Pub Date : 2023-03-20DOI: 10.20965/jaciii.2023.p0198
Wei Liu, Zhang Tao, Yun Ma, Longsheng Wei
In this study, we present a three-dimensional (3D) object detection algorithm based on monocular images by constructing an end-to-end network, that incorporates depth information. The entire network consists of three parts. The first part includes the basic object detection neural network as the main body, that uses the region proposal network to obtain the two-dimensional (2D) region proposal of the object. The second part is the depth estimation branch network, that obtains the depth information of the object pixels and calculates the corresponding 3D point cloud. In the last part, concatenated features obtained from the aforementioned two parts are fed into the fully-connected layers. Subsequently, 2D and 3D detection results are obtained. Compared with certain existing methods, the accuracy of the detection results is improved in this study.
{"title":"3D Street Object Detection from Monocular Images Using Deep Learning and Depth Information","authors":"Wei Liu, Zhang Tao, Yun Ma, Longsheng Wei","doi":"10.20965/jaciii.2023.p0198","DOIUrl":"https://doi.org/10.20965/jaciii.2023.p0198","url":null,"abstract":"In this study, we present a three-dimensional (3D) object detection algorithm based on monocular images by constructing an end-to-end network, that incorporates depth information. The entire network consists of three parts. The first part includes the basic object detection neural network as the main body, that uses the region proposal network to obtain the two-dimensional (2D) region proposal of the object. The second part is the depth estimation branch network, that obtains the depth information of the object pixels and calculates the corresponding 3D point cloud. In the last part, concatenated features obtained from the aforementioned two parts are fed into the fully-connected layers. Subsequently, 2D and 3D detection results are obtained. Compared with certain existing methods, the accuracy of the detection results is improved in this study.","PeriodicalId":45921,"journal":{"name":"Journal of Advanced Computational Intelligence and Intelligent Informatics","volume":"71 1","pages":"198-206"},"PeriodicalIF":0.7,"publicationDate":"2023-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84932090","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 : 2023-03-20DOI: 10.20965/jaciii.2023.p0135
Dianjun Wang, Meng Xu, Ya Chen, Haoxiang Zhong, Y. Zhu, Zilong Wang, Linlin Gao
To reduce the error of the odometer positioning system and improve the positioning accuracy of four-wheel-steering mobile robots, three types of coupling errors are considered, based on the University of Michigan Benchmark (UMBmark) method: unequal track width, unequal wheel diameter, and speed difference of ipsilateral wheels. A “dual direction square path experiment” is designed to decouple the error, a new system error model is defined, and an improved UMBmark method for a four-wheel mobile robot is proposed. In the mobile robot positioning system, a laser tracker is used to measure the absolute positions of the starting and ending points of the robot. The positioning test results of the robot using the improved UMBmark method show that the odometer system error is 69.103 mm, which is 2.6 times less than that in the traditional UMBmark method. Hence, the improved UMBmark can better compensate for the system error of four-wheel-steering mobile robots.
{"title":"Positioning Method of Four-Wheel-Steering Mobile Robots Based on Improved UMBmark of Michigan Benchmark Algorithm","authors":"Dianjun Wang, Meng Xu, Ya Chen, Haoxiang Zhong, Y. Zhu, Zilong Wang, Linlin Gao","doi":"10.20965/jaciii.2023.p0135","DOIUrl":"https://doi.org/10.20965/jaciii.2023.p0135","url":null,"abstract":"To reduce the error of the odometer positioning system and improve the positioning accuracy of four-wheel-steering mobile robots, three types of coupling errors are considered, based on the University of Michigan Benchmark (UMBmark) method: unequal track width, unequal wheel diameter, and speed difference of ipsilateral wheels. A “dual direction square path experiment” is designed to decouple the error, a new system error model is defined, and an improved UMBmark method for a four-wheel mobile robot is proposed. In the mobile robot positioning system, a laser tracker is used to measure the absolute positions of the starting and ending points of the robot. The positioning test results of the robot using the improved UMBmark method show that the odometer system error is 69.103 mm, which is 2.6 times less than that in the traditional UMBmark method. Hence, the improved UMBmark can better compensate for the system error of four-wheel-steering mobile robots.","PeriodicalId":45921,"journal":{"name":"Journal of Advanced Computational Intelligence and Intelligent Informatics","volume":"65 1","pages":"135-142"},"PeriodicalIF":0.7,"publicationDate":"2023-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76562596","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 : 2023-03-20DOI: 10.20965/jaciii.2023.p0148
Toumi Ohara, Fumiya Kinoshita
In recent years, as eye movement measurement devices have become relatively cheap, many attempts have been made to quantitatively evaluate covert attention by focusing on microsaccades. However, the measurement of microsaccades still has many unclear points, and a unified analysis method is still lacking. As such, the interpretation of results differs among different research groups. To solve this problem, it is important to conduct empirical studies on microsaccades to evaluate them using a unified method. In this study, we conducted an empirical experiment on the effects of alcohol consumption on microsaccades by temporarily suppressing cerebellar activity with alcohol consumption. The results showed that the frequency of microsaccades was significantly reduced after 30, 50, and 70 min of drinking compared to after drinking (p< 0.05). These results suggest that the decrease in brain function caused by alcohol consumption suppresses the frequency of microsaccades, and that this may be the cause of constriction in the peripheral visual field when drinking.
{"title":"Effect of Alcohol Consumption on the Frequency of Microsaccades","authors":"Toumi Ohara, Fumiya Kinoshita","doi":"10.20965/jaciii.2023.p0148","DOIUrl":"https://doi.org/10.20965/jaciii.2023.p0148","url":null,"abstract":"In recent years, as eye movement measurement devices have become relatively cheap, many attempts have been made to quantitatively evaluate covert attention by focusing on microsaccades. However, the measurement of microsaccades still has many unclear points, and a unified analysis method is still lacking. As such, the interpretation of results differs among different research groups. To solve this problem, it is important to conduct empirical studies on microsaccades to evaluate them using a unified method. In this study, we conducted an empirical experiment on the effects of alcohol consumption on microsaccades by temporarily suppressing cerebellar activity with alcohol consumption. The results showed that the frequency of microsaccades was significantly reduced after 30, 50, and 70 min of drinking compared to after drinking (p< 0.05). These results suggest that the decrease in brain function caused by alcohol consumption suppresses the frequency of microsaccades, and that this may be the cause of constriction in the peripheral visual field when drinking.","PeriodicalId":45921,"journal":{"name":"Journal of Advanced Computational Intelligence and Intelligent Informatics","volume":"19 1","pages":"148-153"},"PeriodicalIF":0.7,"publicationDate":"2023-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86949661","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 : 2023-03-20DOI: 10.20965/jaciii.2023.p0173
Nozomu Ohta, Shin Kawai, H. Nobuhara
A capsule network (CapsNet) is a deep learning model for image classification that provides robustness to changes in the poses of objects in the images. A capsule is a vector whose direction represents the presence, position, size, and pose of an object. However, with CapsNet, the distribution of capsules is concentrated in a class, and the number of capsules increases with the number of classes. In addition, learning is computationally expensive for a CapsNet. We proposed a method to increase the diversity of capsule directions and decrease the computational cost of CapsNet training by allowing a single capsule to represent multiple object classes. To determine the distance between classes, we used an additive angular margin loss called ArcFace. To validate the proposed method, the distribution of the capsules was determined using principal component analysis to validate the proposed method. In addition, using the MNIST, fashion-MNIST, EMNIST, SVHN, and CIFAR-10 datasets, as well as the corresponding affine-transformed datasets, we determined the accuracy and training time of the proposed method and original CapsNet. The accuracy of the proposed method improved by 8.91% on the CIFAR-10 dataset, and the training time reduced by more than 19% for each dataset compared with those of the original CapsNets.
{"title":"Capsule Network Extension Based on Metric Learning","authors":"Nozomu Ohta, Shin Kawai, H. Nobuhara","doi":"10.20965/jaciii.2023.p0173","DOIUrl":"https://doi.org/10.20965/jaciii.2023.p0173","url":null,"abstract":"A capsule network (CapsNet) is a deep learning model for image classification that provides robustness to changes in the poses of objects in the images. A capsule is a vector whose direction represents the presence, position, size, and pose of an object. However, with CapsNet, the distribution of capsules is concentrated in a class, and the number of capsules increases with the number of classes. In addition, learning is computationally expensive for a CapsNet. We proposed a method to increase the diversity of capsule directions and decrease the computational cost of CapsNet training by allowing a single capsule to represent multiple object classes. To determine the distance between classes, we used an additive angular margin loss called ArcFace. To validate the proposed method, the distribution of the capsules was determined using principal component analysis to validate the proposed method. In addition, using the MNIST, fashion-MNIST, EMNIST, SVHN, and CIFAR-10 datasets, as well as the corresponding affine-transformed datasets, we determined the accuracy and training time of the proposed method and original CapsNet. The accuracy of the proposed method improved by 8.91% on the CIFAR-10 dataset, and the training time reduced by more than 19% for each dataset compared with those of the original CapsNets.","PeriodicalId":45921,"journal":{"name":"Journal of Advanced Computational Intelligence and Intelligent Informatics","volume":"269 1","pages":"173-181"},"PeriodicalIF":0.7,"publicationDate":"2023-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78743763","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 : 2023-03-20DOI: 10.20965/jaciii.2023.p0190
Yankun Shen, Yaya Su, L. Wang, Dongli Jia
In recent years, the use of deep learning in image inpainting has yielded positive results. However, existing image inpainting algorithms do not pay sufficient attention to the structural and textural features of the image when inpainting, which leads to issues in the inpainting results such as blurring and distortion. To solve the above problems, a channel attention mechanism was introduced to emphasize the importance of structure and texture after extraction by the convolutional network. A bidirectional gated feature fusion module was employed to exchange and fuse the structural and textural features, ensuring the overall consistency of the image. In addition, the features of the image were better captured by selecting a deformable convolution that can adapt the receptive field to replace the ordinary convolution in the contextual feature aggregation module. This resulted in highly vivid and realistic restoration results with more reasonable details. The experiments showed that, compared with the current mainstream network, the repair results of this algorithm were more realistic, and the superiority of this algorithm was proved by qualitative and quantitative experiments.
{"title":"Research on Image Inpainting Algorithms Based on Attention Guidance","authors":"Yankun Shen, Yaya Su, L. Wang, Dongli Jia","doi":"10.20965/jaciii.2023.p0190","DOIUrl":"https://doi.org/10.20965/jaciii.2023.p0190","url":null,"abstract":"In recent years, the use of deep learning in image inpainting has yielded positive results. However, existing image inpainting algorithms do not pay sufficient attention to the structural and textural features of the image when inpainting, which leads to issues in the inpainting results such as blurring and distortion. To solve the above problems, a channel attention mechanism was introduced to emphasize the importance of structure and texture after extraction by the convolutional network. A bidirectional gated feature fusion module was employed to exchange and fuse the structural and textural features, ensuring the overall consistency of the image. In addition, the features of the image were better captured by selecting a deformable convolution that can adapt the receptive field to replace the ordinary convolution in the contextual feature aggregation module. This resulted in highly vivid and realistic restoration results with more reasonable details. The experiments showed that, compared with the current mainstream network, the repair results of this algorithm were more realistic, and the superiority of this algorithm was proved by qualitative and quantitative experiments.","PeriodicalId":45921,"journal":{"name":"Journal of Advanced Computational Intelligence and Intelligent Informatics","volume":"18 1","pages":"190-197"},"PeriodicalIF":0.7,"publicationDate":"2023-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91052634","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 : 2023-01-20DOI: 10.20965/jaciii.2023.p0027
Mike Louie C. Enriquez, Ronnie S. Concepcion, R. Relano, Kate G. Francisco, Jonah Jahara G. Baun, Adrian Genevie G. Janairo, R. Baldovino, R. R. Vicerra, A. Bandala, E. Dadios
In performing underground imaging surveying, applying a coating in the antenna dipole plates with robust and durable material to stay protected against rough road features is vital to consider. By doing this, the mechanical properties of the metallic antenna dipole can be improved and be shielded from deterioration. With that, this study has developed an indentation-flexure algorithm optimized using a hybrid Lichtenberg algorithm (LA) and artificial neural network (ANN) that can predict the indentation-flexure as a function of the coating material’s elastic modulus, Poisson ratio, and thickness as well as the load antenna weight. Acrylic, epoxy, nylon 101, high-density polyethylene, and polyvinyl chloride were chosen as the top five most popular coating materials. A 120° titanium cone indenter with a 0.5-inch-diameter, slightly rounded point, and a constant compressive force of 200 N in the center was employed to plot and use a nonlinear mechanical finite element analysis on an antenna dipole plate using SolidWorks. Nature-inspired and evolutionary metaheuristics such as African vultures, Lichtenberg, and gorilla troop optimization algorithm including genetic algorithm (GA) were employed as optimized models for the hardness indentation for capacitively coupled antenna dipoles. Based on the results, the hybrid LA-ANN solution with a hidden neurons of 3000 and a sigmoid activation function is the best performing model as it acquired a MSE score of 0.0061 in validation and 0.1478 in testing compare to the other model with 0.1610 for GA with 100 hidden neurons with sigmoid activation function. Thus, LA-ANN model is considered as the optIFnet as it exhibited the best prediction performance and fastest convergence among all optimizers used.
{"title":"optIFnet: A Capacitive Antenna Dipole Indention-Flexure Predictive Model Optimized Using Hybrid Lichtenberg Algorithm and Neural Network","authors":"Mike Louie C. Enriquez, Ronnie S. Concepcion, R. Relano, Kate G. Francisco, Jonah Jahara G. Baun, Adrian Genevie G. Janairo, R. Baldovino, R. R. Vicerra, A. Bandala, E. Dadios","doi":"10.20965/jaciii.2023.p0027","DOIUrl":"https://doi.org/10.20965/jaciii.2023.p0027","url":null,"abstract":"In performing underground imaging surveying, applying a coating in the antenna dipole plates with robust and durable material to stay protected against rough road features is vital to consider. By doing this, the mechanical properties of the metallic antenna dipole can be improved and be shielded from deterioration. With that, this study has developed an indentation-flexure algorithm optimized using a hybrid Lichtenberg algorithm (LA) and artificial neural network (ANN) that can predict the indentation-flexure as a function of the coating material’s elastic modulus, Poisson ratio, and thickness as well as the load antenna weight. Acrylic, epoxy, nylon 101, high-density polyethylene, and polyvinyl chloride were chosen as the top five most popular coating materials. A 120° titanium cone indenter with a 0.5-inch-diameter, slightly rounded point, and a constant compressive force of 200 N in the center was employed to plot and use a nonlinear mechanical finite element analysis on an antenna dipole plate using SolidWorks. Nature-inspired and evolutionary metaheuristics such as African vultures, Lichtenberg, and gorilla troop optimization algorithm including genetic algorithm (GA) were employed as optimized models for the hardness indentation for capacitively coupled antenna dipoles. Based on the results, the hybrid LA-ANN solution with a hidden neurons of 3000 and a sigmoid activation function is the best performing model as it acquired a MSE score of 0.0061 in validation and 0.1478 in testing compare to the other model with 0.1610 for GA with 100 hidden neurons with sigmoid activation function. Thus, LA-ANN model is considered as the optIFnet as it exhibited the best prediction performance and fastest convergence among all optimizers used.","PeriodicalId":45921,"journal":{"name":"Journal of Advanced Computational Intelligence and Intelligent Informatics","volume":"3 1","pages":"27-34"},"PeriodicalIF":0.7,"publicationDate":"2023-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79281931","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 : 2023-01-20DOI: 10.20965/jaciii.2023.p0054
Etsuro Nakamura, Y. Kageyama, Satoshi Hirose
In recent years, several Japanese companies have attempted to improve the efficiency of their meetings, which has been a significant challenge. For instance, voice recognition technology is used to considerably improve meeting minutes creation. In an automatic minutes-creating system, identifying the speaker to add speaker information to the text would substantially improve the overall efficiency of the process. Therefore, a few companies and research groups have proposed speaker estimation methods; however, it includes challenges, such as requiring advance preparation, special equipment, and multiple microphones. These problems can be solved by using speech sections that are extracted from lip movements and voice information. When a person speaks, voice and lip movements occur simultaneously. Therefore, the speaker’s speech section can be extracted from videos by using lip movement and voice information. However, when this speech section contains only voice information, the voiceprint information of each meeting participant is required for speaker identification. When using lip movements, the speech section and speaker position can be extracted without the voiceprint information. Therefore, in this study, we propose a speech-section extraction method that uses image and voice information in Japanese for speaker identification. The proposed method consists of three processes: i) the extraction of speech frames using lip movements, ii) the extraction of speech frames using voices, and iii) the classification of speech sections using these extraction results. We used video data to evaluate the functionality of the method. Further, the proposed method was compared with state-of-the-art techniques. The average F-measure of the proposed method is determined to be higher than that of the conventional methods that are based on state-of-the-art techniques. The evaluation results showed that the proposed method achieves state-of-the-art performance using a simpler process compared to the conventional method.
{"title":"Speech-Section Extraction Using Lip Movement and Voice Information in Japanese","authors":"Etsuro Nakamura, Y. Kageyama, Satoshi Hirose","doi":"10.20965/jaciii.2023.p0054","DOIUrl":"https://doi.org/10.20965/jaciii.2023.p0054","url":null,"abstract":"In recent years, several Japanese companies have attempted to improve the efficiency of their meetings, which has been a significant challenge. For instance, voice recognition technology is used to considerably improve meeting minutes creation. In an automatic minutes-creating system, identifying the speaker to add speaker information to the text would substantially improve the overall efficiency of the process. Therefore, a few companies and research groups have proposed speaker estimation methods; however, it includes challenges, such as requiring advance preparation, special equipment, and multiple microphones. These problems can be solved by using speech sections that are extracted from lip movements and voice information. When a person speaks, voice and lip movements occur simultaneously. Therefore, the speaker’s speech section can be extracted from videos by using lip movement and voice information. However, when this speech section contains only voice information, the voiceprint information of each meeting participant is required for speaker identification. When using lip movements, the speech section and speaker position can be extracted without the voiceprint information. Therefore, in this study, we propose a speech-section extraction method that uses image and voice information in Japanese for speaker identification. The proposed method consists of three processes: i) the extraction of speech frames using lip movements, ii) the extraction of speech frames using voices, and iii) the classification of speech sections using these extraction results. We used video data to evaluate the functionality of the method. Further, the proposed method was compared with state-of-the-art techniques. The average F-measure of the proposed method is determined to be higher than that of the conventional methods that are based on state-of-the-art techniques. The evaluation results showed that the proposed method achieves state-of-the-art performance using a simpler process compared to the conventional method.","PeriodicalId":45921,"journal":{"name":"Journal of Advanced Computational Intelligence and Intelligent Informatics","volume":"1 1","pages":"54-63"},"PeriodicalIF":0.7,"publicationDate":"2023-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89687225","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}
It is becoming increasingly attractive to execute workflows in the cloud, as the cloud environment enables scientific applications to utilize elastic computing resources on demand. However, despite being a key to efficiently managing application execution in the cloud, traditional workflow scheduling algorithms face significant challenges in the cloud environment. The gradient-based optimizer (GBO) is a newly proposed evolutionary algorithm with a search engine based on the Newton’s method. It employs a set of vectors to search in the solution space. This study designs a gradient-based scheduler by using GBO for workflow scheduling to minimize the usage costs of workflows under given deadline constraints. Extensive experiments are conducted on well-known scientific workflows of different sizes and types using WorkflowSim. The experimental results show that the proposed scheduling algorithm outperforms five other state-of-the-art algorithms in terms of both the constraint satisfiability and cost optimization, thereby verifying its advantages in addressing workflow scheduling problems.
{"title":"Gradient-Based Scheduler for Scientific Workflows in Cloud Computing","authors":"Danjing Wang, Huifang Li, Youwei Zhang, Baihai Zhang","doi":"10.20965/jaciii.2023.p0064","DOIUrl":"https://doi.org/10.20965/jaciii.2023.p0064","url":null,"abstract":"It is becoming increasingly attractive to execute workflows in the cloud, as the cloud environment enables scientific applications to utilize elastic computing resources on demand. However, despite being a key to efficiently managing application execution in the cloud, traditional workflow scheduling algorithms face significant challenges in the cloud environment. The gradient-based optimizer (GBO) is a newly proposed evolutionary algorithm with a search engine based on the Newton’s method. It employs a set of vectors to search in the solution space. This study designs a gradient-based scheduler by using GBO for workflow scheduling to minimize the usage costs of workflows under given deadline constraints. Extensive experiments are conducted on well-known scientific workflows of different sizes and types using WorkflowSim. The experimental results show that the proposed scheduling algorithm outperforms five other state-of-the-art algorithms in terms of both the constraint satisfiability and cost optimization, thereby verifying its advantages in addressing workflow scheduling problems.","PeriodicalId":45921,"journal":{"name":"Journal of Advanced Computational Intelligence and Intelligent Informatics","volume":"65 1","pages":"64-73"},"PeriodicalIF":0.7,"publicationDate":"2023-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87738399","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 : 2023-01-20DOI: 10.20965/jaciii.2023.p0003
R. Relano, Kate G. Francisco, Ronnie S. Concepcion, Mike Louie C. Enriquez, Jonah Jahara G. Baun, Adrian Genevie G. Janairo, R. R. Vicerra, A. Bandala, E. Dadios
Land surveying has been one of the core operations in performing underground imaging. It is known that dynamic and continuous resistivity readings were employed through this technique using the array of capacitive electrodes being towed with a light vehicle. However, the main challenge in doing subsurface surveying is the change in speed of the system when there are inevitable obstacles and sloping road surfaces. To address it, this study will develop prediction models using different computational intelligence such as multigene symbolic regression genetic programming (MSRGP), regression-based decision tree (RTree), and feed forward neural network (FFNN) that will result in a smart speed controller system that can maintain the constant speed of the towed subterranean system. The best performing prediction model will be considered as the SpeedX. The expected output is a correction factor that will signal the speed controller in slow down or inclined plane road environment to maintain a constant speed of 1.6667 m/s for avoidance of data distortion on land surveying. Thus, the MSEs for MSRGP, FFNN, and RTree are 0.00163, 0.00178, and 0.00240, respectively. This results in MSRGP as the best performing model and was considered as the SpeedX model. Other evaluation metrics were employed such as the MAE and R2 which signify the advantage of SpeedX. Furthermore, the comparison between the CI-controlled and uncontrolled towed subterranean imaging trailer system, as well as its advantages clearly highlight the advantage of embedded SpeedX in the system.
{"title":"SpeedX: Smart Speed Controller Model of Towed Subterranean Imaging System for Resistivity Data Distortion Reduction Using Computational Intelligence","authors":"R. Relano, Kate G. Francisco, Ronnie S. Concepcion, Mike Louie C. Enriquez, Jonah Jahara G. Baun, Adrian Genevie G. Janairo, R. R. Vicerra, A. Bandala, E. Dadios","doi":"10.20965/jaciii.2023.p0003","DOIUrl":"https://doi.org/10.20965/jaciii.2023.p0003","url":null,"abstract":"Land surveying has been one of the core operations in performing underground imaging. It is known that dynamic and continuous resistivity readings were employed through this technique using the array of capacitive electrodes being towed with a light vehicle. However, the main challenge in doing subsurface surveying is the change in speed of the system when there are inevitable obstacles and sloping road surfaces. To address it, this study will develop prediction models using different computational intelligence such as multigene symbolic regression genetic programming (MSRGP), regression-based decision tree (RTree), and feed forward neural network (FFNN) that will result in a smart speed controller system that can maintain the constant speed of the towed subterranean system. The best performing prediction model will be considered as the SpeedX. The expected output is a correction factor that will signal the speed controller in slow down or inclined plane road environment to maintain a constant speed of 1.6667 m/s for avoidance of data distortion on land surveying. Thus, the MSEs for MSRGP, FFNN, and RTree are 0.00163, 0.00178, and 0.00240, respectively. This results in MSRGP as the best performing model and was considered as the SpeedX model. Other evaluation metrics were employed such as the MAE and R2 which signify the advantage of SpeedX. Furthermore, the comparison between the CI-controlled and uncontrolled towed subterranean imaging trailer system, as well as its advantages clearly highlight the advantage of embedded SpeedX in the system.","PeriodicalId":45921,"journal":{"name":"Journal of Advanced Computational Intelligence and Intelligent Informatics","volume":"36 1","pages":"3-11"},"PeriodicalIF":0.7,"publicationDate":"2023-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91164932","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 : 2023-01-20DOI: 10.20965/jaciii.2023.p0012
J. A. D. Leon, Ronnie S. Concepcion, R. Billones, Jonah Jahara G. Baun, Jose Miguel F. Custodio, R. R. Vicerra, A. Bandala, E. Dadios
Electrical resistivity tomography (ERT) has been seen as an appropriate instrument in several works to monitor and aid in the control of seawater intrusion (SWI) in coastal groundwater systems. This study seeks to discuss the synthesis of a digital twin that couples information between the physical space through ERT as a monitoring sensor and the digital space using SWI simulations to accurately model the behavior of SWI in the present and future settings. To showcase the concept, a Python-based simulation was presented that shows (a) the joint forward modeling-simulation scheme for calculating expected ERT apparent resistivity values from simulated SWI and (b) the calibration of the digital coastal aquifer system through genetic algorithm to accurately match the outputs of the SWI simulations with the ERT measurements.
{"title":"Digital Twin Concept Utilizing Electrical Resistivity Tomography for Monitoring Seawater Intrusion","authors":"J. A. D. Leon, Ronnie S. Concepcion, R. Billones, Jonah Jahara G. Baun, Jose Miguel F. Custodio, R. R. Vicerra, A. Bandala, E. Dadios","doi":"10.20965/jaciii.2023.p0012","DOIUrl":"https://doi.org/10.20965/jaciii.2023.p0012","url":null,"abstract":"Electrical resistivity tomography (ERT) has been seen as an appropriate instrument in several works to monitor and aid in the control of seawater intrusion (SWI) in coastal groundwater systems. This study seeks to discuss the synthesis of a digital twin that couples information between the physical space through ERT as a monitoring sensor and the digital space using SWI simulations to accurately model the behavior of SWI in the present and future settings. To showcase the concept, a Python-based simulation was presented that shows (a) the joint forward modeling-simulation scheme for calculating expected ERT apparent resistivity values from simulated SWI and (b) the calibration of the digital coastal aquifer system through genetic algorithm to accurately match the outputs of the SWI simulations with the ERT measurements.","PeriodicalId":45921,"journal":{"name":"Journal of Advanced Computational Intelligence and Intelligent Informatics","volume":"20 1","pages":"12-18"},"PeriodicalIF":0.7,"publicationDate":"2023-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83497663","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：