<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>In the field of computer vision development, the core problem is how to obtain high-level semantic information from the lowest level of original video data, and make a correct understanding. At present, the research topics proposed at home and abroad are mainly focused on the visual analysis of moving objects. Especially for image research in automobile field, visual analysis of moving objects is one of the most common topics. Its focus is to use computer vision technology to detect moving objects, and describe and understand them after tracking and recognizing relevant user portraits and behaviors. In this paper, taking vehicle driving posture preference as an example, linear dimension parameters were used instead of human joint Angle parameters to simply present the driving posture of vehicle users, and corresponding dimension parameters were obtained to present the driving posture characteristics of vehicle users. By inviting 50 drivers to participate in the test of driving attitude preference, cluster analysis was carried out on three sample data representing upper body attitude, and the clustering results were presented by combining with three-dimensional images. The final results show that the k-means clustering model can eliminate the influence of body size difference, accurately distinguish the preference characteristics of driving posture, and scientifically design steering wheel and car seat according to the measurement data of target users.</p>
</abstract>

In the field of computer vision development, the core problem is how to obtain high-level semantic information from the lowest level of original video data, and make a correct understanding. At present, the research topics proposed at home and abroad are mainly focused on the visual analysis of moving objects. Especially for image research in automobile field, visual analysis of moving objects is one of the most common topics. Its focus is to use computer vision technology to detect moving objects, and describe and understand them after tracking and recognizing relevant user portraits and behaviors. In this paper, taking vehicle driving posture preference as an example, linear dimension parameters were used instead of human joint Angle parameters to simply present the driving posture of vehicle users, and corresponding dimension parameters were obtained to present the driving posture characteristics of vehicle users. By inviting 50 drivers to participate in the test of driving attitude preference, cluster analysis was carried out on three sample data representing upper body attitude, and the clustering results were presented by combining with three-dimensional images. The final results show that the k-means clustering model can eliminate the influence of body size difference, accurately distinguish the preference characteristics of driving posture, and scientifically design steering wheel and car seat according to the measurement data of target users.

Construction and application of automobile user portrait based on k-mean clustering model

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

In the field of computer vision development, the core problem is how to obtain high-level semantic information from the lowest level of original video data,...

Construction and application of automobile user portrait based on k-mean clustering model

Online veröffentlicht: 15. Juli 2022

Seitenbereich: 909 - 918

Eingereicht: 22. Feb. 2022

Akzeptiert: 27. Apr. 2022

DOI: https://doi.org/10.2478/amns.2022.2.0077

Schlüsselwörter
K-means clustering model, User portrait, Driving attitude, Clustering analysis

© 2023 Da Wei et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Figure 1

Figure 2

Construction and application of automobile user portrait based on k-mean clustering model

Online veröffentlicht: 15. Juli 2022

Seitenbereich: 909 - 918

Eingereicht: 22. Feb. 2022

Akzeptiert: 27. Apr. 2022

DOI: https://doi.org/10.2478/amns.2022.2.0077

SchlüsselwörterK-means clustering model, User portrait, Driving attitude, Clustering analysis

© 2023 Da Wei et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Figure 1

Figure 2

Schlüsselwörter
K-means clustering model, User portrait, Driving attitude, Clustering analysis