Abstract

Aircraft recognition has great application value, but aircraft in remote sensing images have some problems such as low resolution, poor contrasts, poor sharpness, and lack of details caused by the vertical view, which make the aircraft recognition very difficult. Especially when there are many kinds of aircraft and the differences between aircraft are subtle, the fine-grained recognition of aircraft is more challenging. In this paper, we propose a non-locally enhanced feature fusion network(NLFFNet) and attempt to make full use of the features from discriminative parts of aircraft. First, according to the long-distance self-correlation in aircraft images, we adopt non-locally enhanced operation and guide the network to pay more attention to the discriminating areas and enhance the features beneficial to classification. Second, we propose a part-level feature fusion mechanism(PFF), which crops 5 parts of the aircraft on the shared feature maps, then extracts the subtle features inside the parts through the part full connection layer(PFC) and fuses the features of these parts together through the combined full connection layer(CFC). In addition, by adopting the improved loss function, we can enhance the weight of hard examples in the loss function meanwhile reducing the weight of excessively hard examples, which improves the overall recognition ability of the network. The dataset includes 47 categories of aircraft, including many aircraft of the same family with slight differences in appearance, and our method can achieve 89.12% accuracy on the test dataset, which proves the effectiveness of our method.

Highlights

  • With the development of space technology, the remote sensing image has become an effective means to survey and monitor resources, environment, urban layout, and traffic facilities, playing an increasingly important role in these fields

  • In order to observe the impact of non-locally enhanced module and part feature fusion (PFF) method on classification results separately, we conduct an ablation experiment

  • A non-locally enhanced feature fusion network is designed for the remote sensing image dataset with 47 categories of aircraft

Read more

Summary

Introduction

With the development of space technology, the remote sensing image has become an effective means to survey and monitor resources, environment, urban layout, and traffic facilities, playing an increasingly important role in these fields. As a subtask of remote sensing image processing, aircraft recognition is of great practical demand and application value. Our research is a fine-grained recognition task of aircraft in remote sensing images, which is very challenging. Aircraft recognition in remote sensing images is more difficult than in ordinary optical images. Remote sensing images of aircraft are acquired at different times and on different platforms, and the light condition, atmospheric transparency, and sensor performance will cause great differences in the imaging effect. (3) In addition, due to the limitation of the vertical view, many details of aircraft in the vertical direction are occluded. These factors will make it difficult to recognize the type of aircraft

Methods
Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call