One of the challenges faced by Diffused Metal-Oxide Semiconductor (DMOs) is how to track the behavior of tourists and provide more comfortable experience for tourists. Nowadays, multi-source tourism big data provides many available information for improving tourists’ experience. For management organizations, in order to achieve better publicity effect, how to choose the appropriate image as the representative of the destination image has become a problem. Based on the review text and image data, this paper proposes a method, Scale-invariant feature transform KMeans (SIFT-KMeans) of selecting the representative image of tourism destination. This method uses the text and image data generated by tourists to carry out a series of analysis and processing, and then feeds back the results to tourists, so as to reflect the greatest interest of tourists. The accuracy and stability of this method is wonderful, and the change of destination image can be reflected through the change of time. The destination management organization can carry out corresponding construction and publicity based on the obtained results.
{"title":"An image selection method for image representation of tourism destination based on comment text and image data","authors":"Xiaojia Huang, Yong Yang, Yezhou Yang, Chen Wang, Liang Guo","doi":"10.1109/CoST57098.2022.00012","DOIUrl":"https://doi.org/10.1109/CoST57098.2022.00012","url":null,"abstract":"One of the challenges faced by Diffused Metal-Oxide Semiconductor (DMOs) is how to track the behavior of tourists and provide more comfortable experience for tourists. Nowadays, multi-source tourism big data provides many available information for improving tourists’ experience. For management organizations, in order to achieve better publicity effect, how to choose the appropriate image as the representative of the destination image has become a problem. Based on the review text and image data, this paper proposes a method, Scale-invariant feature transform KMeans (SIFT-KMeans) of selecting the representative image of tourism destination. This method uses the text and image data generated by tourists to carry out a series of analysis and processing, and then feeds back the results to tourists, so as to reflect the greatest interest of tourists. The accuracy and stability of this method is wonderful, and the change of destination image can be reflected through the change of time. The destination management organization can carry out corresponding construction and publicity based on the obtained results.","PeriodicalId":135595,"journal":{"name":"2022 International Conference on Culture-Oriented Science and Technology (CoST)","volume":"113 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117250162","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 : 2022-08-01DOI: 10.1109/cost57098.2022.00025
Xianting Feng, Jianming Hu, Xin Zhu
With the development of the Internet industry, the information age presents a trend of “information overload”, and people's efficiency in extracting effective information is getting lower and lower. In order to relieve people's browsing pressure, this paper implements a collaborative filtering algorithm based on machine learning for the movie recommendation, citing the principle of personalized recommendation system proposed by Robert Armstrong and others in the United States in 1995. First, the rating data is preprocessed and visualized in consideration of the user's real behavior. Then implement the algorithm mentioned above, and use the test indicators to measure the performance of the recommender system and optimize the system parameters. Finally, using software engineering and Java front-end knowledge based on Spring+SpringMVC+Mybaits (SSM) to conduct demand analysis, functional analysis, non-functional analysis and establish a database. At last, use java database connectivity (JDBC) to link the database Mysql, and finally realized a movie recommender system with basic functions.
{"title":"Machine Learning Based Personalized Movie Research and Implementation of Recommendation System","authors":"Xianting Feng, Jianming Hu, Xin Zhu","doi":"10.1109/cost57098.2022.00025","DOIUrl":"https://doi.org/10.1109/cost57098.2022.00025","url":null,"abstract":"With the development of the Internet industry, the information age presents a trend of “information overload”, and people's efficiency in extracting effective information is getting lower and lower. In order to relieve people's browsing pressure, this paper implements a collaborative filtering algorithm based on machine learning for the movie recommendation, citing the principle of personalized recommendation system proposed by Robert Armstrong and others in the United States in 1995. First, the rating data is preprocessed and visualized in consideration of the user's real behavior. Then implement the algorithm mentioned above, and use the test indicators to measure the performance of the recommender system and optimize the system parameters. Finally, using software engineering and Java front-end knowledge based on Spring+SpringMVC+Mybaits (SSM) to conduct demand analysis, functional analysis, non-functional analysis and establish a database. At last, use java database connectivity (JDBC) to link the database Mysql, and finally realized a movie recommender system with basic functions.","PeriodicalId":135595,"journal":{"name":"2022 International Conference on Culture-Oriented Science and Technology (CoST)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131797876","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 : 2022-08-01DOI: 10.1109/cost57098.2022.00060
Fangfang Yu, Shengyu Jian, Chen Shen, Wen Xue, Yu Fu
In the process of digital transformation of human society, the application of Digital Human has gradually entered daily life. Limited by technology, the current virtual Digital Human is still just a digital image with only appearance but no autonomous thoughts. The degree of intelligence that drives Digital Human is the key factor restricting its subsequent development. If Digital Human is to be put into commercial application, at least it must be able to interact with people based on the scene. This paper makes a broad classification of digital humans and outlines their future evolutionary trends. It is necessary to broadly categorize Digital Human and outline their evolutionary status in the future. As a companion product of the digital transformation of society, Digital Human plays a very important role in mobilizing human resources across domains and achieving more efficient collaboration in the future.
{"title":"On the Issue of “Digital Human” in the context of digital transformation","authors":"Fangfang Yu, Shengyu Jian, Chen Shen, Wen Xue, Yu Fu","doi":"10.1109/cost57098.2022.00060","DOIUrl":"https://doi.org/10.1109/cost57098.2022.00060","url":null,"abstract":"In the process of digital transformation of human society, the application of Digital Human has gradually entered daily life. Limited by technology, the current virtual Digital Human is still just a digital image with only appearance but no autonomous thoughts. The degree of intelligence that drives Digital Human is the key factor restricting its subsequent development. If Digital Human is to be put into commercial application, at least it must be able to interact with people based on the scene. This paper makes a broad classification of digital humans and outlines their future evolutionary trends. It is necessary to broadly categorize Digital Human and outline their evolutionary status in the future. As a companion product of the digital transformation of society, Digital Human plays a very important role in mobilizing human resources across domains and achieving more efficient collaboration in the future.","PeriodicalId":135595,"journal":{"name":"2022 International Conference on Culture-Oriented Science and Technology (CoST)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130208294","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 : 2022-08-01DOI: 10.1109/cost57098.2022.00039
Jie Wang, Chang Lu
In the protection and development of architectural heritage, in addition to the traditional protection means, digital technology is even more imperative. The digitization process of cultural heritage has developed by leaps and bounds in this century. The application of digital technology in the protection and inheritance of architectural heritage has made remarkable achievements, which has brought significant changes to the protection, dissemination and development mode of architectural heritage. With the continuous improvement of the digitization of cultural heritage, we should actively carry out the digital protection of architectural heritage, and always pay attention to the development and application of digital technology, and constantly explore and practice the new forms of digital technology in the field of architectural heritage protection. In this paper, CiteSpace software is used to visualize and compare the keywords of relevant literature at home and abroad, so as to explore the research direction and research hotspot, analyze the existing research status, and put forward the limitations of the research, and introduce the application and development of a variety of high-tech means in the field of architectural heritage protection. Digital technology as an indispensable and important technical means in the field of architectural cultural heritage protection, which has attracted numerous experts and scholars to engage in research.
{"title":"Research on the Development and Practice of Digital Technology in Architectural Heritage","authors":"Jie Wang, Chang Lu","doi":"10.1109/cost57098.2022.00039","DOIUrl":"https://doi.org/10.1109/cost57098.2022.00039","url":null,"abstract":"In the protection and development of architectural heritage, in addition to the traditional protection means, digital technology is even more imperative. The digitization process of cultural heritage has developed by leaps and bounds in this century. The application of digital technology in the protection and inheritance of architectural heritage has made remarkable achievements, which has brought significant changes to the protection, dissemination and development mode of architectural heritage. With the continuous improvement of the digitization of cultural heritage, we should actively carry out the digital protection of architectural heritage, and always pay attention to the development and application of digital technology, and constantly explore and practice the new forms of digital technology in the field of architectural heritage protection. In this paper, CiteSpace software is used to visualize and compare the keywords of relevant literature at home and abroad, so as to explore the research direction and research hotspot, analyze the existing research status, and put forward the limitations of the research, and introduce the application and development of a variety of high-tech means in the field of architectural heritage protection. Digital technology as an indispensable and important technical means in the field of architectural cultural heritage protection, which has attracted numerous experts and scholars to engage in research.","PeriodicalId":135595,"journal":{"name":"2022 International Conference on Culture-Oriented Science and Technology (CoST)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126403910","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 : 2022-08-01DOI: 10.1109/CoST57098.2022.00037
Jing He, M. Zhang, Feng Xu, Qiang Liu, He Li
The kinematic parameters of camera robot are described by Unified Robot Description Format (URDF). Compared to Denavit-Hartenberg (DH) method, URDF is more flexible. But DH method has an advantage in inverse solution control at position level. Sometimes only URDF file can be obtained due to the engineering project requirements, such as modularity and security. It is easy to convert DH to URDF. There is a lack of research for inverse conversion. In this paper, three conversion principles were proposed. The relationship of base coordinate system and joint coordinate system was divided into six situations. The DH transformation parameters were solved based on the three principles. Finally the extend conversion algorithm for serial joints chain was supplied. The robot kinematic conversion algorithm from URDF to DH provides a mathematical tool for requirements of engineering projects.
{"title":"A Model Conversion Algorithm from URDF to DH for Camera Robot","authors":"Jing He, M. Zhang, Feng Xu, Qiang Liu, He Li","doi":"10.1109/CoST57098.2022.00037","DOIUrl":"https://doi.org/10.1109/CoST57098.2022.00037","url":null,"abstract":"The kinematic parameters of camera robot are described by Unified Robot Description Format (URDF). Compared to Denavit-Hartenberg (DH) method, URDF is more flexible. But DH method has an advantage in inverse solution control at position level. Sometimes only URDF file can be obtained due to the engineering project requirements, such as modularity and security. It is easy to convert DH to URDF. There is a lack of research for inverse conversion. In this paper, three conversion principles were proposed. The relationship of base coordinate system and joint coordinate system was divided into six situations. The DH transformation parameters were solved based on the three principles. Finally the extend conversion algorithm for serial joints chain was supplied. The robot kinematic conversion algorithm from URDF to DH provides a mathematical tool for requirements of engineering projects.","PeriodicalId":135595,"journal":{"name":"2022 International Conference on Culture-Oriented Science and Technology (CoST)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130703050","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}
Nowadays, the use of video editing software has increased dramatically. However, there is a problem of insufficient intelligence in the recommendation of clip templates in these software. Therefore, this paper addresses this problem and devotes to combining machine learning algorithms and deep learning to achieve optimization of video clip template recommendation, and proposes the design of a clip template recommendation system based on cross-modal retrieval technology. Firstly, the Requests module is used to crawl some data from Baidu images and NetEase cloud music websites and store them persistently as components of user templates to make the templates diverse and meet the needs of more users. Secondly, the algorithm network construction based on PyTorch framework was completed to realize background replacement and music matching, improve the template matching mechanism for users, and generate videos from images; finally, the Android Studio platform was used to develop the APP for Android system, and the Web server was built to realize the data interaction between the client side and the server side, so that users can easily use the APP to get functional experience.
{"title":"Recommendation of Clip Templates Based on Cross-Modal Retrieval","authors":"Zhiyi Zhu, Xiaoyu Wu, Xueting Yang, Kai Zhang, Haoyi Yu, Xiangshan Chen","doi":"10.1109/cost57098.2022.00071","DOIUrl":"https://doi.org/10.1109/cost57098.2022.00071","url":null,"abstract":"Nowadays, the use of video editing software has increased dramatically. However, there is a problem of insufficient intelligence in the recommendation of clip templates in these software. Therefore, this paper addresses this problem and devotes to combining machine learning algorithms and deep learning to achieve optimization of video clip template recommendation, and proposes the design of a clip template recommendation system based on cross-modal retrieval technology. Firstly, the Requests module is used to crawl some data from Baidu images and NetEase cloud music websites and store them persistently as components of user templates to make the templates diverse and meet the needs of more users. Secondly, the algorithm network construction based on PyTorch framework was completed to realize background replacement and music matching, improve the template matching mechanism for users, and generate videos from images; finally, the Android Studio platform was used to develop the APP for Android system, and the Web server was built to realize the data interaction between the client side and the server side, so that users can easily use the APP to get functional experience.","PeriodicalId":135595,"journal":{"name":"2022 International Conference on Culture-Oriented Science and Technology (CoST)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122592664","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 : 2022-08-01DOI: 10.1109/cost57098.2022.00089
Tingting Wu, Jianming Hu, Xin Zhu
Research on automatic character analysis of novels can help to achieve automatic Q & A with fictional characters. In this paper, a corpus containing 1435 novel texts was constructed with Chinese martial arts novel characters as the research object, and a total of 57026 characters were extracted. The character vectors were generated by Skip-gram model training, and the effect of applying the character vectors was explored. Similarity calculation and K-means clustering were performed on the persona vectors, and the experimental results showed that people from the same author usually have similarity. The gender classification prediction was performed using logistic regression and support vector machine for the persona vectors respectively, and the experimental results showed that both classification algorithms could predict the gender of the new sample characters well.
{"title":"Character Feature Extraction for Novels Based on Text Analysis","authors":"Tingting Wu, Jianming Hu, Xin Zhu","doi":"10.1109/cost57098.2022.00089","DOIUrl":"https://doi.org/10.1109/cost57098.2022.00089","url":null,"abstract":"Research on automatic character analysis of novels can help to achieve automatic Q & A with fictional characters. In this paper, a corpus containing 1435 novel texts was constructed with Chinese martial arts novel characters as the research object, and a total of 57026 characters were extracted. The character vectors were generated by Skip-gram model training, and the effect of applying the character vectors was explored. Similarity calculation and K-means clustering were performed on the persona vectors, and the experimental results showed that people from the same author usually have similarity. The gender classification prediction was performed using logistic regression and support vector machine for the persona vectors respectively, and the experimental results showed that both classification algorithms could predict the gender of the new sample characters well.","PeriodicalId":135595,"journal":{"name":"2022 International Conference on Culture-Oriented Science and Technology (CoST)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124839688","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 : 2022-08-01DOI: 10.1109/CoST57098.2022.00021
Yuliang Xue, Nan Zhong, Zhenxing Qian, Xinpeng Zhang
Neural style transfer (NST) is a technique based on deep learning that preserves the content of an image and converts its style to a target style. In recent years, NST has been widely used to generate new artworks based on existent styles to promote cultural communication. However, there is little research that considers the protection of copyright during the generation of stylised images. To this end, we propose an end-to-end protectable style transfer network based on steganography, called PSTNet. This network, including a pair of encoder and decoder, takes a content image and copyright information as input. The encoder embeds copyright information directly into the input content image and render the content image in a specific style. When the copyright needs to be verified, only the corresponding decoder can extract copyright information correctly. Furthermore, an elaborated designed noise layer is added between the encoder and decoder to improve the robustness of the copyright protection method. Experiments show that the protectable stylised images generated by PSTNet have significant visual effects and the undetectability of copyright information is proved by steganalysis. In addition, our method is robust enough that the copyright of generated stylised images can still be proved even after spreading on real social networks. We hope this work will raise awareness of the protection of artworks created by NST.
{"title":"PSTNet: Protectable Style Transfer Network Based on Steganography","authors":"Yuliang Xue, Nan Zhong, Zhenxing Qian, Xinpeng Zhang","doi":"10.1109/CoST57098.2022.00021","DOIUrl":"https://doi.org/10.1109/CoST57098.2022.00021","url":null,"abstract":"Neural style transfer (NST) is a technique based on deep learning that preserves the content of an image and converts its style to a target style. In recent years, NST has been widely used to generate new artworks based on existent styles to promote cultural communication. However, there is little research that considers the protection of copyright during the generation of stylised images. To this end, we propose an end-to-end protectable style transfer network based on steganography, called PSTNet. This network, including a pair of encoder and decoder, takes a content image and copyright information as input. The encoder embeds copyright information directly into the input content image and render the content image in a specific style. When the copyright needs to be verified, only the corresponding decoder can extract copyright information correctly. Furthermore, an elaborated designed noise layer is added between the encoder and decoder to improve the robustness of the copyright protection method. Experiments show that the protectable stylised images generated by PSTNet have significant visual effects and the undetectability of copyright information is proved by steganalysis. In addition, our method is robust enough that the copyright of generated stylised images can still be proved even after spreading on real social networks. We hope this work will raise awareness of the protection of artworks created by NST.","PeriodicalId":135595,"journal":{"name":"2022 International Conference on Culture-Oriented Science and Technology (CoST)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125382245","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 : 2022-08-01DOI: 10.1109/CoST57098.2022.00079
Maoyuan Xu, Tao Jia, Hongyang Zhou, Xiaobin Zhu
The conditional normalized flow model can alleviate the ill-posed nature of super-resolution problems by learning the conditional distribution of the output for a given low-resolution input, which allows multiple predictions for a given low-resolution image. However, the model may generate confusing artifacts in the high-frequency region of the image. While giving a smaller reduction in temperature would solve this problem, it would also smooth out the model's output, which would lead to a decrease in perceptual quality. In this paper, we propose a novel conditional normalizing flow-based distribution adaptation strategy for image Super-Resolution. More specifically, we demonstrate that the part of the latent variable that differs significantly between the latent variables of HR and the Bicubic of LR, contains mainly high-frequency information of the image. We adopt a Bicubic image of LR and a continuous threshold function to evaluate different temperatures in different latent variables. In this way, We can further alleviate generation of confusing artifacts without reducing perceptual quality. Extensive experiments show that our model can outperform state-of-the-art methods and generate more visually favorable results.
{"title":"Rethinking the Adaptiveness of Conditional Flow A Novel Distribution Adaptation Strategy for Image Super-Resolution","authors":"Maoyuan Xu, Tao Jia, Hongyang Zhou, Xiaobin Zhu","doi":"10.1109/CoST57098.2022.00079","DOIUrl":"https://doi.org/10.1109/CoST57098.2022.00079","url":null,"abstract":"The conditional normalized flow model can alleviate the ill-posed nature of super-resolution problems by learning the conditional distribution of the output for a given low-resolution input, which allows multiple predictions for a given low-resolution image. However, the model may generate confusing artifacts in the high-frequency region of the image. While giving a smaller reduction in temperature would solve this problem, it would also smooth out the model's output, which would lead to a decrease in perceptual quality. In this paper, we propose a novel conditional normalizing flow-based distribution adaptation strategy for image Super-Resolution. More specifically, we demonstrate that the part of the latent variable that differs significantly between the latent variables of HR and the Bicubic of LR, contains mainly high-frequency information of the image. We adopt a Bicubic image of LR and a continuous threshold function to evaluate different temperatures in different latent variables. In this way, We can further alleviate generation of confusing artifacts without reducing perceptual quality. Extensive experiments show that our model can outperform state-of-the-art methods and generate more visually favorable results.","PeriodicalId":135595,"journal":{"name":"2022 International Conference on Culture-Oriented Science and Technology (CoST)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130131112","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 : 2022-08-01DOI: 10.1109/CoST57098.2022.00040
Tianyu Hua, Shuang Wang, Jingyu Liu, Jian Jiang
This article is based on the theoretical achievements of design psychology, takes graphic poster design as the research object, and combines experimental psychology methods with information processing technology to construct a visual cognitive model that can serve art design. This article firstly starts from the laws of formal beauty in design psychology, and screens the formal beauty features of posters. Secondly, a subjective evaluation experiment for the formal beauty characteristics of the poster materials was conducted, and correlation analysis and factor analysis on the processed experimental data were completed. Finally, this article uses a variety of machine learning algorithms to construct a visual cognitive prediction model. This article summarizes seven low-level and four high-level feature descriptors of the beauty of poster form, and constructs a prediction model from low-level features to high-level features. In addition, this article also quantifies the “Balance” feature which has the highest word frequency in the low-level features, realizing the calculation of the balance degree of the poster image.
{"title":"Research on Visual Cognitive Model Modeling based on Design Psychology","authors":"Tianyu Hua, Shuang Wang, Jingyu Liu, Jian Jiang","doi":"10.1109/CoST57098.2022.00040","DOIUrl":"https://doi.org/10.1109/CoST57098.2022.00040","url":null,"abstract":"This article is based on the theoretical achievements of design psychology, takes graphic poster design as the research object, and combines experimental psychology methods with information processing technology to construct a visual cognitive model that can serve art design. This article firstly starts from the laws of formal beauty in design psychology, and screens the formal beauty features of posters. Secondly, a subjective evaluation experiment for the formal beauty characteristics of the poster materials was conducted, and correlation analysis and factor analysis on the processed experimental data were completed. Finally, this article uses a variety of machine learning algorithms to construct a visual cognitive prediction model. This article summarizes seven low-level and four high-level feature descriptors of the beauty of poster form, and constructs a prediction model from low-level features to high-level features. In addition, this article also quantifies the “Balance” feature which has the highest word frequency in the low-level features, realizing the calculation of the balance degree of the poster image.","PeriodicalId":135595,"journal":{"name":"2022 International Conference on Culture-Oriented Science and Technology (CoST)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117237338","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：