Share with others

Criminal Face Detection System


Project Objective

            Criminal Face Detection System project aims to build a Criminal Face Detection system by levering the human ability to recall minute facial details. Identification of criminals at the scene of a crime can be achieved in many ways like fingerprinting, DNA matching or eye witness accounts. Out of these methods eye witness accounts are preferred because it stands scrutiny in court and it is a cost – effective method. It is possible that witnesses to a crime have seen the criminal though in most cases it may not be possible to completely see the face of the perpetrator.

            In most crimes, criminals may wear a mask which covers a majority of their face, they may first bind and blind – fold their victims or they may confine victims in a room before crimes are committed. In other cases a witness may simply notice the perpetrator in a passing glance and thus may not be able to recall many facial details. In such cases it is useful to show witnesses slices of the face like the nose, ears, lips, eyes. etc to see which triggers memory.

            The Criminal Face Detection System will be built of an existing criminal database. High resolution images from the criminal database will be sliced down into the different components and flashed in front of the witness. Witnesses may recall specific features and seeing high resolution images of the same will help identify the perpetrator of a crime. Each sliced image will be connected to a criminal profile and thus can give a vital clue in difficult criminal investigations. In the event that a witness is unable to definitively point to one criminal face, the system will also allow for the generation of a possible face for the criminal using the sliced images available in the system.

Criminal Face Detection System Snapshot

Existing system

            There is no dedicated Criminal Face Detection System to assist in facial detection of criminals rather police technicians have to go through to different pictures of criminals and manually slice each picture to generate images, this will usually lead to the generation of low resolution and blurred images. Linking of each sliced image to the original image is also a herculean task. The Criminal Face Detection System is ineffective because a witness will not be able to continually peruse the different images rather they will receive a broken stream of images and randomness of the sliced image is not achievable.

Proposed system

            This Criminal Face Detection System is most effective when a witness can go through sliced images in a constant flow at one sitting. In the proposed system all the criminal images are sliced beforehand and kept ready for instant viewing. Since the images are preloaded into the system the images can be randomly viewed thus making the system more effective. Every sliced image will be linked to the original image and details of the criminal. In addition a new face can be generated by using the different sliced images to create a possible face for the criminal.


Witness: The witness module is where an eyewitness interacts with the system and identifies sliced images.

Investigator: The investigating officer or a subordinate will use this module to maintain the criminal database by adding, deleting or updating criminal records with pictures. This module can also be used to generate a possible face if a witness is unable to identify it based on the sliced images.

Admin: The admin will control the access to the system. The admin can create user ids and passwords for investigating officer and ensures authenticated access.

Software Requirements

  • Windows XP
  • Apache Tomcat Web Server
  • Oracle

Technology Used

  • Java

Hardware Requirements

  • Hard Disk – 2 GB
  • RAM – 1 GB
  • Processor – Dual Core or Above
  • Mouse
  • Printer

Download Project

Abstract Download Download Abstract
Criminal Face Detection Source Code Download Source Code

Share with others