Abstract
Foreign object intrusion is a great threat to high-speed railway safety operations. Accurate foreign object intrusion detection is particularly important. As a result of the lack of intruding foreign object samples during the operational period, artificially generated ones will greatly benefit the development of the detection methods. In this paper, we propose a novel method to generate railway intruding object images based on an improved conditional deep convolutional generative adversarial network (C-DCGAN). It consists of a generator and multi-scale discriminators. Loss function is also improved so as to generate samples with a high quality and authenticity. The generator is extracted in order to generate foreign object images from input semantic labels. We synthesize the generated objects to the railway scene. To make the generated objects more similar to real objects, on scale in different positions of a railway scene, a scale estimation algorithm based on the gauge constant is proposed. The experimental results on the railway intruding object dataset show that the proposed C-DCGAN model outperforms several state-of-the-art methods and achieves a higher quality (the pixel-wise accuracy, mean intersection-over-union (mIoU), and mean average precision (mAP) are 80.46%, 0.65, and 0.69, respectively) and diversity (the Fréchet-Inception Distance (FID) score is 26.87) of generated samples. The mIoU of the real-generated pedestrian pairs reaches 0.85, and indicates a higher scale of accuracy for the generated intruding objects in the railway scene.
Highlights
Foreign objects intruding railway clearance, such as pedestrians and large livestock, are a major hazard to the safety of railway operations
The generative adversarial networks (GAN) image generating method has the problem of low quality, and has not been used generating method has the problem of low quality, and has not been used in the field of railway in the field of railway intruding object image generating
We propose a novel railway intruding object image generating method of high quality and authenticity, based on an improved conditional Deep convolutional GAN (DCGAN) (C-DCGAN), which consists of quality and authenticity, based on an improved conditional DCGAN (C-DCGAN), which consists of a generator and multi-scale discriminators
Summary
Foreign objects intruding railway clearance, such as pedestrians and large livestock, are a major hazard to the safety of railway operations. It is of great significance to detect intruding foreign objects quickly and accurately. Numerous intruding object samples are needed for detection algorithm development and testing. Foreign object intrusion events are rare in daily operation. Experiments on operating high-speed railways are not permitted. Generated railway images with intruding objects will benefit detection algorithm development and testing. Pedestrian intruding railway clearance classification algorithm based on improved deep convolutional network.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have