Fingerprint Image Identification for Crime Detection using Convolutional neural networks
Dadithota Ganesh, Dasika Akshitha, C. Gayathri, S. Sujana · 2022 3rd International Conference for Emerging Technology (INCET) · 2022
Fingerprint photographs taken at the site of a crime are critical clues in solving serial crimes. We demonstrate a comprehensive fingerprint recognition system for crime scenes using Convolutional Neural Networks (CNN). Images are retrieved from the crime scene and preserved in the database utilising methods ranging from simple physical processing techniques to advanced physiochemical processing techniques. Pre-processing the fingerprint pictures requires the use of appropriate enhancing techniques. The characteristics of prepossessed information are used as input to train the classifier the CNN. After training, it compares and determines whether a picture is a crime or not depending on its accuracy. Our suggested technique was tested using SOCOFing, FVC2004 DB1, and images gathered and saved as dataset names. It will get recognition rate of 97 percent our proposed models.