Name
Title | ||
|---|---|---|
Real-Time Garbage Object Detection With Data Augmentation and Feature Fusion Using SUAV Low-Altitude Remote Sensing Images |
Abstract | ||
|---|---|---|
Recently, a number of nature reserves have been shut down because of serious pollution from tourist garbage. Garbage monitoring in high-altitude natural reserves using small unmanned aerial vehicle (SUAV) remote sensing is an important and urgent need for environmental protection. In order to help cleaners to eliminate garbage more conveniently and quickly, a novel approach is proposed to detect scattered garbage regions in real time using low-altitude remote sensing videos captured by SUAVs. First, the high-resolution, low-altitude, multitemporal remote sensing images and videos containing scattered garbage were collected through SUAV and then proposed a data augmentation method to expand the training samples. Second, the Yolov4 detection network was used to classify the scattered garbage regions. Finally, the location of the object was roughly calculated according to the altitude, flight direction, global positioning system, and digital elevation model (DEM). Then, the garbage object was marked on the video, while the object location was marked on the map. Experimental results show that the proposed method achieves a mean accuracy of 91.34% and provides better performances on the real data set compared with state-of-the-art methods. |
Year | DOI | Venue |
|---|---|---|
2022 | 10.1109/LGRS.2021.3074415 | IEEE GEOSCIENCE AND REMOTE SENSING LETTERS |
Keywords | DocType | Volume |
Remote sensing, Videos, Object detection, Training, Real-time systems, Feature extraction, Deep learning, Low-altitude remote sensing of small unmanned aerial vehicle (SUAV), real-time object detection | Journal | 19 |
ISSN | Citations | PageRank |
1545-598X | 0 | 0.34 |
References | Authors | |
0 | 6 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Weiyang Chen | 1 | 0 | 0.34 |
| Haifeng Wang | 2 | 0 | 1.01 |
| Hao Li | 3 | 0 | 0.34 |
| Quanjing Li | 4 | 0 | 0.34 |
| Yang Yang | 5 | 1960 | 104.48 |
| Kun Yang | 6 | 64 | 18.24 |