Airport Detection Based on Improved Faster RCNN in Large Scale Remote Sensing Images
- PDF / 1,674,879 Bytes
- 13 Pages / 439.37 x 666.142 pts Page_size
- 27 Downloads / 345 Views
Airport Detection Based on Improved Faster RCNN in Large Scale Remote Sensing Images Shoulin Yin1 · Hang Li1 · Lin Teng1 Received: 14 July 2020 / Revised: 31 August 2020 / Accepted: 29 September 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract The extensive acquisition and using of high-resolution remote sensing images have greatly promoted the development of airport detection. However, due to the complex shape, background and different scale of the airport location, the real-time and accuracy of airport detection are also facing major challenges. These factors can reduce the detection accuracy. To solve the above problems, we propose an improved faster region-based convolutional neural network (RCNN) detection method for airport detection in large scale remote sensing images. Multi-scale training is applied to faster RCNN to enhance the robustness of network for detecting airport with different sizes. Meanwhile, we adopt the modified multitask loss function to improve the accuracy of airport detection. Online hard example mining strategy is introduced to decrease the redundant negative samples in the training process. Then the non-maximum suppression method is used to remove the redundant boxes of the detected airport. Finally, we conduct sufficient experiments with the airport data obtained from Google Earth and make comparison with the state-of-the-art airport detection methods. The results show that the proposed method can accurately detect different airports under complex background with high detection rate, low false alarm rate and short running time. Keywords Airport detection · Faster RCNN · Multi-scale training · Loss function · Online hard example mining strategy · Non-maximum suppression
* Hang Li [email protected] * Lin Teng [email protected] Shoulin Yin [email protected] 1
Software College, Shenyang Normal University, Shenyang 110034, China
13
Vol.:(0123456789)
49
Page 2 of 13
Sensing and Imaging
(2020) 21:49
1 Introduction Typical object detection in remote sensing image is one of the research hotspots in the field of image processing [1]. As an important military and civil infrastructure, airport plays an important role in aircraft landing, transportation and energy supply. In the military field, the airport has many strategic functions such as energy supply, transit and aircraft parking, which is an important object to be struck by the enemy. In the civil field, airport, as an important node of transportation energy allocation, plays a very significant role in promoting economic development. However, airport detection also faces many problems and challenges, such as different airports shapes, complex backgrounds, interference with urban roads and other ground objects, vast remote sensing image data [2]. Airport detection and recognition in remote sensing images can provide reference for the understanding of the whole image and application, which is a good information extraction point. Therefore, the research of airport search, detection, localizat
Data Loading...
