A fast template matching-based algorithm for railway bolts detection
- PDF / 861,394 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 43 Downloads / 335 Views
ORIGINAL ARTICLE
A fast template matching-based algorithm for railway bolts detection Yunguang Dou • Yaping Huang • Qingyong Li Siwei Luo
•
Received: 23 May 2013 / Accepted: 10 December 2013 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Railway bolts detection is an important task in railway maintenance and some techniques based on traditional feature extraction and classification have been used in this application. However, these techniques have two critical disadvantages, i.e., manual collection of training data set and time-consuming training process; furthermore, trained classifiers are hard to generalize from a specific railway to the others. In order to overcome these problems, we propose a fast template matching-based algorithm, named FTM, in this paper. Firstly, we use a template matching method to locate the bolts with constrains of the railway geometric structure. Then, we use a nearest neighbor classifier to determine whether a bolt is in position or not. At last, we use GPU with CUDA architecture to accelerate the most timeconsuming part of FTM. The experiments demonstrate that our proposed FTM algorithm achieves the accuracy of 98.57 % in average, and the average false positive is only 0.89 %. The overall speedup of FTM by GPU is 6.11, and the most time-consuming part gets speedup of 17.73. Furthermore, FTM only need to collect several samples in a new railway without laborious training work. Keywords Feature extraction Pattern classification Railway maintenance Bolts detection Object detection GPU acceleration Y. Dou Y. Huang (&) Q. Li S. Luo Beijing Key Lab of Traffic Data Analysis and Mining, Beijing Jiaotong University, Beijing, China e-mail: [email protected] Y. Dou e-mail: [email protected] Q. Li e-mail: [email protected] S. Luo e-mail: [email protected]
1 Introduction Recently, the railway transportation is experiencing high growth rates all over the world, especially, in China. Indeed, the Railway Ministry of China plans to invest 2.5 trillion RMB to expand the railway network from 91,000 to 120,000 km in 2015, including 45,000 km high-speed railway, in the 12th-five-year-plan [1]. However, there are more serious security problems in high-speed railway than traditional one, and railway maintenance has become a more and more important issue in railway development. In the past decades, railway maintenance was usually performed by trained workers to detect rail defects and missing bolts. However, the manual operation always leads to low speed and efficiency. In order to tackle this problem, many companies [2, 3] and researchers [4–8] are interested in proposing a suitable vision-based automatic detection algorithm to perform railway inspection. In our previous work, we have developed an algorithm to detect the rail defects [4]. So in this paper, we further propose a new algorithm to implement another important rail aspect—detecting missing bolts. Although many works [2, 5–7] have been done to detect missing bolts, there are several disadvantages in them. Some geometric
Data Loading...
