Name
Abstract | ||
|---|---|---|
Although Faster R-CNN has excellent performance in object detection, it still has some difficulties in detecting small targets and slightly overlapped targets in UAV (Unmanned Aerial Vehicle) images. Based on Faster R-CNN, this paper uses ResNet101 as a feature extractor. We increase the number of anchors from 9 to 15 in RPN so that the small targets can match more anchors and get sufficient training. Due to the increasement of anchors, this paper introduces a 1×1 convolution layer to integrate features and reduce the feature map channels. We also apply RoIAlign to avoid the misalignment caused by RoIPool. The improved model effectively increases the detection rate of small targets and slightly overlapped targets so that it can be applied to human detection under UAV. The improved model can detect small targets with a size of about 30×80 pixels on aerial images with resolution of 3840×2160 pixels. Compared with Faster R-CNN, the improved model increases AP (Average Precision) from 74.31% to 79.77% on the WILDTRACK dataset. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1109/ICSAI.2018.8599511 | 2018 5th International Conference on Systems and Informatics (ICSAI) |
Keywords | Field | DocType |
deep learning,Faster R-CNN,human detection | Object detection,Computer vision,Control theory,Computer science,Convolution,Communication channel,Artificial intelligence,Extractor,Pixel,Deep learning | Conference |
ISSN | ISBN | Citations |
2474-0217 | 978-1-7281-0121-7 | 0 |
PageRank | References | Authors |
0.34 | 5 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Hanshan Zhu | 1 | 0 | 0.34 |
| Yayun Qi | 2 | 0 | 0.34 |
| Haochen Shi | 3 | 3 | 1.10 |
| Ning Li | 4 | 145 | 48.40 |
| Huiyu Zhou | 5 | 1303 | 111.91 |