Fault Detection Based on Multi-Scale Local Binary Patterns Operator and Improved Teaching-Learning-Based Optimization Algorithm

Abstract

Aiming to effectively recognize train center plate bolt loss faults, this paper presents an improved fault detection method. A multi-scale local binary pattern operator containing the local texture information of different radii is designed to extract more efficient discrimination information. An improved teaching-learning-based optimization algorithm is established to optimize the classification results in the decision level. Two new phases including the worst recombination phase and the cuckoo search phase are incorporated to improve the diversity of the population and enhance the exploration. In the worst recombination phase, the worst solution is updated by a crossover recombination operation to prevent the premature convergence. The cuckoo search phase is adopted to escape the local optima. Experimental results indicate that the recognition accuracy is up to 98.9% which strongly demonstrates the effectiveness and reliability of the proposed detection method.

Multi-Strategy Discrete Teaching–Learning-Based Optimization Algorithm to Solve No-Wait Flow-Shop-Scheduling Problem

“…Although the BRD values obtained in the Rec37, 39, 41 problem sets were not optimal, the algorithm still showed a better performance. It can also be seen that the average BRD of the MSDTLBO was significantly better than that of the other 30,2,36,24,42,31,27,40,41,4,35,6,10,25,20,7,12,47,19,48,15,14,44,13,18,23,39,26,17,43,1,21,33,38,37,8,28,45,32,49,9,0,5,29,11,34,3,16 comparison algorithms, except for HMM-FPA. Compared to the ARD, the MSDTLBO algorithm did not obtain optimal values on the Rec37 and 39 problem sets, while for the rest of the problem sets, its average ARD value was better than that of the other algorithms.…”

“…in various optimization problems. For example, Zhang et al [16] presented a more effective approach to detecting faults. They implemented an enhanced teaching-learning-based optimization algorithm to refine the classification outcomes at the decision level.…”

An anisotropic heat diffusion model for enhancing the extraction of underground rail track fasteners under extremely low and uneven illumination conditions

“…10,31 The LBP dictates the pixel-level binary patterns of the fastener at a local level. 32,33 As shown in Figure 4, it is straightforward to construct these descriptors based on the outlook of the fasteners from the image captured with ample illumination. These descriptors are then further fed into the training pipeline, e.g.…”

An Automatic Method for Detecting Side Frame Key Using Histogram of Oriented Gradient and Local Binary Pattern

“…So, for TFDS, fault recognition is accomplished automatically by computers or man-machines, which greatly enhances the efficiency and reliability of fault detection [5]. Generally, the main fault types include Center-Plate-Bolt losing, Safety Chain dropping, Brake-Shoe-Key losing, Side-Frame-Key losing and so on [5,6,7,8]. In these failures, the most common failure is Side-Frame-Key(SFK) losing fault.…”