Xin Huang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xin Huang has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Hao Bai
College of Information Engineering, Hainan Vocational University of Science and Technology, Haikou, Hainan, China
1 shared publication - Jianfeng Jiang
Westlake Institute for Optoelectronics, Hangzhou, Zhejiang, CHINA
1 shared publication - Lifeng Jiang
College of Information Engineering, Hainan Vocational University of Science and Technology, Haikou, Hainan, China
1 shared publication - Meixia Chen
College of Information Engineering, Hainan Vocational University of Science and Technology, Haikou, Hainan, China
1 shared publication - Qianyu Liu
College of Information Engineering, Hainan Vocational University of Science and Technology, Haikou, Hainan, China
1 shared publication - Zhiguo Zheng
School of Information and Communication Engineering, Hainan University, Haikou, Hainan, China
1 shared publication
Latest Publications
- Open Access
- 18 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 270-274
Research on marine litter identification model based on convolutional neural network
The problem of marine litter is becoming more and more serious, which poses a great threat to the marine ecological environment. The research of marine litter identification and classification problem is imminent, and the identification and classification of images are especially critical. The advantage of convolutional neural network, one of the deep learning models, lies in its ability to automatically learn the features in the image, thus improving the accuracy of image classification. Therefore, the study of marine litter recognition model based on convolutional neural network is of some significance. Collecting and pre-processing multiple types of marine litter image data, extracting features in terms of colour, shape and texture, constructing a marine litter recognition model based on convolutional neural network, training and analysing the comparison results, effectively verifying the accuracy and precision of the model. The proposed model has high accuracy and stability in marine litter identification, which can provide effective technical support for marine environmental protection as well as a technical basis for the protection of marine ecosystems.
- 18 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 270-274
Open AccessResearch on marine litter identification model based on convolutional neural network
The problem of marine litter is becoming more and more serious, which poses a great threat to the marine ecological environment. The research of marine litter identification and classification problem is imminent, and the identification and classification of images are especially critical. The advantage of convolutional neural network, one of the deep learning models, lies in its ability to automatically learn the features in the image, thus improving the accuracy of image classification. Therefore, the study of marine litter recognition model based on convolutional neural network is of some significance. Collecting and pre-processing multiple types of marine litter image data, extracting features in terms of colour, shape and texture, constructing a marine litter recognition model based on convolutional neural network, training and analysing the comparison results, effectively verifying the accuracy and precision of the model. The proposed model has high accuracy and stability in marine litter identification, which can provide effective technical support for marine environmental protection as well as a technical basis for the protection of marine ecosystems.
- Open Access
- 18 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 775-780
A Modeling Framework for the Digitization of Intelligent Transportation Standards
Traditional document-based standards can no longer meet the high-quality development needs of intelligent transportation, making the digitization of standards an inevitable trend. This paper aims to develop a framework for achieving machine-readable intelligent transportation standards by analyzing the evolution and forms of digital standards in intelligent transportation. A model-based systems engineering (MBSE) method is proposed to construct a digital modeling framework for intelligent transportation standards. This framework is designed to transform traditional document-based standards into digital formats. It comprises three stages: formal modeling, structured modeling, and information reengineering modeling. To validate the effectiveness of the modeling framework, this paper applies the framework to a case study involving the development of a Level 3 Conditional Automated Driving system based on the national standard "Automotive Driving Automation Grading" (GB/T 40429-2021). The results demonstrate that the proposed framework is a viable solution for digitizing intelligent transportation standards, offering valuable insights and guidance for future research and practical applications.
- 18 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 775-780
Open AccessA Modeling Framework for the Digitization of Intelligent Transportation Standards
Traditional document-based standards can no longer meet the high-quality development needs of intelligent transportation, making the digitization of standards an inevitable trend. This paper aims to develop a framework for achieving machine-readable intelligent transportation standards by analyzing the evolution and forms of digital standards in intelligent transportation. A model-based systems engineering (MBSE) method is proposed to construct a digital modeling framework for intelligent transportation standards. This framework is designed to transform traditional document-based standards into digital formats. It comprises three stages: formal modeling, structured modeling, and information reengineering modeling. To validate the effectiveness of the modeling framework, this paper applies the framework to a case study involving the development of a Level 3 Conditional Automated Driving system based on the national standard "Automotive Driving Automation Grading" (GB/T 40429-2021). The results demonstrate that the proposed framework is a viable solution for digitizing intelligent transportation standards, offering valuable insights and guidance for future research and practical applications.