Automatic Classification and Recognition of Spatiotemporal High-resolution Image Data Based on Deep Neural Networks

Authors

  • Yetong Wang College of Information Engineering, Hainan Vocational University of Science and Technology, Haikou 571126, Hainan, China
  • Guoz Hang College of Information Engineering, Hainan Vocational University of Science and Technology, Haikou 571126, Hainan, China
  • Kongduo Xing College of Information Engineering, Hainan Vocational University of Science and Technology, Haikou 571126, Hainan, China
  • Rayner Alfred Fakulti Komputeran Dan Informatik, Universiti Malaysia Sabah, Sabah, Malaysia

Abstract

Due to the huge volume of data obtained from spatiotemporal high-resolution images, traditional data classification methods cannot meet the requirements. This paper applied an automatic classification and recognition method to such data based on a deep convolutional neural network (DCNN). This method used the convolutional neural network (CNN) to extract the spatial and temporal features of the image, and combined the full connection layer for classification and recognition. Finally, the Softmax classifier was used to complete the classification of high-resolution images. For the experiments, 21 high-resolution remote sensing image (RS) datasets were selected as the research subjects. For the experimental analysis, the classification accuracy, precision, and recall of different classification algorithms were tested. Finally, the recognition rate of the neural network was adjusted by setting different expected errors. The data showed that the DCNN algorithm achieved a classification accuracy of 100% for the spatiotemporal high-resolution RS of shrubs and intersections, and the average precision and recall of DCNN were between 0.90 and 0.95, respectively. The image classification time after training was also the shortest, at 0.06ms. The experimental results show that for the classification and recognition of spatiotemporal high-resolution image data, the use of DCNN can achieve better performance and effect, and improve the efficiency
and accuracy of image classification.

Keywords: Spatiotemporal High-resolution Image Data, Remote Sensing Images, Deep Convolutional Neural Network,
Image Classification and Recognition, Feature Extraction

Cite As

Y. Wang, G. Hang, K. Xing, R. Alfred, "Automatic Classification and Recognition of Spatiotemporal High-resolution
Image Data Based on Deep Neural Networks", Engineering Intelligent Systems, vol. 30 no. 4, pp. 309-317, 2024.



Downloads

Published

2024-07-01

Issue

Vol. 32 No. 4 (2024): International Journal of Engineering Intelligent Systems

Section

Article

License

The submission of a paper implies that, if accepted for publication, it will not be published elsewhere in the same form, in any language, without the prior consent of the publisher. Before publication, authors are requested to assign copyright to CRL Publishing Ltd. This allows CRL to sanction photocopying, and to authorize the reprinting of issues or volumes according to demand. Authors' traditional rights will not be jeopardized by assigning Copyright in this way, as they retain the right to reuse the material following publication, and to veto third-party publication.