Abstract
Object detection is a focal point in remote sensing applications. Remote sensing images typically contain a large number of small objects and a wide range of orientations across objects. This results in great challenges to small object detection approaches based on remote sensing images. Methods directly employ channel relations with equal weights to construct information features leads to inadequate feature representation in complex image small object detection tasks. Multiscale detection methods improve the speed and accuracy of detection, while small objects themselves contain limited information, and the features are easily lost following down-sampling. During the detection, the feature images are independent across scales, resulting in a discontinuity at the detection scale. In this article, we propose the multiscale context and enhanced channel attention (MSCCA) model. MSCCA employs PeleeNet as the backbone network. In particular, the feature image channel attention is enhanced and the multiscale context information is fused with multiscale detection methods to improve the characterization ability of the convolutional neural network. The proposed MSCCA method is evaluated on two real datasets. Results show that for 512 × 512 input images, MSCCA was able to achieve 80.4% and 94.4% mAP on the DOTA and NWPU VHR-10, respectively. Meanwhile, the model size of MSCCA is 21% smaller than that of its predecessor. MSCCA can be considered as a practical lightweight oriented object detection model in remote sensing images.
Highlights
T HE object detection plays a key role in remote sensing algorithms and applications
We propose the multiscale context and enhanced channel attention (MSCCA) model
The feature image channel attention is enhanced and the multiscale context information is fused with multiscale detection methods to improve the characterization ability of the convolutional neural network (CNN) [31]
Summary
T HE object detection plays a key role in remote sensing algorithms and applications. RAN et al.: LIGHTWEIGHT ORIENTED OBJECT DETECTION USING MSCCA IN REMOTE SENSING IMAGES the bounding boxes and category of each region This process greatly improves the detection speed, yet reduces the detection accuracy compared to two-stage detectors. YOLO [14] is a typical one-stage object detection method that treats object detection as the solution of a regression problem, applying a single CNN to the full image This network simultaneously predicts the bounding boxes and category for each region. One-stage detection methods employ multiscale detection that extracts multiscale feature maps from different layers of the network for predictions This does not increase the number of calculations, the small object itself has less pixel information and is lost during downsampling [30].
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have