Ivan Nunez-Garcia, Rocio A. Lizarraga-Morales, Geovanni Hernandez-Gomez
Considering the massive increase in digitized databases of visual art, the classification of each artistic element, either by style or genre, is an important task for its correct administration and understanding. In this paper, an automatic system for classification of paintings by artistic genre is proposed. In our approach, we use a combination of color represented in perceptual color spaces and texture descriptors. Other methods use isolated information of color or texture, in our approach, we relate them from a perceptual point of view. Using an artificial neural network, the proposed system classifies 7 different genres which are: Abstract Expressionism, Cubism, Impressionism, Pop art, Renaissance, Romanticism, and Mexican muralism. Experiments show that the synergistic integration of features in this framework results in better accuracy, in comparison with other related state-of-the-art approaches.
{"title":"Classification of Paintings by Artistic Genre Integrating Color and Texture Descriptors","authors":"Ivan Nunez-Garcia, Rocio A. Lizarraga-Morales, Geovanni Hernandez-Gomez","doi":"10.1145/3268866.3268885","DOIUrl":"https://doi.org/10.1145/3268866.3268885","url":null,"abstract":"Considering the massive increase in digitized databases of visual art, the classification of each artistic element, either by style or genre, is an important task for its correct administration and understanding. In this paper, an automatic system for classification of paintings by artistic genre is proposed. In our approach, we use a combination of color represented in perceptual color spaces and texture descriptors. Other methods use isolated information of color or texture, in our approach, we relate them from a perceptual point of view. Using an artificial neural network, the proposed system classifies 7 different genres which are: Abstract Expressionism, Cubism, Impressionism, Pop art, Renaissance, Romanticism, and Mexican muralism. Experiments show that the synergistic integration of features in this framework results in better accuracy, in comparison with other related state-of-the-art approaches.","PeriodicalId":285628,"journal":{"name":"Proceedings of the 2018 International Conference on Artificial Intelligence and Pattern Recognition","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129481812","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}
The electronic health record (EHR) analysis has become an increasingly important application for artificial intelligence (AI) algorithms to leverage the insight from the big data for improving the quality of human healthcare. In a lot of Chinese EHR analysis applications, it is very important to categorize the patients' diseases according to the medical coding standard. In this paper, we develop NLP and machine learning algorithms to automatically categorize each patient's individual diseases into the ICD-10 coding standard. Experimental results show that the support vector machine algorithm (SVM) accomplishes very promising classification results.
{"title":"Categorization of Patient Disease into ICD-10 with NLP and SVM for Chinese Electronic Health Record Analysis","authors":"J. Zhong, Chuangui Gao, X. Yi","doi":"10.1145/3268866.3268877","DOIUrl":"https://doi.org/10.1145/3268866.3268877","url":null,"abstract":"The electronic health record (EHR) analysis has become an increasingly important application for artificial intelligence (AI) algorithms to leverage the insight from the big data for improving the quality of human healthcare. In a lot of Chinese EHR analysis applications, it is very important to categorize the patients' diseases according to the medical coding standard. In this paper, we develop NLP and machine learning algorithms to automatically categorize each patient's individual diseases into the ICD-10 coding standard. Experimental results show that the support vector machine algorithm (SVM) accomplishes very promising classification results.","PeriodicalId":285628,"journal":{"name":"Proceedings of the 2018 International Conference on Artificial Intelligence and Pattern Recognition","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115549040","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}
Aiming at the problem of lower recognition accuracy of video airport targets under complex conditions, the paper proposes a video airport target recognition method. The paper uses clustering method to extract the key-frames containing airport targets. According to the morphological processing results and the extracted contour features, the paper recognizes multiple potential areas including airport targets, and adopts Adaboost method based on Support Vector Machine (SVM) to recognize airport targets. The experimental results show the method can accurately recognize video airport targets.
{"title":"A Video Airport Target Recognition Method","authors":"Yongmei Zhang, Chao Feng, Kuo Xing, Jiong Peng","doi":"10.1145/3268866.3268869","DOIUrl":"https://doi.org/10.1145/3268866.3268869","url":null,"abstract":"Aiming at the problem of lower recognition accuracy of video airport targets under complex conditions, the paper proposes a video airport target recognition method. The paper uses clustering method to extract the key-frames containing airport targets. According to the morphological processing results and the extracted contour features, the paper recognizes multiple potential areas including airport targets, and adopts Adaboost method based on Support Vector Machine (SVM) to recognize airport targets. The experimental results show the method can accurately recognize video airport targets.","PeriodicalId":285628,"journal":{"name":"Proceedings of the 2018 International Conference on Artificial Intelligence and Pattern Recognition","volume":"147 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133678637","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}
In this paper, a novel Deep Discriminant Model, DDM is proposed for predicting imminent collisions caused by dangerous lane change, which can be utilized as a collision avoidance control strategy for advanced driver assistance system. Different from previous work, the proposed approach incorporates multiple visual information about the driving environment, as well as the vehicle state and driver's physiological information, information about the uncertainty inherent, and decision making from the spatio-temporal information to the task. In particular, a special network, ConvLSTMs is presented, which is a combination of convolutional and recurrent layers, to process the input image sensor data in both time and spatial domain. The DDM has the ability of extracting features from multiple data sources (e.g., visual, vehicle state and physiological data) in a deep network. Experiments in a simulation environment showed that the DDM can learn to predict impending collisions with an accuracy of 80.8%, especially when multiple modality sensor data are used as input.
{"title":"Collision Avoidance Control for Advanced Driver Assistance System Based on Deep Discriminant Model","authors":"Jun Gao, Honghui Zhu, Y. Murphey","doi":"10.1145/3268866.3268872","DOIUrl":"https://doi.org/10.1145/3268866.3268872","url":null,"abstract":"In this paper, a novel Deep Discriminant Model, DDM is proposed for predicting imminent collisions caused by dangerous lane change, which can be utilized as a collision avoidance control strategy for advanced driver assistance system. Different from previous work, the proposed approach incorporates multiple visual information about the driving environment, as well as the vehicle state and driver's physiological information, information about the uncertainty inherent, and decision making from the spatio-temporal information to the task. In particular, a special network, ConvLSTMs is presented, which is a combination of convolutional and recurrent layers, to process the input image sensor data in both time and spatial domain. The DDM has the ability of extracting features from multiple data sources (e.g., visual, vehicle state and physiological data) in a deep network. Experiments in a simulation environment showed that the DDM can learn to predict impending collisions with an accuracy of 80.8%, especially when multiple modality sensor data are used as input.","PeriodicalId":285628,"journal":{"name":"Proceedings of the 2018 International Conference on Artificial Intelligence and Pattern Recognition","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130516613","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}
As a special professional group, the seafarers' psychological health is very important to the safety of shipping, and has drawn great attention from the industry. As an emerging neuroimaging technique, functional magnetic resonance image (fMRI) has been widely applied to the field of mental health research. In this paper, in order to explore the possible influence of maritime working and living environments on the brain functional network of seafarers, a new method of functional connectivity detection is proposed to obtain more accurate brain functional networks, which is implemented by using the classical independent component analysis (ICA) method with intrinsic priori information from the fMRI data itself. Finally, the experimental results of real seafarers' fMRI data demonstrate that the functional connectivity of seafarers has been changed before and after sailing.
{"title":"The Study of Seafarer's Brain Functional Connectivity Before and After Sailling Using fMRI","authors":"Yuhu Shi, Weiming Zeng","doi":"10.1145/3268866.3268876","DOIUrl":"https://doi.org/10.1145/3268866.3268876","url":null,"abstract":"As a special professional group, the seafarers' psychological health is very important to the safety of shipping, and has drawn great attention from the industry. As an emerging neuroimaging technique, functional magnetic resonance image (fMRI) has been widely applied to the field of mental health research. In this paper, in order to explore the possible influence of maritime working and living environments on the brain functional network of seafarers, a new method of functional connectivity detection is proposed to obtain more accurate brain functional networks, which is implemented by using the classical independent component analysis (ICA) method with intrinsic priori information from the fMRI data itself. Finally, the experimental results of real seafarers' fMRI data demonstrate that the functional connectivity of seafarers has been changed before and after sailing.","PeriodicalId":285628,"journal":{"name":"Proceedings of the 2018 International Conference on Artificial Intelligence and Pattern Recognition","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127012484","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}
In view of the face recognition system implemented on the traditional computer, the face recognition technology is combined with the embedded system, which is not easy to carry and work inefficiency. The current mainstream embedded systems have the advantages of high chip integration, minimization of hardware and software, high automation, concurrent processing, real-time response, and stability and reliability. This system can not only play the advantages of biometric identification, but also make full use of the characteristics of the embedded system with small body volume, low cost and stable reliability. It is the development trend of face recognition system. In view of the above two points, the face recognition system based on embedded processor is deeply researched, and a more accurate recognition result is obtained.
{"title":"Research on Face Recognition System based on Embedded Processor and Deep Neural Network","authors":"Bowen Du, Xiaoxia Guo, Y. Chen","doi":"10.1145/3268866.3268880","DOIUrl":"https://doi.org/10.1145/3268866.3268880","url":null,"abstract":"In view of the face recognition system implemented on the traditional computer, the face recognition technology is combined with the embedded system, which is not easy to carry and work inefficiency. The current mainstream embedded systems have the advantages of high chip integration, minimization of hardware and software, high automation, concurrent processing, real-time response, and stability and reliability. This system can not only play the advantages of biometric identification, but also make full use of the characteristics of the embedded system with small body volume, low cost and stable reliability. It is the development trend of face recognition system. In view of the above two points, the face recognition system based on embedded processor is deeply researched, and a more accurate recognition result is obtained.","PeriodicalId":285628,"journal":{"name":"Proceedings of the 2018 International Conference on Artificial Intelligence and Pattern Recognition","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132672356","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}
Histopathological examination of tissues is vital to diagnosis of tumor. Statistical model has been proved to be efficiently in detecting of DAB staining in histopathology images. However, there is no statement of which color space is the best for statistical model in detecting of DAB staining. This paper have tested statistical model with 50 pairwisely re-combined color components from 17 color spaces. The experimental results have shown the combination CM of CMYK color space achieves the best. We also compared the statistical model with current popular DAB staining detection methods, and demonstrated the statistical model is the best.
{"title":"The Selection of Best Color Components Combination for Statistical Model with Application to DAB Staining Detection","authors":"Jie Shu, Yang Wang, Lei Jiang","doi":"10.1145/3268866.3268870","DOIUrl":"https://doi.org/10.1145/3268866.3268870","url":null,"abstract":"Histopathological examination of tissues is vital to diagnosis of tumor. Statistical model has been proved to be efficiently in detecting of DAB staining in histopathology images. However, there is no statement of which color space is the best for statistical model in detecting of DAB staining. This paper have tested statistical model with 50 pairwisely re-combined color components from 17 color spaces. The experimental results have shown the combination CM of CMYK color space achieves the best. We also compared the statistical model with current popular DAB staining detection methods, and demonstrated the statistical model is the best.","PeriodicalId":285628,"journal":{"name":"Proceedings of the 2018 International Conference on Artificial Intelligence and Pattern Recognition","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127673748","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}
Gamma knife radiosurgery is one of the main methods for tumor treatment, but the presence of set-up error results in a certain deviation between the irradiation target and the preselected target in radiotherapy. In this paper, a method for the positioning verification of irradiation target in Gamma knife radiosurgery based on the minimum projection error is proposed. By comparing the projection of the actual irradiation target with that of each spot in the small area around the tumor spot, the location of the actual irradiation target in the radiation treatment is achieved. The simulation results show our method is feasible and stability.
{"title":"Positioning Verification of Irradiation Target in Gamma Knife Radiosurgery Based on Minimum Projection Error","authors":"Xiuqing Li, Bingzhen Lei, Jun Zhang, Junhai Wen","doi":"10.1145/3268866.3268881","DOIUrl":"https://doi.org/10.1145/3268866.3268881","url":null,"abstract":"Gamma knife radiosurgery is one of the main methods for tumor treatment, but the presence of set-up error results in a certain deviation between the irradiation target and the preselected target in radiotherapy. In this paper, a method for the positioning verification of irradiation target in Gamma knife radiosurgery based on the minimum projection error is proposed. By comparing the projection of the actual irradiation target with that of each spot in the small area around the tumor spot, the location of the actual irradiation target in the radiation treatment is achieved. The simulation results show our method is feasible and stability.","PeriodicalId":285628,"journal":{"name":"Proceedings of the 2018 International Conference on Artificial Intelligence and Pattern Recognition","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123901810","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}
Wireless sensor networks that include large numbers of autonomous sensors are attracting increasing attention over the recent years since it can potentially play an important role in many applications such as healthcare monitoring, environmental/earth sensing, human detection in disasters, etc. This paper presents a novel wireless sensor network prototype based on commercial GSM network for data transfer. Each sensor node integrates with a microprocessor and a GPS module, and therefore, can collect information tagged with the current location. The sensor nodes are powered by electromagnetic energy harvesting technology and therefore eliminates the need for battery replacement. Thanks to the wide coverage of GSM network, such a wireless sensor network is particularly fit for applications where the sensors are spread in expected to operate over a long period, such as environmental monitoring and structural health monitoring.
{"title":"A Wireless Sensor Network Prototype Based on GSM Technology for Remote Data Collection","authors":"Xinghua Ren, Yi Zou, Shifan Luo","doi":"10.1145/3268866.3268874","DOIUrl":"https://doi.org/10.1145/3268866.3268874","url":null,"abstract":"Wireless sensor networks that include large numbers of autonomous sensors are attracting increasing attention over the recent years since it can potentially play an important role in many applications such as healthcare monitoring, environmental/earth sensing, human detection in disasters, etc. This paper presents a novel wireless sensor network prototype based on commercial GSM network for data transfer. Each sensor node integrates with a microprocessor and a GPS module, and therefore, can collect information tagged with the current location. The sensor nodes are powered by electromagnetic energy harvesting technology and therefore eliminates the need for battery replacement. Thanks to the wide coverage of GSM network, such a wireless sensor network is particularly fit for applications where the sensors are spread in expected to operate over a long period, such as environmental monitoring and structural health monitoring.","PeriodicalId":285628,"journal":{"name":"Proceedings of the 2018 International Conference on Artificial Intelligence and Pattern Recognition","volume":"35 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120967795","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}
We present a new method for action recognition that employs the co-occurrence of concept features as semantic geometric context. Firstly, the semantic concept codebook is learnt by an improved subspace clustering, then the spatio-temporal interest points are labelled as meaningful features, namely concept features. After that, Multi-scale co-occurrence statistics that embeds the relative distance and direction of pairwise concept features is constructed. Unlike the traditional k-means, the features labelled by the concept codebook well represent the ingredients of objects and ensure temporal consistency. Moreover, the relative layout is the semantic geometric context that describes the changes of geometric relationships. Using the popular KTH and UCF-sports action datasets, experimental results show that the relative layouts combined with the STIPs have discriminative power for action recognition. Our method obtains promising recognition performance compared with other state-of-the-art algorithms.
{"title":"Discriminative Co-Occurrence of Concept Features for Action Recognition","authors":"Tongchi Zhou, Qinjun Xu, A. Hamdulla","doi":"10.1145/3268866.3268871","DOIUrl":"https://doi.org/10.1145/3268866.3268871","url":null,"abstract":"We present a new method for action recognition that employs the co-occurrence of concept features as semantic geometric context. Firstly, the semantic concept codebook is learnt by an improved subspace clustering, then the spatio-temporal interest points are labelled as meaningful features, namely concept features. After that, Multi-scale co-occurrence statistics that embeds the relative distance and direction of pairwise concept features is constructed. Unlike the traditional k-means, the features labelled by the concept codebook well represent the ingredients of objects and ensure temporal consistency. Moreover, the relative layout is the semantic geometric context that describes the changes of geometric relationships. Using the popular KTH and UCF-sports action datasets, experimental results show that the relative layouts combined with the STIPs have discriminative power for action recognition. Our method obtains promising recognition performance compared with other state-of-the-art algorithms.","PeriodicalId":285628,"journal":{"name":"Proceedings of the 2018 International Conference on Artificial Intelligence and Pattern Recognition","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114626119","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：