Pub Date : 2022-12-15DOI: 10.1109/ICNSC55942.2022.10004081
Yawen Kou, Yangming Zhou, Mengchu Zhou
A soft-clustered vehicle routing problem (SoftClu-VRP) is an important variant of the well-known capacitated vehicle routing problem, where customers are partitioned into clusters and all customers of the same cluster must be served by the same vehicle. As a highly useful model for parcel delivery in courier companies, SoftCluVRP is NP-hard. In this work, we propose an efficient metaheuristic algorithm for solving it. Starting from an initial population, it iterates by using a solution recombination operator (to generate a promising offspring solution), a hybrid neighborhood search (to find a high-quality local optimum), and a population updating strategy (to manage a healthy population). Experiments on two groups of 378 widely-used benchmark instances show that it achieves highly competitive performance compared to state-of-the-art algorithms. In particular, our algorithm finds the best upper bounds on 320 instances.
{"title":"An Efficient Metaheuristic Algorithm for Solving Soft-clustered Vehicle Routing Problems","authors":"Yawen Kou, Yangming Zhou, Mengchu Zhou","doi":"10.1109/ICNSC55942.2022.10004081","DOIUrl":"https://doi.org/10.1109/ICNSC55942.2022.10004081","url":null,"abstract":"A soft-clustered vehicle routing problem (SoftClu-VRP) is an important variant of the well-known capacitated vehicle routing problem, where customers are partitioned into clusters and all customers of the same cluster must be served by the same vehicle. As a highly useful model for parcel delivery in courier companies, SoftCluVRP is NP-hard. In this work, we propose an efficient metaheuristic algorithm for solving it. Starting from an initial population, it iterates by using a solution recombination operator (to generate a promising offspring solution), a hybrid neighborhood search (to find a high-quality local optimum), and a population updating strategy (to manage a healthy population). Experiments on two groups of 378 widely-used benchmark instances show that it achieves highly competitive performance compared to state-of-the-art algorithms. In particular, our algorithm finds the best upper bounds on 320 instances.","PeriodicalId":230499,"journal":{"name":"2022 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"120 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124508842","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 : 2022-12-15DOI: 10.1109/ICNSC55942.2022.10004131
Xuhang Li, Jiliang Luo, Jun Li, Sijia Yi, Chunrong Pan
Parallel Petri nets are a class of Petri nets which can take into account scheduling and control issues of manufacturing systems [1]. However, Its designed relies on the manual effort, which is very difficult and boring for real manufacturing systems. Therefore, this work defines an improved planning domain definition language (PDDL) to automatically translate it to a parallel Petri net. In details, the PDDL syntax is extended to make it more accurate and convenient to describe actions, tasks and conditions. Elements of an extended PDDL are automatically represented by places and transitions of a parallel Petri net. Finally, an experiment is taken to illustrate and verify our method.
{"title":"Parallel Petri Nets Modeling Method Of Manufacturing System Based On The improved PDDL","authors":"Xuhang Li, Jiliang Luo, Jun Li, Sijia Yi, Chunrong Pan","doi":"10.1109/ICNSC55942.2022.10004131","DOIUrl":"https://doi.org/10.1109/ICNSC55942.2022.10004131","url":null,"abstract":"Parallel Petri nets are a class of Petri nets which can take into account scheduling and control issues of manufacturing systems [1]. However, Its designed relies on the manual effort, which is very difficult and boring for real manufacturing systems. Therefore, this work defines an improved planning domain definition language (PDDL) to automatically translate it to a parallel Petri net. In details, the PDDL syntax is extended to make it more accurate and convenient to describe actions, tasks and conditions. Elements of an extended PDDL are automatically represented by places and transitions of a parallel Petri net. Finally, an experiment is taken to illustrate and verify our method.","PeriodicalId":230499,"journal":{"name":"2022 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"26 Suppl 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126408285","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}
Multichannel EEG is generally used to collect brain activities from various locations across the brain. However, BCIs using lesser channels will be more convenient for subjects. What's more, information acquired from adjacent channels is usually inter-correlated or irrelevant to the task. And some channels are noisy. This paper proposes a novel channel selection algorithm based on reinforcement learning. It can adaptively transform the full-channel EEG data to the optimal-channel-number EEG format conditioned on different input trials to make a trade-off between brain decoding accuracy and efficiency. Experimen-tal results showed that the proposed model can improve the classification accuracy by 2% ~ 6% compared to channel set ${C3,C4,Cz}$.
{"title":"EEG channel selection algorithm based on Reinforcement Learning","authors":"Yingxin Jin, Shaohua Shang, Liwei Tang, Lianzhua He, Mengchu Zhou","doi":"10.1109/ICNSC55942.2022.10004161","DOIUrl":"https://doi.org/10.1109/ICNSC55942.2022.10004161","url":null,"abstract":"Multichannel EEG is generally used to collect brain activities from various locations across the brain. However, BCIs using lesser channels will be more convenient for subjects. What's more, information acquired from adjacent channels is usually inter-correlated or irrelevant to the task. And some channels are noisy. This paper proposes a novel channel selection algorithm based on reinforcement learning. It can adaptively transform the full-channel EEG data to the optimal-channel-number EEG format conditioned on different input trials to make a trade-off between brain decoding accuracy and efficiency. Experimen-tal results showed that the proposed model can improve the classification accuracy by 2% ~ 6% compared to channel set ${C3,C4,Cz}$.","PeriodicalId":230499,"journal":{"name":"2022 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130926090","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 : 2022-12-15DOI: 10.1109/ICNSC55942.2022.10004182
Yijie Yuan, Shaopeng Jia, Fei Wang, Xiong Chen
For the few-shot classification problem of low-resolution(LR) images, we propose a meta-learning method based on prototypical networks called LR-ProtoNet. The metric-based meta-learning algorithm mainly extracts the features of the support and query samples through the feature encoder and obtains the prediction categories from a metric module. Our core idea is to add feature-affine layers in the feature encoder to increase the feature distribution of LR images, and use Brownian Distance Covariance(BDC) in the metric module to capture the joint distribution and nonlinear relationship between different affine transformations. We down-sample standard few-shot image datasets to simulate LR images and conduct extensive ablation experiments and comparative studies of other meta methods in general image recognition and fine-grained classification. Experimental results demonstrate that our proposed model can effectively utilize low-resolution image information, achieving state-of-the-art performance compared to baseline works.
{"title":"LR-ProtoNet: Meta-Learning for Low-Resolution Few-Shot Recognition and Classification","authors":"Yijie Yuan, Shaopeng Jia, Fei Wang, Xiong Chen","doi":"10.1109/ICNSC55942.2022.10004182","DOIUrl":"https://doi.org/10.1109/ICNSC55942.2022.10004182","url":null,"abstract":"For the few-shot classification problem of low-resolution(LR) images, we propose a meta-learning method based on prototypical networks called LR-ProtoNet. The metric-based meta-learning algorithm mainly extracts the features of the support and query samples through the feature encoder and obtains the prediction categories from a metric module. Our core idea is to add feature-affine layers in the feature encoder to increase the feature distribution of LR images, and use Brownian Distance Covariance(BDC) in the metric module to capture the joint distribution and nonlinear relationship between different affine transformations. We down-sample standard few-shot image datasets to simulate LR images and conduct extensive ablation experiments and comparative studies of other meta methods in general image recognition and fine-grained classification. Experimental results demonstrate that our proposed model can effectively utilize low-resolution image information, achieving state-of-the-art performance compared to baseline works.","PeriodicalId":230499,"journal":{"name":"2022 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130998312","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 : 2022-12-15DOI: 10.1109/ICNSC55942.2022.10004079
Qing Zhao, Zhilong Ma, Jianqiang Li
Automatically extract relations between medical entity pairs is fundamental in biomedical research. Since the annotated dataset is very expensive, distant supervision provides an efficient solution to reduce the cost of annotation by utilizing rough corpus labeled with semantic knowledge base. However, two same entities mentioned in different sentences may express different relations, it is difficult for the traditional distant supervision methods to distinguish these different relations. In this paper, we propose a new model for biomedical relation extraction in Chinese EMRs. First, the distant supervision is used for coarse-grained relation labeling. Then, the fine-grained relations are annotated initially by measuring the distance between the contextual information of the relation instance to the semantic profile of each candidate fine-grained relation category. Finally, the high confidence fine-grained relation instances are selected as initial training set for PCNN model, in addition, a bootstrap learning is introduced in the training process to enhance the performance of fine-grained relation extraction. Experiments conducted on a real-word dataset and the results show that our method outperforms all baseline systems.
{"title":"Distant supervision for fine-grained biomedical relation extraction from Chinese EMRs","authors":"Qing Zhao, Zhilong Ma, Jianqiang Li","doi":"10.1109/ICNSC55942.2022.10004079","DOIUrl":"https://doi.org/10.1109/ICNSC55942.2022.10004079","url":null,"abstract":"Automatically extract relations between medical entity pairs is fundamental in biomedical research. Since the annotated dataset is very expensive, distant supervision provides an efficient solution to reduce the cost of annotation by utilizing rough corpus labeled with semantic knowledge base. However, two same entities mentioned in different sentences may express different relations, it is difficult for the traditional distant supervision methods to distinguish these different relations. In this paper, we propose a new model for biomedical relation extraction in Chinese EMRs. First, the distant supervision is used for coarse-grained relation labeling. Then, the fine-grained relations are annotated initially by measuring the distance between the contextual information of the relation instance to the semantic profile of each candidate fine-grained relation category. Finally, the high confidence fine-grained relation instances are selected as initial training set for PCNN model, in addition, a bootstrap learning is introduced in the training process to enhance the performance of fine-grained relation extraction. Experiments conducted on a real-word dataset and the results show that our method outperforms all baseline systems.","PeriodicalId":230499,"journal":{"name":"2022 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132276681","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 : 2022-12-15DOI: 10.1109/ICNSC55942.2022.10004086
Jie Zong, Zhijun Ding
Link prediction is to predict whether there is a link between two nodes in the graph, it is a very important application and plays a great role in various industries. In recent years, with the development of graph neural network technology, many algorithms make effort to study the expression of each node from the original graph data and use them to infer new links. However, most of the existing algorithms have a common problem when facing heterogeneous graphs, which is, they do not consider the weight of edges in graphs. Instead, they put all their energies into computing node features. Although a few algorithms such as RGCN are trying to take the influence of different link types into account while extracting node features, these implicit feature extractions do not start from the global information, but just calculate independently for each node. In other words, in these algorithms, even the same type of links will be abstracted into different features on different nodes. This is obviously inconsistent with reality. On the same map, the feature of the same link should be relatively fixed and should not be changed just because of different positions. In addition, when the current graph neural network algorithm is applied to link prediction, the link type to be predicted must be specified in advance, which makes the algorithm extremely inflexible. In order to solve these problems, we propose an edge weight calculation algorithm that extracts the edge feature from the whole graph. We also propose the edge-weight-based link prediction algorithm. By introducing edge weight into the MLP, there is no need to specify the target link type at the beginning of model training. It improves both the performance and efficiency of the link prediction model. Experiments on two datasets show that this edge-weight-based link prediction algorithm performs better than current algorithms and reaches SOTA.
{"title":"Edge-weight-Based link prediction in heterogeneous graph","authors":"Jie Zong, Zhijun Ding","doi":"10.1109/ICNSC55942.2022.10004086","DOIUrl":"https://doi.org/10.1109/ICNSC55942.2022.10004086","url":null,"abstract":"Link prediction is to predict whether there is a link between two nodes in the graph, it is a very important application and plays a great role in various industries. In recent years, with the development of graph neural network technology, many algorithms make effort to study the expression of each node from the original graph data and use them to infer new links. However, most of the existing algorithms have a common problem when facing heterogeneous graphs, which is, they do not consider the weight of edges in graphs. Instead, they put all their energies into computing node features. Although a few algorithms such as RGCN are trying to take the influence of different link types into account while extracting node features, these implicit feature extractions do not start from the global information, but just calculate independently for each node. In other words, in these algorithms, even the same type of links will be abstracted into different features on different nodes. This is obviously inconsistent with reality. On the same map, the feature of the same link should be relatively fixed and should not be changed just because of different positions. In addition, when the current graph neural network algorithm is applied to link prediction, the link type to be predicted must be specified in advance, which makes the algorithm extremely inflexible. In order to solve these problems, we propose an edge weight calculation algorithm that extracts the edge feature from the whole graph. We also propose the edge-weight-based link prediction algorithm. By introducing edge weight into the MLP, there is no need to specify the target link type at the beginning of model training. It improves both the performance and efficiency of the link prediction model. Experiments on two datasets show that this edge-weight-based link prediction algorithm performs better than current algorithms and reaches SOTA.","PeriodicalId":230499,"journal":{"name":"2022 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115382093","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}
Since the rapid development of deep learning, image inpainting techniques have also improved significantly. Although these techniques have been able to reconstruct semantically coherent and visually plausible masked regions compared to traditional techniques, the results of these works are commonly blurry due to lack fine details. This paper proposes a novel model including an image completion network and an edge matching module. The image completion network is a Generative Adversarial Network framework added skip-connection for contextual feature fusion, and the edge matching network facilitates the image inpainting network by constraining the edge of results. We evaluate our model on the publicly available datasets CelebA and Places2. Results show that our method outperforms current representative technique quantitatively and qualitatively.
{"title":"Generative Image Inpainting for Fine Details","authors":"Xueqing Yang, Xiaoxin Fang, Xiong Chen, Zhenyu Shan","doi":"10.1109/ICNSC55942.2022.10004107","DOIUrl":"https://doi.org/10.1109/ICNSC55942.2022.10004107","url":null,"abstract":"Since the rapid development of deep learning, image inpainting techniques have also improved significantly. Although these techniques have been able to reconstruct semantically coherent and visually plausible masked regions compared to traditional techniques, the results of these works are commonly blurry due to lack fine details. This paper proposes a novel model including an image completion network and an edge matching module. The image completion network is a Generative Adversarial Network framework added skip-connection for contextual feature fusion, and the edge matching network facilitates the image inpainting network by constraining the edge of results. We evaluate our model on the publicly available datasets CelebA and Places2. Results show that our method outperforms current representative technique quantitatively and qualitatively.","PeriodicalId":230499,"journal":{"name":"2022 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115751162","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 : 2022-12-15DOI: 10.1109/ICNSC55942.2022.10004144
Chao Ma, Jing An, Xiang-En Bai, Hanqiu Bao
Semi-supervised node classification of imbalanced graphs is one of the important tasks in the field of graph neural network (GNN). Most of the current methods focus on how to aggregate feature information from neighbor nodes, but they do not distinguish the importance of minority class and majority class samples in the process of aggregation. To this end, this paper introduces an attention mechanism in the process of aggregating feature information, which flexibly assigns individualized weights to minority and majority class samples. At the same time, we improve the loss function using cost-sensitive techniques to increase the minority class misclassification cost. Finally, we design experiments to verify the effectiveness of the proposed method.
{"title":"Attention and Cost-Sensitive Graph Neural Network for Imbalanced Node Classification","authors":"Chao Ma, Jing An, Xiang-En Bai, Hanqiu Bao","doi":"10.1109/ICNSC55942.2022.10004144","DOIUrl":"https://doi.org/10.1109/ICNSC55942.2022.10004144","url":null,"abstract":"Semi-supervised node classification of imbalanced graphs is one of the important tasks in the field of graph neural network (GNN). Most of the current methods focus on how to aggregate feature information from neighbor nodes, but they do not distinguish the importance of minority class and majority class samples in the process of aggregation. To this end, this paper introduces an attention mechanism in the process of aggregating feature information, which flexibly assigns individualized weights to minority and majority class samples. At the same time, we improve the loss function using cost-sensitive techniques to increase the minority class misclassification cost. Finally, we design experiments to verify the effectiveness of the proposed method.","PeriodicalId":230499,"journal":{"name":"2022 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114348420","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 : 2022-12-15DOI: 10.1109/ICNSC55942.2022.10004186
Sunsi Fu, Xiaoxin Fang, Xiong Chen
As a kind of special nonlinear phenomenon, chaos has obtained much attention due to its interesting characteristics, such as randomness, sensibility, and complexity. How to predict chaos effectively and accurately is a significant issue in the nonlinear area. In this paper, a hybrid enhanced echo state network (HEESN) is proposed for the nonlinear prediction of multivariate chaotic time series. The HEESN scheme is contributed by three interactional aspects: output weight regularization, initial parameter optimization, and chaotic signal reconstruction. First, to enhance noise robustness, a sparse regression based on L2 regularization is employed to finely learn the output weights of ESN. Second, vital reservoir parameters (i.e., global scaling factor, reservoir size, scaling coefficient, and sparsity degree) are learned by a linear-weighted particle swarm optimization (LW-PSO) to further improve prediction accuracy and reliability. Third, recommendations of key settings in the signal reconstruction stage (i.e., embedding dimension and time delay) are studied and given according to the temporal complexity and signal-to-noise ratio of the predicted time series. Extensive experiments about computational complexity and three evaluating metrics are carried out on one chaotic benchmark. The analyzed results indicate that the proposed HEESN performs promisingly on multivariate chaotic time series prediction.
{"title":"Hybrid Enhanced Echo State Network for Nonlinear Prediction of Multivariate Chaotic Time Series","authors":"Sunsi Fu, Xiaoxin Fang, Xiong Chen","doi":"10.1109/ICNSC55942.2022.10004186","DOIUrl":"https://doi.org/10.1109/ICNSC55942.2022.10004186","url":null,"abstract":"As a kind of special nonlinear phenomenon, chaos has obtained much attention due to its interesting characteristics, such as randomness, sensibility, and complexity. How to predict chaos effectively and accurately is a significant issue in the nonlinear area. In this paper, a hybrid enhanced echo state network (HEESN) is proposed for the nonlinear prediction of multivariate chaotic time series. The HEESN scheme is contributed by three interactional aspects: output weight regularization, initial parameter optimization, and chaotic signal reconstruction. First, to enhance noise robustness, a sparse regression based on L2 regularization is employed to finely learn the output weights of ESN. Second, vital reservoir parameters (i.e., global scaling factor, reservoir size, scaling coefficient, and sparsity degree) are learned by a linear-weighted particle swarm optimization (LW-PSO) to further improve prediction accuracy and reliability. Third, recommendations of key settings in the signal reconstruction stage (i.e., embedding dimension and time delay) are studied and given according to the temporal complexity and signal-to-noise ratio of the predicted time series. Extensive experiments about computational complexity and three evaluating metrics are carried out on one chaotic benchmark. The analyzed results indicate that the proposed HEESN performs promisingly on multivariate chaotic time series prediction.","PeriodicalId":230499,"journal":{"name":"2022 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123212488","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 : 2022-12-15DOI: 10.1109/ICNSC55942.2022.10004126
Xueyi Chi, Huiliang Shang, Xiong-Zi Chen
Most of the commonly used learning target detection algorithms require a large amount of data sets and time for training, and if the target has to be changed, the network needs to be retrained. In response to this problem, we aim to build a vision-based grasping system, which acquires target features through multi-angle demonstration, and can select an appropriate matching method according to the geometric shape of the target to detect more accurately. The method involves improved template matching, comparing the means of BGR channels and shape parameter with the features from demonstration. Our improvements to the template matching algorithm solve the shortcomings of its inability to recognize rotated targets. We also combine 2D recognition with 3D point clouds to obtain the grasping point. It has been verified by simulation experiments that our vision guided manipulation system can learn and extract the target features through a few demonstrations, and select an appropriate method to detect the target, the robotic arm performs manipulations such as grasping the target.
{"title":"Vision Guided Manipulation by Learning from Demonstration","authors":"Xueyi Chi, Huiliang Shang, Xiong-Zi Chen","doi":"10.1109/ICNSC55942.2022.10004126","DOIUrl":"https://doi.org/10.1109/ICNSC55942.2022.10004126","url":null,"abstract":"Most of the commonly used learning target detection algorithms require a large amount of data sets and time for training, and if the target has to be changed, the network needs to be retrained. In response to this problem, we aim to build a vision-based grasping system, which acquires target features through multi-angle demonstration, and can select an appropriate matching method according to the geometric shape of the target to detect more accurately. The method involves improved template matching, comparing the means of BGR channels and shape parameter with the features from demonstration. Our improvements to the template matching algorithm solve the shortcomings of its inability to recognize rotated targets. We also combine 2D recognition with 3D point clouds to obtain the grasping point. It has been verified by simulation experiments that our vision guided manipulation system can learn and extract the target features through a few demonstrations, and select an appropriate method to detect the target, the robotic arm performs manipulations such as grasping the target.","PeriodicalId":230499,"journal":{"name":"2022 IEEE International Conference on Networking, Sensing and Control (ICNSC)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128519900","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：