Open AccessResearch progress and trend analysis of computer vision based on cite space
Author(s) -
Yuzhi Liu
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1941/1/012089
Subject(s) - computer science , bibliometrics , artificial intelligence , convolutional neural network , data science , deep learning , world wide web
In recent years, with the rapid development of artificial intelligence and the proposal of efficient algorithms such as deep learning, the ability of computer vision to deal with complex problems has made a qualitative leap. Therefore, the study of computer vision is deepenly expanded from academia to industry, and the application of industry continues to be enriched. It involves many disciplines in natural science and social science. The main applications and technologies include image classification, image recognition, deep learning, convolutional neural network, etc. In this paper, the bibliometrics software CiteSpace was used to conduct visualization processing based on the literatures in the core database of Web of Science with the time span from 2010 to 2020 and the subject words as computer vision. The current situation of scientific research cooperation is analyzed from the perspectives of countries, institutions, individuals and cited literatures themselves. The involvement of computer vision discipline is analyzed from the perspectives of natural sciences and social sciences in front, and analyzed from the funding institutions. Funded institutions are analyzed from the perspective of computer vision discipline. The current research hotspots and development frontiers are analyzed by using maps. It is found that the future computer vision will focus on deep learning, transfer learning and semantic segmentation. Finally, it points out that there are some problems in the current research, such as the lack of close scientific research cooperation, the lack of comprehensive discipline involved, and the difficulty of technology implementation. Three suggestions are put forward to solve these problems: (1) Holding trans-regional academic exchange meetings to promote the sharing of academic achievements. (2) Raising the attention to the partial humanities and promote the depth of disciplines. (3) Considering the cost and efficiency in practical application to accelerate the realization of scenario application.