Digitized railway ticket verification using facial recognition

The history of the Indian railway dates back to over 160 years. Since then, manual verification of passenger tickets has been practiced by the ticket examiners. This process is very wasteful, in terms of resources, and also tends to be inconvenient for the passengers at times. This project aims at alleviating the problems and modernizing the conventional ticket checking system by the application of computer vision techniques. The system's goal is to create a scalable and efficient approach in order to eliminate the need for carrying a ticket which reduces the usage of paper as well as improves the convenience of passengers. This digitization also provides a clean and organized system throughout the ticket verification life-cycle. With the recent advancements in the field of computer vision and the astounding improvements in the capacity of deep learning algorithms to impart the abilities of human eyes and brain to computer systems, the concept of facial recognition was applied as a solution to railway ticketing modernization. Various face detection and recognition algorithms were studied and a model was created using Haar Cascade Classifiers for human face detection and Local Binary Pattern Histogram (LBPH) method for face recognition. Using this model, this project aims at converting the ticket checking process into a digitized and paperless system, which would be much more convenient for the passengers and efficient for the ticket examiners.