increasing the number of ANN hidden layers up to 3; and marginally improved by adding the first and second order differences of the x time series, as well as adding steps for calculating the first and second order differences of the input time series.
{"title":"Lorenz Chaotic System Artificial Neural Network Training with Single Time Series Input and Multiple Time Series Outputs for EEG Prediction","authors":"Lei Zhang","doi":"10.1109/ICMLA.2018.00221","DOIUrl":"https://doi.org/10.1109/ICMLA.2018.00221","url":null,"abstract":"The goal of this research is to develop an efficient artificial neural network (ANN) architecture to predict three chaotic time series outputs for Lorenz system using single time series input. The training performances are evaluated and compared for different ANN architectures with multiple hidden layers, as well as for input data with different combination of time series, including the first and second order differences of the time series. It is found that given the same ANN architecture, the training results of multiple time series outputs using single time series (x) input are much worse than those using multiple time series inputs. However, the training results can be improved significantly by increasing the number of ANN hidden layers up to 3; and marginally improved by adding the first and second order differences of the x time series, as well as adding steps for calculating the first and second order differences of the input time series.","PeriodicalId":6533,"journal":{"name":"2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"149 1","pages":"1358-1365"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79423964","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 : 2018-12-01DOI: 10.1109/ICMLA.2018.00217
Ion Freeman, Ashley Haigler, Suzanna E. Schmeelk, Lisa Ellrodt, Tonya Fields
This paper examines industry-based doctoral dissertation research in a professional computing doctoral program for full time working professionals through the lens of different machine learning algorithms to understand topics explored by full time working industry professionals. This research paper examines machine learning algorithms and the IBM Watson Discovery machine learning tool to categorize dissertation research topics defended at Pace University. The research provides insights into differences in machine learning algorithm categorization using natural language processing.
{"title":"What are they Researching? Examining Industry-Based Doctoral Dissertation Research through the Lens of Machine Learning","authors":"Ion Freeman, Ashley Haigler, Suzanna E. Schmeelk, Lisa Ellrodt, Tonya Fields","doi":"10.1109/ICMLA.2018.00217","DOIUrl":"https://doi.org/10.1109/ICMLA.2018.00217","url":null,"abstract":"This paper examines industry-based doctoral dissertation research in a professional computing doctoral program for full time working professionals through the lens of different machine learning algorithms to understand topics explored by full time working industry professionals. This research paper examines machine learning algorithms and the IBM Watson Discovery machine learning tool to categorize dissertation research topics defended at Pace University. The research provides insights into differences in machine learning algorithm categorization using natural language processing.","PeriodicalId":6533,"journal":{"name":"2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"58 1","pages":"1338-1340"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84353855","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 : 2018-12-01DOI: 10.1109/ICMLA.2018.00039
Youngha Hwang, S. Gelfand
Dynamic time warping (DTW) has been applied to a wide range of machine learning problems involving the comparison of time series. An important feature of such time series is that they can sometimes be sparse in the sense that the data takes zero value at many epochs. This corresponds for example to quiet periods in speech or to a lack of physical activity. However, employing conventional DTW for such sparse time series runs a full search ignoring the zero data. So a fast dynamic time warping algorithm that is exactly equivalent to DTW was developed for the unconstrained case where there is no global constraint on the permissible warping path. It was called sparse dynamic time warping (SDTW). In this paper we focus on the development and analysis of a fast dynamic time warping algorithm for the constrained case where there is a global constraint on the permissible warping path, specifically limit the width along the diagonal of the permissible path domain. We call this constrained sparse dynamic time warping (CSDTW). A careful formulation and analysis are performed to determine exactly how CSDTW should treat the zero data. It is shown that CSDTW reduces the computational complexity relative to constrained DTW by about three times the sparsity ratio, which is defined as the arithmetic mean of the fraction of non-zero's in the two time series. Numerical experiments confirm the speed advantage of CSDTW relative to constrained DTW for sparse time series with sparsity ratio up to 0.2-0.3. This study provides a benchmark and also background to potentially understand how to exploit such sparsity when the underlying time series is approximated to reduce complexity.
{"title":"Constrained Sparse Dynamic Time Warping","authors":"Youngha Hwang, S. Gelfand","doi":"10.1109/ICMLA.2018.00039","DOIUrl":"https://doi.org/10.1109/ICMLA.2018.00039","url":null,"abstract":"Dynamic time warping (DTW) has been applied to a wide range of machine learning problems involving the comparison of time series. An important feature of such time series is that they can sometimes be sparse in the sense that the data takes zero value at many epochs. This corresponds for example to quiet periods in speech or to a lack of physical activity. However, employing conventional DTW for such sparse time series runs a full search ignoring the zero data. So a fast dynamic time warping algorithm that is exactly equivalent to DTW was developed for the unconstrained case where there is no global constraint on the permissible warping path. It was called sparse dynamic time warping (SDTW). In this paper we focus on the development and analysis of a fast dynamic time warping algorithm for the constrained case where there is a global constraint on the permissible warping path, specifically limit the width along the diagonal of the permissible path domain. We call this constrained sparse dynamic time warping (CSDTW). A careful formulation and analysis are performed to determine exactly how CSDTW should treat the zero data. It is shown that CSDTW reduces the computational complexity relative to constrained DTW by about three times the sparsity ratio, which is defined as the arithmetic mean of the fraction of non-zero's in the two time series. Numerical experiments confirm the speed advantage of CSDTW relative to constrained DTW for sparse time series with sparsity ratio up to 0.2-0.3. This study provides a benchmark and also background to potentially understand how to exploit such sparsity when the underlying time series is approximated to reduce complexity.","PeriodicalId":6533,"journal":{"name":"2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"38 1","pages":"216-222"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82307584","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 : 2018-12-01DOI: 10.1109/ICMLA.2018.00065
Myeongseob Ko, Donghyun Kim, Kwangtaek Kim
The development of vision technology for observation of skin surface and diagnosis of skin disease for preventing secondary infections caused by direct skin touch has consistently been in the medical field spotlight. Many studies have been conducted to acquire three dimensional (3D) data through stereo images, multiple images, and lasers because (3D) data of in-vivo skin image is essential for accurate medical diagnosis. However, stereo vision systems or 3D laser systems for obtaining 3D information require high cost and have high computational complexity, and hence they have not been used universally. Additionally, the use of such systems is still not preferred in the medical field due to limitations on visual decision making. Therefore, a haptic diagnosis system that can blend vision information from a camera and palpation information from a dermatologist has been considered. In this study, we propose a 3D skin surface reconstruction method using a light field camera for haptic rendering and palpation. To achieve this goal, we addressed the low resolution problem, which has been consistently present in light field cameras, through the generative adversarial nets (GANs)-based super resolution method, and exploited the light field system which has been applied only to the object scene for obtaining 3D skin surface texture. Experimental results show that the method proposed in this study is promising and offers sufficient potential for haptic diagnosis.
{"title":"GAN-Based Super Resolution for Accurate 3D Surface Reconstruction from Light Field Skin Images Towards Haptic Palpation","authors":"Myeongseob Ko, Donghyun Kim, Kwangtaek Kim","doi":"10.1109/ICMLA.2018.00065","DOIUrl":"https://doi.org/10.1109/ICMLA.2018.00065","url":null,"abstract":"The development of vision technology for observation of skin surface and diagnosis of skin disease for preventing secondary infections caused by direct skin touch has consistently been in the medical field spotlight. Many studies have been conducted to acquire three dimensional (3D) data through stereo images, multiple images, and lasers because (3D) data of in-vivo skin image is essential for accurate medical diagnosis. However, stereo vision systems or 3D laser systems for obtaining 3D information require high cost and have high computational complexity, and hence they have not been used universally. Additionally, the use of such systems is still not preferred in the medical field due to limitations on visual decision making. Therefore, a haptic diagnosis system that can blend vision information from a camera and palpation information from a dermatologist has been considered. In this study, we propose a 3D skin surface reconstruction method using a light field camera for haptic rendering and palpation. To achieve this goal, we addressed the low resolution problem, which has been consistently present in light field cameras, through the generative adversarial nets (GANs)-based super resolution method, and exploited the light field system which has been applied only to the object scene for obtaining 3D skin surface texture. Experimental results show that the method proposed in this study is promising and offers sufficient potential for haptic diagnosis.","PeriodicalId":6533,"journal":{"name":"2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"227 1","pages":"392-397"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80172944","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 : 2018-12-01DOI: 10.1109/ICMLA.2018.00209
V. Vadakattu, S. Suthaharan
The goal of this paper is to perform an experimental research and show that simple statistical predictors can reveal usage patterns of the electrical appliances from smart meter and sensor readings. We used an open data set of Smart* project and its real power and apparent power variability to accomplish this goal. We generated the predictors using block-based statistical information of the real power and apparent power associated with each appliance class type. We constructed five machine learning models using these predictors and evaluated them using random forest classification and the qualitative measures – classification accuracy, out-of-bag error, and misclassification error. Our finding is that the simple statistical predictors that reveal smart home occupants appliance usage patterns and energy consumption details can be obtained through smart home data analytics. Our finding includes that the statistical predictors generated from apparent power can improve the accuracy of the significantly-imbalanced smart home data classification.
{"title":"Feature Extraction Using Apparent Power and Real Power for Smart Home Data Classification","authors":"V. Vadakattu, S. Suthaharan","doi":"10.1109/ICMLA.2018.00209","DOIUrl":"https://doi.org/10.1109/ICMLA.2018.00209","url":null,"abstract":"The goal of this paper is to perform an experimental research and show that simple statistical predictors can reveal usage patterns of the electrical appliances from smart meter and sensor readings. We used an open data set of Smart* project and its real power and apparent power variability to accomplish this goal. We generated the predictors using block-based statistical information of the real power and apparent power associated with each appliance class type. We constructed five machine learning models using these predictors and evaluated them using random forest classification and the qualitative measures – classification accuracy, out-of-bag error, and misclassification error. Our finding is that the simple statistical predictors that reveal smart home occupants appliance usage patterns and energy consumption details can be obtained through smart home data analytics. Our finding includes that the statistical predictors generated from apparent power can improve the accuracy of the significantly-imbalanced smart home data classification.","PeriodicalId":6533,"journal":{"name":"2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"2 1","pages":"1290-1295"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80752821","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：