Pub Date : 2021-10-12DOI: 10.23919/ICCAS52745.2021.9649878
S. Kim, Jong Hyun Lee, Dong Hun Wang, Insoo Lee
Currently, lithium-ion batteries (a type of secondary battery) are used as the primary sources of power in many applications due to their low energy loss as a result of their high energy density and low self-discharge rate, and their ability to store energy for a long time. However, due to the frequent charging and discharging of such batteries, overcharging is inevitable. This can cause system shutdowns, accidents, or property damage due to explosions. Therefore, it is necessary to accurately predict the state of charge (SOC) of batteries for stable and efficient usage. Hence, in this paper, we propose a SOC estimation method using a vehicle driving simulator. After manufacturing the simulator to perform the battery discharge experiment, voltage, current, and discharge-time data were collected. Using the collected data as input parameters for an RNN-based LSTM, we estimated the SOC of the battery and compared the errors to. We then used the developed LSTM surrogate model to conduct discharge experiments and simultaneously estimate the SOC in real-time.
{"title":"LSTM-Based Real-Time SOC Estimation of Lithium-Ion Batteries Using a Vehicle Driving Simulator","authors":"S. Kim, Jong Hyun Lee, Dong Hun Wang, Insoo Lee","doi":"10.23919/ICCAS52745.2021.9649878","DOIUrl":"https://doi.org/10.23919/ICCAS52745.2021.9649878","url":null,"abstract":"Currently, lithium-ion batteries (a type of secondary battery) are used as the primary sources of power in many applications due to their low energy loss as a result of their high energy density and low self-discharge rate, and their ability to store energy for a long time. However, due to the frequent charging and discharging of such batteries, overcharging is inevitable. This can cause system shutdowns, accidents, or property damage due to explosions. Therefore, it is necessary to accurately predict the state of charge (SOC) of batteries for stable and efficient usage. Hence, in this paper, we propose a SOC estimation method using a vehicle driving simulator. After manufacturing the simulator to perform the battery discharge experiment, voltage, current, and discharge-time data were collected. Using the collected data as input parameters for an RNN-based LSTM, we estimated the SOC of the battery and compared the errors to. We then used the developed LSTM surrogate model to conduct discharge experiments and simultaneously estimate the SOC in real-time.","PeriodicalId":411064,"journal":{"name":"2021 21st International Conference on Control, Automation and Systems (ICCAS)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115237679","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 : 2021-10-12DOI: 10.23919/ICCAS52745.2021.9649930
Kota Takahashi, H. Madokoro, Satoshi Yamamoto, Yoshiteru Nishimura, Stephanie Nix, Hanwool Woo, T. K. Saito, Kazuhito Sato
Recognition accuracy obtained using deep learning drops precipitously when the training data are insufficient. This paper presents a data-expansion method for training of the transfer learning source domain. Using expanding images generated from weights on a category map as source data, we compared accuracies obtained from five derivative models and our previously reported method. Moreover, we obtained the result of domain adaptation between actual images and synthetic images using weights obtained during transfer learning. Based on those results, we verify whether the amount of training data can be expanded quantitatively and qualitatively. Experiment results obtained from two open benchmark datasets and our original benchmark dataset demonstrated that our proposed method outperforms the previous method under a guarantee of sufficient accuracy for the synthetic images.
{"title":"Domain Adaptation for Agricultural Image Recognition and Segmentation Using Category Maps","authors":"Kota Takahashi, H. Madokoro, Satoshi Yamamoto, Yoshiteru Nishimura, Stephanie Nix, Hanwool Woo, T. K. Saito, Kazuhito Sato","doi":"10.23919/ICCAS52745.2021.9649930","DOIUrl":"https://doi.org/10.23919/ICCAS52745.2021.9649930","url":null,"abstract":"Recognition accuracy obtained using deep learning drops precipitously when the training data are insufficient. This paper presents a data-expansion method for training of the transfer learning source domain. Using expanding images generated from weights on a category map as source data, we compared accuracies obtained from five derivative models and our previously reported method. Moreover, we obtained the result of domain adaptation between actual images and synthetic images using weights obtained during transfer learning. Based on those results, we verify whether the amount of training data can be expanded quantitatively and qualitatively. Experiment results obtained from two open benchmark datasets and our original benchmark dataset demonstrated that our proposed method outperforms the previous method under a guarantee of sufficient accuracy for the synthetic images.","PeriodicalId":411064,"journal":{"name":"2021 21st International Conference on Control, Automation and Systems (ICCAS)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116665977","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 : 2021-10-12DOI: 10.23919/ICCAS52745.2021.9650054
Youngbin Song, Minhwan Seo, Shina Park, S. W. Kim
Initial parameter variances between cells in battery packs occur in a manufacturing process. Furthermore, this difference is intensified as the pack is being used, resulting in differences in capacity and the state of charge (SOC) between cells. Cell inconsistencies decrease the energy efficiency, and low-capacity cells in packs can occur an internal short circuit (ISC) fault which causes a thermal runaway in severe cases. However, the ISC may be misdiagnosed as cell inconsistencies and vice versa because the impacts of cell inconsistencies and the ISC are similar in particular charge/discharge. In this paper, a model-based cell inconsistency classification method is proposed. The equivalent circuit model of the fresh cell is used as a reference model, making it possible to save efforts in constructing parameter look-up tables for various degrees of aging. In addition, we use the SOC difference feature that can clearly distinguish the effects of inconsistencies and ISC using the reference SOC calculated by the nominal capacity. The proposed method was verified in simulation for various types and degrees of cell inconsistencies and ISC, and accurately identified inconsistent cells and ISC cells, thereby leading to efficient energy use and early detection of the ISC fault.
{"title":"Cell Inconsistency Classification for Lithium-Ion Battery Packs Considering Internal Short Circuit Fault","authors":"Youngbin Song, Minhwan Seo, Shina Park, S. W. Kim","doi":"10.23919/ICCAS52745.2021.9650054","DOIUrl":"https://doi.org/10.23919/ICCAS52745.2021.9650054","url":null,"abstract":"Initial parameter variances between cells in battery packs occur in a manufacturing process. Furthermore, this difference is intensified as the pack is being used, resulting in differences in capacity and the state of charge (SOC) between cells. Cell inconsistencies decrease the energy efficiency, and low-capacity cells in packs can occur an internal short circuit (ISC) fault which causes a thermal runaway in severe cases. However, the ISC may be misdiagnosed as cell inconsistencies and vice versa because the impacts of cell inconsistencies and the ISC are similar in particular charge/discharge. In this paper, a model-based cell inconsistency classification method is proposed. The equivalent circuit model of the fresh cell is used as a reference model, making it possible to save efforts in constructing parameter look-up tables for various degrees of aging. In addition, we use the SOC difference feature that can clearly distinguish the effects of inconsistencies and ISC using the reference SOC calculated by the nominal capacity. The proposed method was verified in simulation for various types and degrees of cell inconsistencies and ISC, and accurately identified inconsistent cells and ISC cells, thereby leading to efficient energy use and early detection of the ISC fault.","PeriodicalId":411064,"journal":{"name":"2021 21st International Conference on Control, Automation and Systems (ICCAS)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117137107","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 : 2021-10-12DOI: 10.23919/ICCAS52745.2021.9649796
K. Shimizu, T. Hirogaki, E. Aoyama
With the recent advances in industrial robot the motion control, there is a demand for a method that can easily measure the accuracy of the synchronous control of dual-arm selective compliance assembly robot arm (SCARA) robots. Therefore, in this study, a measurement method was developed in which a ball rolling motion is created in a circular orbit on the work plate and the rolling motion error with respect to the reference circle is utilized. An experiment was conducted on a teaching method for controlling the rolling motion of the ball at a position away from the center of the robot on the work plate. It was found that an error occurs in the peripheral speed in the rotation of the plate. When the work plate is grasped and the movement is taught, the program using the master-slave method was the primary influence on the motion error, confirming the possibility of correction.
{"title":"Working plate operation for graspless handling technology with an industrial dual-arm SCARA robot","authors":"K. Shimizu, T. Hirogaki, E. Aoyama","doi":"10.23919/ICCAS52745.2021.9649796","DOIUrl":"https://doi.org/10.23919/ICCAS52745.2021.9649796","url":null,"abstract":"With the recent advances in industrial robot the motion control, there is a demand for a method that can easily measure the accuracy of the synchronous control of dual-arm selective compliance assembly robot arm (SCARA) robots. Therefore, in this study, a measurement method was developed in which a ball rolling motion is created in a circular orbit on the work plate and the rolling motion error with respect to the reference circle is utilized. An experiment was conducted on a teaching method for controlling the rolling motion of the ball at a position away from the center of the robot on the work plate. It was found that an error occurs in the peripheral speed in the rotation of the plate. When the work plate is grasped and the movement is taught, the program using the master-slave method was the primary influence on the motion error, confirming the possibility of correction.","PeriodicalId":411064,"journal":{"name":"2021 21st International Conference on Control, Automation and Systems (ICCAS)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121063547","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 : 2021-10-12DOI: 10.23919/ICCAS52745.2021.9650037
Jae-Hyeon Park, D. Chang
With the advancement of neural network technology, many researchers are trying to find a clever way to apply neural network to a fault detection and isolation area for satisfactory and safer operations of the system. Some researchers detect system faults by combining a concrete model of the system with neural network, generating residuals by neural network, or training neural network with specific sensor signals of the system. In this article, we make a fault detection and isolation neural network algorithm that uses only inherent sensor measurements and control inputs of the system. This algorithm does not need a model of the system, residual generations, or additional sensors. We obtain sensor measurements and control inputs in a discrete-time manner, cut signals with a sliding window approach, and label data with one-hot vectors representing a normal or fault classes. We train our neural network model with the labeled training data. We give 2 neural network models: a stacked long short-term memory neural network and a multilayer perceptron. We test our algorithm with the quadrotor fault simulation and the real experiment. Our algorithm gives nice performance on a fault detection and isolation of the quadrotor.
{"title":"Data-driven fault detection and isolation of system with only state measurements and control inputs using neural networks","authors":"Jae-Hyeon Park, D. Chang","doi":"10.23919/ICCAS52745.2021.9650037","DOIUrl":"https://doi.org/10.23919/ICCAS52745.2021.9650037","url":null,"abstract":"With the advancement of neural network technology, many researchers are trying to find a clever way to apply neural network to a fault detection and isolation area for satisfactory and safer operations of the system. Some researchers detect system faults by combining a concrete model of the system with neural network, generating residuals by neural network, or training neural network with specific sensor signals of the system. In this article, we make a fault detection and isolation neural network algorithm that uses only inherent sensor measurements and control inputs of the system. This algorithm does not need a model of the system, residual generations, or additional sensors. We obtain sensor measurements and control inputs in a discrete-time manner, cut signals with a sliding window approach, and label data with one-hot vectors representing a normal or fault classes. We train our neural network model with the labeled training data. We give 2 neural network models: a stacked long short-term memory neural network and a multilayer perceptron. We test our algorithm with the quadrotor fault simulation and the real experiment. Our algorithm gives nice performance on a fault detection and isolation of the quadrotor.","PeriodicalId":411064,"journal":{"name":"2021 21st International Conference on Control, Automation and Systems (ICCAS)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124836072","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 : 2021-10-12DOI: 10.23919/ICCAS52745.2021.9650015
Yew Ji Hao, Lee Koon Teck, Chua Ying Xiang, Enoch Jeevanraj, S. Srigrarom
This paper aims to introduce the method of detection of high-speed drones using both Single Shot Detector (SSD) and YOLOv3 (You Only Look Once)v3. After conducting experiments and obtaining footage of the fast-flying drones, the software and algorithms are being put to the test. In a motion detector, there are 3 main fundamentals - unmanned aerial vehicle (UAV) detection, UAV identification and tracking of the UAV, which will be introduced as a preliminary UAV detection system to spark of the use of other more advanced image recognition based detector. The alternative of using SSD and YOLOv3 will be the main discussion to target high-speed drones.
本文旨在介绍使用Single Shot Detector (SSD)和YOLOv3 (You Only Look Once)v3对高速无人机进行检测的方法。在进行实验并获得快速飞行的无人机的镜头后，软件和算法正在进行测试。在一个运动探测器中，有3个主要的基础——无人机(UAV)探测、无人机识别和无人机跟踪，这将作为无人机探测系统的初步介绍，以激发其他更先进的基于图像识别的探测器的使用。针对高速无人机，将主要讨论使用SSD和YOLOv3的替代方案。
{"title":"Fast Drone Detection using SSD and YoloV3","authors":"Yew Ji Hao, Lee Koon Teck, Chua Ying Xiang, Enoch Jeevanraj, S. Srigrarom","doi":"10.23919/ICCAS52745.2021.9650015","DOIUrl":"https://doi.org/10.23919/ICCAS52745.2021.9650015","url":null,"abstract":"This paper aims to introduce the method of detection of high-speed drones using both Single Shot Detector (SSD) and YOLOv3 (You Only Look Once)v3. After conducting experiments and obtaining footage of the fast-flying drones, the software and algorithms are being put to the test. In a motion detector, there are 3 main fundamentals - unmanned aerial vehicle (UAV) detection, UAV identification and tracking of the UAV, which will be introduced as a preliminary UAV detection system to spark of the use of other more advanced image recognition based detector. The alternative of using SSD and YOLOv3 will be the main discussion to target high-speed drones.","PeriodicalId":411064,"journal":{"name":"2021 21st International Conference on Control, Automation and Systems (ICCAS)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125180725","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 : 2021-10-12DOI: 10.23919/ICCAS52745.2021.9649775
Bhivraj Suthar, Seul Jung
Humans hand has a complex functionality to achieve stable and reliable gripping of irregular-shaped objects. This paper aims to design a three-finger anthropomorphic gripper and endow the designed hand with natural grasping functions. We introduce a passively adjustable flexion finger mechanism that can change a flexion finger angle according to the irregular-shaped objects. Each finger of the gripper has a torsional spring and has placed co-axially, connected in series on an equilateral triangular palm. The required stiffness of the flexion angle and object safety condition is analyzed. The variation of the flexion joint stiffness for the minimum to maximum flexion angle is evaluated. Practical experiments using a wide range of obj ects under different grasping scenarios are performed to demonstrate the grasping capability of the integrated gripper.
{"title":"Design and Development of a Co-axial Passive Flexion Mechanism-based Gripper for Irregular Objects","authors":"Bhivraj Suthar, Seul Jung","doi":"10.23919/ICCAS52745.2021.9649775","DOIUrl":"https://doi.org/10.23919/ICCAS52745.2021.9649775","url":null,"abstract":"Humans hand has a complex functionality to achieve stable and reliable gripping of irregular-shaped objects. This paper aims to design a three-finger anthropomorphic gripper and endow the designed hand with natural grasping functions. We introduce a passively adjustable flexion finger mechanism that can change a flexion finger angle according to the irregular-shaped objects. Each finger of the gripper has a torsional spring and has placed co-axially, connected in series on an equilateral triangular palm. The required stiffness of the flexion angle and object safety condition is analyzed. The variation of the flexion joint stiffness for the minimum to maximum flexion angle is evaluated. Practical experiments using a wide range of obj ects under different grasping scenarios are performed to demonstrate the grasping capability of the integrated gripper.","PeriodicalId":411064,"journal":{"name":"2021 21st International Conference on Control, Automation and Systems (ICCAS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125889511","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 : 2021-10-12DOI: 10.23919/ICCAS52745.2021.9650043
Yi-Hsin Lin, Wei-Yu Chiu
Striking a balance between power supply and demand is the most imperative target for any electricity grid system. In order to address variability of renewable energy in the modern grid, a robust and elastic balancing scheme is required. Conventional model-based approaches can suffer from great performance degradation given the uncertainty induced by the renewable energy. As such, this study explores a model-free approach by proposing a reinforcement learning based pricing scheme that balances the power supply and demand. A price signal is considered as the control signal for the balance management. Case studies involving different market parameters and different time resolutions were conducted to show the effectiveness of the proposed methodology.
{"title":"Reinforcement Learning Based Electricity Price Controller in Smart Grids","authors":"Yi-Hsin Lin, Wei-Yu Chiu","doi":"10.23919/ICCAS52745.2021.9650043","DOIUrl":"https://doi.org/10.23919/ICCAS52745.2021.9650043","url":null,"abstract":"Striking a balance between power supply and demand is the most imperative target for any electricity grid system. In order to address variability of renewable energy in the modern grid, a robust and elastic balancing scheme is required. Conventional model-based approaches can suffer from great performance degradation given the uncertainty induced by the renewable energy. As such, this study explores a model-free approach by proposing a reinforcement learning based pricing scheme that balances the power supply and demand. A price signal is considered as the control signal for the balance management. Case studies involving different market parameters and different time resolutions were conducted to show the effectiveness of the proposed methodology.","PeriodicalId":411064,"journal":{"name":"2021 21st International Conference on Control, Automation and Systems (ICCAS)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126005838","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 : 2021-10-12DOI: 10.23919/ICCAS52745.2021.9649908
Hyungtai Kim, Y. Kim, Seung-jong Kim, Munsik Choi
In rehabilitation of the patients after stroke, gait types are important to know the characteristics of the patient. To know gait types, a systematic methodology for direct measurement and interpretation of gait motion are required. In this study, the patient's kinetic data were collected eight times over six months after onset using motion capture equipment. Features for gait type classification were extracted from time series gait cycle data and used for machine learning analysis. We utilized the simultaneous clustering and classification method to determine gait types that ensure classification performance. The optimal number of gait groups was four, which shows 0.1504 and 0.9142 in silhouette score and F1 score. We present a novel work to find the gait groups of patients after stroke, and showed the potential for use in the rehabilitation field.
{"title":"Gait Clustering Analysis in Patients after Stroke using Gait Kinematics Data","authors":"Hyungtai Kim, Y. Kim, Seung-jong Kim, Munsik Choi","doi":"10.23919/ICCAS52745.2021.9649908","DOIUrl":"https://doi.org/10.23919/ICCAS52745.2021.9649908","url":null,"abstract":"In rehabilitation of the patients after stroke, gait types are important to know the characteristics of the patient. To know gait types, a systematic methodology for direct measurement and interpretation of gait motion are required. In this study, the patient's kinetic data were collected eight times over six months after onset using motion capture equipment. Features for gait type classification were extracted from time series gait cycle data and used for machine learning analysis. We utilized the simultaneous clustering and classification method to determine gait types that ensure classification performance. The optimal number of gait groups was four, which shows 0.1504 and 0.9142 in silhouette score and F1 score. We present a novel work to find the gait groups of patients after stroke, and showed the potential for use in the rehabilitation field.","PeriodicalId":411064,"journal":{"name":"2021 21st International Conference on Control, Automation and Systems (ICCAS)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123767423","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 : 2021-10-12DOI: 10.23919/ICCAS52745.2021.9649805
T. Cao, Khoa Van Duong
The process of classifying many types of food from images is an exciting field involving various applications. Especially in tourist, Vietnamese food classification connects us across our cultures and generations. Food classification is not easy, even with people. The reason is the food's extreme diversity between dishes and in the middle variations of the dish. So some traditional approaches with hand-crafted features had been used for food recognition. However, evaluation in deep learning and convolutional neural networks achieved higher accuracy compared to the traditional methods. We propose a new dataset called TypicalVietnameseFoodNet and a proposed model with the best performance for our dataset, called the TypicalVietnameseFood model. Our proposed approach achieves 94.84% on the test set.
{"title":"TypicalVietnameseFoodNet: A Vietnamese Food Image Dataset For Vietnamese Food Classifications","authors":"T. Cao, Khoa Van Duong","doi":"10.23919/ICCAS52745.2021.9649805","DOIUrl":"https://doi.org/10.23919/ICCAS52745.2021.9649805","url":null,"abstract":"The process of classifying many types of food from images is an exciting field involving various applications. Especially in tourist, Vietnamese food classification connects us across our cultures and generations. Food classification is not easy, even with people. The reason is the food's extreme diversity between dishes and in the middle variations of the dish. So some traditional approaches with hand-crafted features had been used for food recognition. However, evaluation in deep learning and convolutional neural networks achieved higher accuracy compared to the traditional methods. We propose a new dataset called TypicalVietnameseFoodNet and a proposed model with the best performance for our dataset, called the TypicalVietnameseFood model. Our proposed approach achieves 94.84% on the test set.","PeriodicalId":411064,"journal":{"name":"2021 21st International Conference on Control, Automation and Systems (ICCAS)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114916605","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：