Abstract
Metro barrier-detection has been one of the most popular research fields. How to detect obstacles quickly and accurately during metro operation is the key issue in the study of automatic train operation. Intelligent monitoring systems based on computer vision not only complete safeguarding tasks efficiently but also save a great deal of human labor. Deep convolutional neural networks (DCNNs) are the most state-of-the-art technology in computer vision tasks. In this paper, we evaluated the effectiveness in classifying the common facility images in metro tunnels based on Google’s Inception V3 DCNN. The model requires fewer computational resources. The number of parameters and the computational complexity are much smaller than similar DCNNs. We changed its architecture (the last softmax layer and the auxiliary classifier) and used transfer learning technology to retrain the common facility images in the metro tunnel. We use mean average precision (mAP) as the metric for performance evaluation. The results indicate that our recognition model achieved 90.81% mAP. Compared with the existing method, this method is a considerable improvement.
Highlights
With the rapid development of urban public transport in recent years, urban rail transport has become the preferred choice for many people because of its various advantages, such as high speed, punctuality, and environmental friendliness
We studied the detection of obstacles in metro tunnels using a modified Deep convolutional neural networks (DCNNs)
Since random forest has been widely used in image classification and object detection [21], we compare our model with random forest
Summary
With the rapid development of urban public transport in recent years, urban rail transport has become the preferred choice for many people because of its various advantages, such as high speed, punctuality, and environmental friendliness. The detection of obstacles in metro tunnels is performed through manual observation. An improvement is automatic detection based on object detection and recognition. There are too many metro tunnel images, and this detection method has very high requirements for the efficiency and accuracy of the algorithm. Due to the diversity of features, it is difficult to extract features standardly and find the best way to represent them [9] This problem could be solved if there were a general method to learn how to extract features automatically. A DCNN has recently developed a new kind of method for classification and recognition.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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
