Footprint Identification using Deep Learning
Criminals find it almost impossible to avoid leaving shoeprints behind at a crime scene as they must enter and exit from the scene.
Foot- and/or shoe-prints, either trace or impression, are often left at a crime scene irrespective of the kind of flooring the scene is made of. These are as useful as fingerprint and DNA left at crime scene. In recent times, footprints have become good forensic evidence used in crime scene investigations, as features from the sole or from shoeprints can be used to determine certain information about the individual that owns the shoe. Also features from the barefoot gives a unique identity of an individual. Organizations with safe rooms that are meant to have limited or restricted access can control access to these rooms and can identify unauthorized access into the rooms via the shoeprints from the inside attackers.
This research proposes an unobtrusive means of acquiring and recording shoeprints of anyone who enters the safe room.. For that reason, it's interesting to use footprint image in the creating of the footprint-based identification system.
In this project, the convolutional neural network training is used for deep learning classification. Convolutional neural networks are essential for deep learning and suited for image recognition.