2022 15th International Conference on Human System Interaction (HSI)最新文献_第2页

Joint Space Based Force Sensorless Bilateral Control with BP Neural Network Gravity Compensation for 6-PSS Parallel Actuator 基于联合空间无力传感器的BP神经网络重力补偿6-PSS并联作动器双边控制

2022 15th International Conference on Human System Interaction (HSI) Pub Date : 2022-07-28 DOI: 10.1109/HSI55341.2022.9869448

Jiangtao Zheng, Yutang Wang, Cheng-rong Lu, Dapeng Tian

引用次数: 0

Local Detail Enhancement Network for CNV Typing in OCT Images 局部细节增强网络在OCT图像CNV分型中的应用

2022 15th International Conference on Human System Interaction (HSI) Pub Date : 2022-07-28 DOI: 10.1109/HSI55341.2022.9869483

Chuanzhen Xu, Xiaoming Xi, Xiao Yang, Liangyun Sun, Lingzhao Meng, Xiushan Nie

{"title":"Local Detail Enhancement Network for CNV Typing in OCT Images","authors":"Chuanzhen Xu, Xiaoming Xi, Xiao Yang, Liangyun Sun, Lingzhao Meng, Xiushan Nie","doi":"10.1109/HSI55341.2022.9869483","DOIUrl":"https://doi.org/10.1109/HSI55341.2022.9869483","url":null,"abstract":"Choroidal neovascularization (CNV) is one of the severe eye disease. The severe results will cause of loss of acuity, scotomata, and distortion of vision. Automatic and accurate classification of CNV with optical coherence tomography (OCT) images can assist doctors in treatment. However, the existing methods ignore the fact that semantic feature maps, used for classification, lose much feature detail information. Therefore, we proposed a local detail enhancement network for CNV classification, which includes both progressive training mode and local detail enhancement (LDE) module. With the progressive training mode, the learned features fuse shallow and stable fine-grained information with high-level semantic information, which promote the diversity of the learned features. In LDE module, the detail feature learn (DFL) module is introduced to learn the underlying detail information and embed it into the semantic feature map. The semantic feature map with detail information is propitious to capture the subtle discrepancy between different CNV types and promote the classification performance. Sufficient experiments are performed on our self-build CNV dataset. Our method excelled existing methods and in evaluation indicators ACC, AUC, SEN, and SPE are 92.3%, 87.1%, 91.5%, and 90.9%.","PeriodicalId":282607,"journal":{"name":"2022 15th International Conference on Human System Interaction (HSI)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134359167","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}

引用次数: 0

Towards Safety in Open-field Agricultural Robotic Applications: A Method for Human Risk Assessment using Classifiers 面向露天农业机器人应用的安全:一种基于分类器的人类风险评估方法

2022 15th International Conference on Human System Interaction (HSI) Pub Date : 2022-07-28 DOI: 10.1109/HSI55341.2022.9869472

José C. Mayoral, Lars Grimstad, P. From, Grzegorz Cielniak

引用次数: 2

ScSer: Supervised Contrastive Learning for Speech Emotion Recognition using Transformers 语音情感识别的监督对比学习

2022 15th International Conference on Human System Interaction (HSI) Pub Date : 2022-07-28 DOI: 10.1109/HSI55341.2022.9869453

Varun Sai Alaparthi, Tejeswara Reddy Pasam, Deepak Abhiram Inagandla, J. Prakash, P. Singh

引用次数: 7

Optimal placement of IMU sensor for the detection of children activity IMU传感器的最佳位置，用于检测儿童活动

2022 15th International Conference on Human System Interaction (HSI) Pub Date : 2022-07-28 DOI: 10.1109/HSI55341.2022.9869442

M. Madej, J. Rumiński

引用次数: 0

Cooee: An Artificial Intelligence Chatbot for Complex Energy Environments 库伊:复杂能源环境下的人工智能聊天机器人

2022 15th International Conference on Human System Interaction (HSI) Pub Date : 2022-07-28 DOI: 10.1109/HSI55341.2022.9869464

Gihan Gamage, Nishan Mills, Prabod Rathnayaka, Andrew Jennings, D. Alahakoon

{"title":"Cooee: An Artificial Intelligence Chatbot for Complex Energy Environments","authors":"Gihan Gamage, Nishan Mills, Prabod Rathnayaka, Andrew Jennings, D. Alahakoon","doi":"10.1109/HSI55341.2022.9869464","DOIUrl":"https://doi.org/10.1109/HSI55341.2022.9869464","url":null,"abstract":"Contemporary energy platforms are leveraging advanced data management and Artificial Intelligence (AI) capabilities in response to the increasing complexity of energy systems and grids. Despite these advances, it is a non-trivial and challenging task to support the decision-making needs of the human operators of such complex energy-related implementations. Conversational agents or chatbots are a potential emerging technology that can be utilized to address this challenge. Although there is a large body of literature on chatbots in general, they are not robust as they rely on predefined conversational pathways that are inadequate to efficiently address the complexities of dynamic data spaces in energy platforms. The capability of generating answers in real-time by communicating with the dynamic dataspace is crucial as energy management decisions are real-time and time sensitive. In this paper, we present the design and development of Cooee, a chatbot for conversational engagement with the dynamic data spaces of complex energy environments. Cooee leverages state-of-art language models along with rule-based language processing methods for a conversational interaction with dynamic data spaces, which consequently supports and enables decision-making by human experts. We have developed Cooee as a standalone application and then integrated into a real-world energy AI platform deployed within a multi-campus tertiary education institution setting. Cooee was empirically evaluated in this setting and compared with several state-of-the-art Q&A approaches.","PeriodicalId":282607,"journal":{"name":"2022 15th International Conference on Human System Interaction (HSI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129132333","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}

引用次数: 1

A Hybrid Tree-Based Ensemble Learning Model for Day-Ahead Peak Load Forecasting 日前高峰负荷预测的混合树集成学习模型

2022 15th International Conference on Human System Interaction (HSI) Pub Date : 2022-07-28 DOI: 10.1109/HSI55341.2022.9869440

Jihoon Moon, Sungwoo Park, Eenjun Hwang, Seungmin Rho

{"title":"A Hybrid Tree-Based Ensemble Learning Model for Day-Ahead Peak Load Forecasting","authors":"Jihoon Moon, Sungwoo Park, Eenjun Hwang, Seungmin Rho","doi":"10.1109/HSI55341.2022.9869440","DOIUrl":"https://doi.org/10.1109/HSI55341.2022.9869440","url":null,"abstract":"Daily peak load forecasting (DPLF) is critical in smart grid applications for security analysis, unit commitment, and scheduling of outages and fuel supplies. Although excellent single machine learning methods using tree-based ensemble learning or deep learning have shown satisfactory performance for DPLF, there is still room for improvement. This study proposes a hybrid tree-based ensemble learning model, called HYTREM, for robust DPLF. We first collected two commercial buildings’ energy consumption data from publicly available datasets. We then performed data preprocessing, such as input variable configuration, for the HYTREM modeling. We divided both datasets into training and test sets and generated the prediction values of several tree-based ensemble learning models, such as gradient boosting machine, extreme gradient boosting, Cubist, and random forest (RF), for each set as novel input variables. We reconstructed datasets using the Boruta algorithm to select all the relevant features and built an online RF model trained on these datasets using time-series cross-validation for day-ahead DPLF. The experimental results showed that the HYTREM performed a better performance than tree-based ensemble and deep learning methods in building-level DPLF in terms of the mean absolute percentage error and normalized root mean square error.","PeriodicalId":282607,"journal":{"name":"2022 15th International Conference on Human System Interaction (HSI)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122340559","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}

引用次数: 1

Identification of Wrist Elastic Moment in Healthy Subjects for Control by Functional Electrical Stimulation 功能电刺激控制下健康受试者手腕弹性矩的识别

2022 15th International Conference on Human System Interaction (HSI) Pub Date : 2022-07-28 DOI: 10.1109/HSI55341.2022.9869499

Akinori Shimomura, S. Katsura

{"title":"Identification of Wrist Elastic Moment in Healthy Subjects for Control by Functional Electrical Stimulation","authors":"Akinori Shimomura, S. Katsura","doi":"10.1109/HSI55341.2022.9869499","DOIUrl":"https://doi.org/10.1109/HSI55341.2022.9869499","url":null,"abstract":"This study is for a functional electrical stimulation. Functional electrical stimulation is a system that uses electrical stimulation to generate specific joint movements. If the nerve is alive, it is possible to promote muscle contraction from the outside by electrical stimulation. If arbitrary movements are generated, to build a control system is necessary for functional electrical stimulation. A model of muscle contraction is needed for control using functional electrical stimulation. It is assumed that the control accuracy will be improved by incorporating the necessary information into the model based on human physiology. In this study, the angle-dependent term in the reaction during muscle contraction is focused as one of the necessary information characteristics. As the angle changes, the muscle condition changes, resulting in responses such as stretch reflexes and repulsive forces. This is important information because if such a response is not taken into consideration, an error will occur between the model and the actual output. In the conventional research, the angle-dependent term was identified by focusing on the lower limbs, and the parameter fitting was realized by using the double exponential function. In this study, the knowledge of the previous study is applied to the upper limbs to build a model. Since the response is different between the upper limbs and the lower limbs, the movements are not always the same. Therefore, verification was performed. From the experimental results, the model based on the double exponential function was a model that took individual differences into consideration even in the upper limbs.","PeriodicalId":282607,"journal":{"name":"2022 15th International Conference on Human System Interaction (HSI)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126942079","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}

引用次数: 0

Environmental and Social monitoring in existing building structures – results of a case study within several historical buildings 现有建筑结构的环境和社会监测——若干历史建筑案例研究的结果

2022 15th International Conference on Human System Interaction (HSI) Pub Date : 2022-07-28 DOI: 10.1109/HSI55341.2022.9869468

A. Redlein, L. Thrainer

{"title":"Environmental and Social monitoring in existing building structures – results of a case study within several historical buildings","authors":"A. Redlein, L. Thrainer","doi":"10.1109/HSI55341.2022.9869468","DOIUrl":"https://doi.org/10.1109/HSI55341.2022.9869468","url":null,"abstract":"One year ago the European Union defined the Environmental, Social and Governance (ESG) directive to foster investment in sustainability. This directive asks for energy monitoring / optimization and actions to foster human well-being. To cover that demand especially historical buildings in Europe need solutions in these respects, to be more energy efficient and at the same time to safeguard the well-being of its users. These buildings have almost no building automation installed. So, the basic prerequisite to identify high energy consumers valid usage data is not available. For that, the following research questions have to be answered: Which data is necessary to reach both, the environmental and social goals? How can data be captured in a valid and efficient way? How can data be made available to optimize energy usage and well-being? To answer these questions, a literature research was executed to define the relevant parameters. Firstly, based on the results of that step and a previous mixed-methods-research project, the relevant tools and the IT architecture were defined. Secondly, based on the ESG-reporting demands, data structure and relevant building structures were defined. Thirdly, a case study was conducted, which is based on the previously defined IT architecture, using IoT measuring devices, two different databases and two analytic tools. As a result, this paper presents the final decision on the database and the analytics tool, based on the users’ interaction and feedback of the case study.","PeriodicalId":282607,"journal":{"name":"2022 15th International Conference on Human System Interaction (HSI)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128182458","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}

引用次数: 1

Person Search via Background and Foreground Contrastive Learning 基于背景与前景对比学习的人物搜索

2022 15th International Conference on Human System Interaction (HSI) Pub Date : 2022-07-28 DOI: 10.1109/HSI55341.2022.9869500

Qing Tang, K. Jo

{"title":"Person Search via Background and Foreground Contrastive Learning","authors":"Qing Tang, K. Jo","doi":"10.1109/HSI55341.2022.9869500","DOIUrl":"https://doi.org/10.1109/HSI55341.2022.9869500","url":null,"abstract":"The specific person search is the foundation of a wide range of applications in intelligent security and surveillance systems. Although detection and re-id have been widely studied, they are difficult to apply to practical applications directly. Therefore, this paper focuses on person search, which aims to solve person detection and person re-identification (re-id) jointly. The common practice is to append the standard detection loss and re-id branches parallelly on Faster RCNN. The traditional re-id utilized Online Instance Matching (OIM) to pull a sample closer to its identity class. However, the relationship among RoIs of an image has not been fully explored in previous methods. To address this issue, we propose Background and Foreground Contrastive Loss (BFCL) to further boost re-id performance. We consider that RoIs from one image have a high probability of containing similar patterns, which might disturb the re-id performance. Therefore, we proposed BFCL to strengthen the learning of distinguishing similar background and foreground by leveraging inter-RoIs pairwise similarity. In summary, our method jointly optimizes the regression loss, classification loss, re-id loss, and the proposed BFCL for achieving optimal performances in person search model. Experiments are performed on two large-scale person search datasets, CUHK-SYSU and PRW. Results show that the proposed BFCL consistently boosts the performance of the baseline framework SeqNet in two datasets. The improved results demonstrate the effectiveness of the proposed BFCL and the necessity of exploring the relationship among RoIs.","PeriodicalId":282607,"journal":{"name":"2022 15th International Conference on Human System Interaction (HSI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125997157","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}

引用次数: 0