Application of Minnan Folk Light and Shadow Animation in Built Environment in Object Detection Algorithm
Autor(en): |
Sichao Wu
Xiaoyu Huang Yiqi Xiong Shengzhen Wu Enlong Li Chen Pan |
---|---|
Medium: | Fachartikel |
Sprache(n): | Englisch |
Veröffentlicht in: | Buildings, 23 Mai 2023, n. 6, v. 13 |
Seite(n): | 1394 |
DOI: | 10.3390/buildings13061394 |
Abstrakt: |
To resolve the problems of deep convolutional neural network models with many parameters and high memory resource consumption, a lightweight network-based algorithm for building detection of Minnan folk light synthetic aperture radar (SAR) images is proposed. Firstly, based on the rotating target detection algorithm R-centernet, the Ghost ResNet network is constructed to reduce the number of model parameters by replacing the traditional convolution in the backbone network with Ghost convolution. Secondly, a channel attention module integrating width and height information is proposed to enhance the network’s ability to accurately locate salient regions in folk light images. Content-aware reassembly of features (CARAFE) up-sampling is used to replace the deconvolution module in the network to fully incorporate feature map information during up-sampling to improve target detection. Finally, the constructed dataset of rotated and annotated light and shadow SAR images is trained and tested using the improved R-centernet algorithm. The experimental results show that the improved algorithm improves the accuracy by 3.8%, the recall by 1.2% and the detection speed by 12 frames/second compared with the original R-centernet algorithm. |
Copyright: | © 2023 by the authors; licensee MDPI, Basel, Switzerland. |
Lizenz: | Dieses Werk wurde unter der Creative-Commons-Lizenz Namensnennung 4.0 International (CC-BY 4.0) veröffentlicht und darf unter den Lizenzbedinungen vervielfältigt, verbreitet, öffentlich zugänglich gemacht, sowie abgewandelt und bearbeitet werden. Dabei muss der Urheber bzw. Rechteinhaber genannt und die Lizenzbedingungen eingehalten werden. |
4.84 MB
- Über diese
Datenseite - Reference-ID
10728396 - Veröffentlicht am:
30.05.2023 - Geändert am:
01.06.2023