Pub Date : 2019-07-01DOI: 10.1109/NICOInt.2019.00030
Deqiang Zhou, Yunlei Fan, Ganran Deng, Fengguang He, Meili Wang
In order to realize automatic cutting of sugarcane seeds in single bud segment, machine vision technology was used to identify sugarcane nodes. Firstly, the sugarcane color image was obtained, and the R component image of RGB color space was separated, and the median value of R component image was filtered and denoised. Second, the contour information of sugarcane image was obtained by FindContours function, and the area of interest of sugarcane image was selected by the width and height of contour. Finally, Sobel edge detection image of region of interest was acquired, and a rectangular detection operator was constructed to perform integral operation on the interested region to obtain a feature description vector of the interested region. Threshold value of the feature description vector was processed. The peak of the feature description vector was defined as the node feature point. The experimental results show that the recognition rate is 93% and the average time is 0.539 seconds.
{"title":"Research on Algorithm of Sugarcane Nodes Identification Based on Machine Vision","authors":"Deqiang Zhou, Yunlei Fan, Ganran Deng, Fengguang He, Meili Wang","doi":"10.1109/NICOInt.2019.00030","DOIUrl":"https://doi.org/10.1109/NICOInt.2019.00030","url":null,"abstract":"In order to realize automatic cutting of sugarcane seeds in single bud segment, machine vision technology was used to identify sugarcane nodes. Firstly, the sugarcane color image was obtained, and the R component image of RGB color space was separated, and the median value of R component image was filtered and denoised. Second, the contour information of sugarcane image was obtained by FindContours function, and the area of interest of sugarcane image was selected by the width and height of contour. Finally, Sobel edge detection image of region of interest was acquired, and a rectangular detection operator was constructed to perform integral operation on the interested region to obtain a feature description vector of the interested region. Threshold value of the feature description vector was processed. The peak of the feature description vector was defined as the node feature point. The experimental results show that the recognition rate is 93% and the average time is 0.539 seconds.","PeriodicalId":436332,"journal":{"name":"2019 Nicograph International (NicoInt)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121583732","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 : 2019-07-01DOI: 10.1109/NICOInt.2019.00025
S. Nakanishi, Shumpei Akahoshi, Mitsunori Matsushita
In this paper, we propose a method of presenting information by manipulating granular material that are placed on two layers. In recent years, substantial displays, which are displays that present information by manipulating some form of material have been gaining traction in the research community. As these displays utilize materials found in nature, they are capable of blending in with the surrounding environment. To create a novel substantial display, we propose a substantial display that utilizes granular material and speakers. In the proposed system, two types of granular material with different colors and sizes are placed on two layers. By emitting sound waves from a speaker installed under the lower layer, the granular material on the lower layer rises to the upper layer and creates a pattern as the color of the granular material on each layer is different. To determine the optimal parameters, we examined four different granular material and varied the sound wave frequency. Finally, we implemented a system that allows users to emit sound waves at any arbitrary position on the canvas with two layers of granular material.
{"title":"Granular Material Display: Information Presentation System using Two Layers of Granular Material and Speakers","authors":"S. Nakanishi, Shumpei Akahoshi, Mitsunori Matsushita","doi":"10.1109/NICOInt.2019.00025","DOIUrl":"https://doi.org/10.1109/NICOInt.2019.00025","url":null,"abstract":"In this paper, we propose a method of presenting information by manipulating granular material that are placed on two layers. In recent years, substantial displays, which are displays that present information by manipulating some form of material have been gaining traction in the research community. As these displays utilize materials found in nature, they are capable of blending in with the surrounding environment. To create a novel substantial display, we propose a substantial display that utilizes granular material and speakers. In the proposed system, two types of granular material with different colors and sizes are placed on two layers. By emitting sound waves from a speaker installed under the lower layer, the granular material on the lower layer rises to the upper layer and creates a pattern as the color of the granular material on each layer is different. To determine the optimal parameters, we examined four different granular material and varied the sound wave frequency. Finally, we implemented a system that allows users to emit sound waves at any arbitrary position on the canvas with two layers of granular material.","PeriodicalId":436332,"journal":{"name":"2019 Nicograph International (NicoInt)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122224205","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 : 2019-07-01DOI: 10.1109/NICOInt.2019.00022
Yuan Feng, Ning Xie, Yuxin Gu, Shixiao Fan, Chun Yang
The hand gesture conveys many information such as hand sign language, detailed operation by hand, and gesture command in policy and army. In VR application, hand motion capture becomes more and more important in order to build the immerse virtual experience. The existing digital hand glove mainly works on the key point tracking of hands in real time. However, the immerse VR system needs not only the changes of the hand position movement but also the meanings of its movement and gesture in order to create the realistic interaction. In this paper, we treat this issue as a classification problem in supervised learning. We use Principal Component Analysis (PCA) to reduce the dimension of original input and analyze its result. We recombining the new data and raw data according to analysis, reduce not only the amount of computation but also the dataset. In the experiment, we implement our method on Raspberry Pi 3 Model B+. The results demonstrate that our method successfully classify 12 different hand gestures in the accuracy rate is over 93.5% less than 1ms in the immerse system and with just 500 training data.
手势传达了许多信息，如手势语，手的详细操作，手势在政策和军队中的指挥。在虚拟现实应用中，手部动作捕捉对于构建沉浸式虚拟体验变得越来越重要。现有的数字手套主要是对手部的关键点进行实时跟踪。然而，沉浸式VR系统不仅需要手部位置运动的变化，还需要其运动和手势的意义，以创造逼真的交互。在本文中，我们将此问题视为监督学习中的分类问题。我们使用主成分分析(PCA)对原始输入进行降维并分析其结果。我们根据分析将新数据和原始数据进行重组，不仅减少了计算量，而且减少了数据集。在实验中，我们在Raspberry Pi 3 Model B+上实现了我们的方法。结果表明，我们的方法在仅500个训练数据的情况下，在不到1ms的时间内成功地对12种不同的手势进行了分类，准确率超过93.5%。
{"title":"The Recombination of Gesture Nodes Enables Higher Accuracy in Small Data Sets","authors":"Yuan Feng, Ning Xie, Yuxin Gu, Shixiao Fan, Chun Yang","doi":"10.1109/NICOInt.2019.00022","DOIUrl":"https://doi.org/10.1109/NICOInt.2019.00022","url":null,"abstract":"The hand gesture conveys many information such as hand sign language, detailed operation by hand, and gesture command in policy and army. In VR application, hand motion capture becomes more and more important in order to build the immerse virtual experience. The existing digital hand glove mainly works on the key point tracking of hands in real time. However, the immerse VR system needs not only the changes of the hand position movement but also the meanings of its movement and gesture in order to create the realistic interaction. In this paper, we treat this issue as a classification problem in supervised learning. We use Principal Component Analysis (PCA) to reduce the dimension of original input and analyze its result. We recombining the new data and raw data according to analysis, reduce not only the amount of computation but also the dataset. In the experiment, we implement our method on Raspberry Pi 3 Model B+. The results demonstrate that our method successfully classify 12 different hand gestures in the accuracy rate is over 93.5% less than 1ms in the immerse system and with just 500 training data.","PeriodicalId":436332,"journal":{"name":"2019 Nicograph International (NicoInt)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124719258","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 : 2019-07-01DOI: 10.1109/NICOInt.2019.00033
Hideyasu Inoue, Yoshina Takano, R. Thawonmas, Tomohiro Harada
In this paper, we apply a curiosity-driven intrinsic reward to reinforcement learning (RL) in a fighting game and verify its effectiveness. An actor-critic model is used for RL. Our experimental results show that the proposed AI has a better learning ability than an AI using a standard actor-critic model.
{"title":"Verification of Applying Curiosity-Driven to Fighting Game AI","authors":"Hideyasu Inoue, Yoshina Takano, R. Thawonmas, Tomohiro Harada","doi":"10.1109/NICOInt.2019.00033","DOIUrl":"https://doi.org/10.1109/NICOInt.2019.00033","url":null,"abstract":"In this paper, we apply a curiosity-driven intrinsic reward to reinforcement learning (RL) in a fighting game and verify its effectiveness. An actor-critic model is used for RL. Our experimental results show that the proposed AI has a better learning ability than an AI using a standard actor-critic model.","PeriodicalId":436332,"journal":{"name":"2019 Nicograph International (NicoInt)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132499208","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 : 2019-07-01DOI: 10.1109/NICOInt.2019.00021
Daiki Fukunaga, K. Ochi, Y. Obuchi
A rhythm action game is a genre of video game in which the player acts in accordance with visual symbols (charts) representing the rhythm of music. This work aims at generating a chart from the music automatically to realize easy development of rhythm action games. A key-sound is a sound object which is played in response to the player's operation. To generate a chart, it is necessary to classify the sound objects into key-sounds and the other sounds. This process can be done by machine learning, but the training data have a wide variety due to the individuality of creators, and the optimal chart is not unique. Accordingly, we divided the training data into clusters based on the classification tendency and prepared multiple models. Classification experiments using multiple models suggested the possibility of improvement, but further investigation is necessary for the selection of the optimum cluster.
{"title":"Training Data Clustering for Key-Sound Estimation in Rhythm Action Games","authors":"Daiki Fukunaga, K. Ochi, Y. Obuchi","doi":"10.1109/NICOInt.2019.00021","DOIUrl":"https://doi.org/10.1109/NICOInt.2019.00021","url":null,"abstract":"A rhythm action game is a genre of video game in which the player acts in accordance with visual symbols (charts) representing the rhythm of music. This work aims at generating a chart from the music automatically to realize easy development of rhythm action games. A key-sound is a sound object which is played in response to the player's operation. To generate a chart, it is necessary to classify the sound objects into key-sounds and the other sounds. This process can be done by machine learning, but the training data have a wide variety due to the individuality of creators, and the optimal chart is not unique. Accordingly, we divided the training data into clusters based on the classification tendency and prepared multiple models. Classification experiments using multiple models suggested the possibility of improvement, but further investigation is necessary for the selection of the optimum cluster.","PeriodicalId":436332,"journal":{"name":"2019 Nicograph International (NicoInt)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121449455","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 : 2019-07-01DOI: 10.1109/NICOInt.2019.00031
Suguru Ito, Tomohiro Harada, R. Thawonmas
In this paper, we perform a visual analysis of gameplay in a fighting game. In recent years, game streaming is actively performed on platforms typified by Twitch. In order to cope with an increase in the number of users watching such streaming, a method for generating gameplay by AI and providing it to spectators has been recently proposed by our group. However, it is necessary to verify beforehand whether the generated gameplay is appropriate for streaming. Therefore, in this paper, we locate such an issue and describe our improvement plan.
{"title":"A Visual Analysis of Gameplay in a Fighting Game","authors":"Suguru Ito, Tomohiro Harada, R. Thawonmas","doi":"10.1109/NICOInt.2019.00031","DOIUrl":"https://doi.org/10.1109/NICOInt.2019.00031","url":null,"abstract":"In this paper, we perform a visual analysis of gameplay in a fighting game. In recent years, game streaming is actively performed on platforms typified by Twitch. In order to cope with an increase in the number of users watching such streaming, a method for generating gameplay by AI and providing it to spectators has been recently proposed by our group. However, it is necessary to verify beforehand whether the generated gameplay is appropriate for streaming. Therefore, in this paper, we locate such an issue and describe our improvement plan.","PeriodicalId":436332,"journal":{"name":"2019 Nicograph International (NicoInt)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115024842","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 : 2019-07-01DOI: 10.1109/NICOInt.2019.00035
Nao Kotani, S. Mizuno
We propose interactive projection mapping using smoke. In this projection mapping, a user enjoys rays rather than images. The user can change color of rays, move rays, and generate sound interactively by touching rays.
{"title":"A Proposal of Interactive Projection Mapping by Touching Rays Visualized by Smoke","authors":"Nao Kotani, S. Mizuno","doi":"10.1109/NICOInt.2019.00035","DOIUrl":"https://doi.org/10.1109/NICOInt.2019.00035","url":null,"abstract":"We propose interactive projection mapping using smoke. In this projection mapping, a user enjoys rays rather than images. The user can change color of rays, move rays, and generate sound interactively by touching rays.","PeriodicalId":436332,"journal":{"name":"2019 Nicograph International (NicoInt)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115210629","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 : 2019-07-01DOI: 10.1109/NICOInt.2019.00024
Yuankai Gao, Masatoshi Tezuka, Y. Seta, M. Kuwahara, Kazuya Ueki
In this paper, survey about e-books for infant that focus at "learning" for the purpose of popularizing e-books. It is a general thing to obtain knowledge and words from books. It is said that children get a lot of things from books in the process of language learning. For example, story time is important. This study looks at achieving a similar effect of parents reading e-books and books by story-telling through an application.
{"title":"A Study of E-Books for Infants to Learning","authors":"Yuankai Gao, Masatoshi Tezuka, Y. Seta, M. Kuwahara, Kazuya Ueki","doi":"10.1109/NICOInt.2019.00024","DOIUrl":"https://doi.org/10.1109/NICOInt.2019.00024","url":null,"abstract":"In this paper, survey about e-books for infant that focus at \"learning\" for the purpose of popularizing e-books. It is a general thing to obtain knowledge and words from books. It is said that children get a lot of things from books in the process of language learning. For example, story time is important. This study looks at achieving a similar effect of parents reading e-books and books by story-telling through an application.","PeriodicalId":436332,"journal":{"name":"2019 Nicograph International (NicoInt)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115167171","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 : 2019-07-01DOI: 10.1109/NICOInt.2019.00029
K. Ueki
To implement semantic segmentation and assign one of the classes to each pixel, a large amount of pixel labelled images are required. However, annotations in existing image datasets are limited both in terms of quantity and diversity owing to the heavy annotation cost. Therefore, in this study, we examined a method to readily add new classes of training images and evaluate feasibility by testing semantic segmentation on car-mounted camera images.
{"title":"Adding New Classes in Semantic Segmentation","authors":"K. Ueki","doi":"10.1109/NICOInt.2019.00029","DOIUrl":"https://doi.org/10.1109/NICOInt.2019.00029","url":null,"abstract":"To implement semantic segmentation and assign one of the classes to each pixel, a large amount of pixel labelled images are required. However, annotations in existing image datasets are limited both in terms of quantity and diversity owing to the heavy annotation cost. Therefore, in this study, we examined a method to readily add new classes of training images and evaluate feasibility by testing semantic segmentation on car-mounted camera images.","PeriodicalId":436332,"journal":{"name":"2019 Nicograph International (NicoInt)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121836130","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 : 2019-07-01DOI: 10.1109/NICOInt.2019.00018
Xiaolong Zhao, Tingting Li, Jiahui Mao, Meili Wang
To improve the precision and efficiency in 3D models management in digital relief editing, this paper proposes a 3D relief models retrieval method based on ontology semantic analysis. Firstly, according to the specific field(Chinese classical story) of relief art, we classified the models, and established the corresponding ontology for these models. Secondly, after analyzing the relationship among these models, we constructed the association between the ontologies and the rule-based semantic reasoning. Then, we used the similarity calculation based on synonyms to solved the common problem if there is no results when search a resemble word. Finally, we developed a relief models retrieval system, which can retrieve 3D models accurately. The experimental results have shown that the proposed method can effectively assist the digital relief editing when retrieving the 3D models, enhance the relativities among the relief models and improve the retrieval efficiency.
{"title":"Ontology-based 3D Relief Models Retrieval","authors":"Xiaolong Zhao, Tingting Li, Jiahui Mao, Meili Wang","doi":"10.1109/NICOInt.2019.00018","DOIUrl":"https://doi.org/10.1109/NICOInt.2019.00018","url":null,"abstract":"To improve the precision and efficiency in 3D models management in digital relief editing, this paper proposes a 3D relief models retrieval method based on ontology semantic analysis. Firstly, according to the specific field(Chinese classical story) of relief art, we classified the models, and established the corresponding ontology for these models. Secondly, after analyzing the relationship among these models, we constructed the association between the ontologies and the rule-based semantic reasoning. Then, we used the similarity calculation based on synonyms to solved the common problem if there is no results when search a resemble word. Finally, we developed a relief models retrieval system, which can retrieve 3D models accurately. The experimental results have shown that the proposed method can effectively assist the digital relief editing when retrieving the 3D models, enhance the relativities among the relief models and improve the retrieval efficiency.","PeriodicalId":436332,"journal":{"name":"2019 Nicograph International (NicoInt)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132722059","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：