Railways are essential to socio-economic development, serving as a critical piece of infrastructure that requires continuous monitoring and upkeep to maintain safety and reliability. Traditional approaches have largely depended on manual inspections and maintenance, necessitating direct physical interaction with railway infrastructure for its installation, maintenance, and repair. Yet, recent advancements in technology have opened the door to more sophisticated, data-driven methods that enhance the safety and precision of railway maintenance. This paper presents a comprehensive approach for monitoring railway tracks and identifying and classifying anomalies in track gauge. By applying a sequence of nonlinear transformations to LiDAR scan data, we align it within a cohesive coordinate system for precise track gauge measurement. We propose an algorithm for detecting abnormal gauges, leveraging both global and local detection techniques. The findings from our case study affirm that the methodology outlined in this study is capable of accurately identifying both global and local gauge irregularities, offering automated classification and in-depth analysis of these discrepancies. This significantly supports the maintenance and operational management of railway systems.