Abstract
Fast and accurate airport detection in remote sensing images is important for many military and civilian applications. However, traditional airport detection methods have low detection rates, high false alarm rates and slow speeds. Due to the power convolutional neural networks in object-detection systems, an end-to-end airport detection method based on convolutional neural networks is proposed in this study. First, based on the common low-level visual features of natural images and airport remote sensing images, region-based convolutional neural networks are chosen to conduct transfer learning for airport images using a limited amount of data. Second, to further improve the detection rate and reduce the false alarm rate, the concepts of “divide and conquer” and “integral loss’’ are introduced to establish cascade region proposal networks and multi-threshold detection networks, respectively. Third, hard example mining is used to improve the object discrimination ability and the training efficiency of the network during sample training. Additionally, a cross-optimization strategy is employed to achieve convolution layer sharing between the cascade region proposal networks and the subsequent multi-threshold detection networks, and this approach significantly decreases the detection time. The results show that the method established in this study can accurately detect various types of airports in complex backgrounds with a higher detection rate, lower false alarm rate, and shorter detection time than existing airport detection methods.
Highlights
Object detection based on remote sensing images is currently a research topic of interest in the field of image processing
A variety of airport detection methods have been proposed, and they can be divided into two categories: edge-based detection [7,8,9,10,11,12] and detection based on region segmentation [13,14]
The training validation set is composed of 1000 airport images, which are obtained through data expansion of the 200 original airport images; another 100 original airport images are used as the test set
Summary
Object detection based on remote sensing images is currently a research topic of interest in the field of image processing. Zhu et al [1] proposed that deep learning has been rapidly developed in image analysis tasks, including image indexing, segmentation, and object classification and detection. Deep learning promotes the development of remote sensing image analysis. Edge-based detection focuses on the characteristics of the lines at edges, and achieves airport detection through the detection of runways This approach is fast and simple but susceptible to interference from non-airport objects with long straight-line features. Airport detection based on region segmentation focuses on the distinct structural features of airports, but such methods have significant efficiency issues that are difficult to overcome due to the problem of overlapping sliding windows. The manually designed features of this method are not sufficiently robust to account for variations derived from airport diversity
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have