Biometric Analysis | Forensic ...

Biometric is the science that deals with identification of individuals based on a person’s physical and behavioral attributes. It is the most advanced technology used in modern world for identification of an human using fingerprints, DNA, face, voice, or iris.

 Basic perspective of biometric authentication is that every person can be accurately found by his or her physical or behavioral traits.

Biometric systems make use of fingerprint, hand geometry, iris, retina, face, hand-vein, facial thermograms, signature, or voice print to identify a person.

Quality of a biometric system is affected by two factors; Authenticity of a sensor used, Degree of freedom offered by features extracted from sensed signals. 

Development of Biometric Technology

The first application of biometrics was the Automated Fingerprint Identification System (AFIS) implemented in 1960, then in 1980s DNA profiling came into play and gradually the development of CCTVs, mobile phone and other electronics which introduced face, iris, fingerprint recognition became an important part of a biometric tool.

Components of a Biometric System

Sensor Setup:The device acquires all types of raw biometric features of a particular attribute and quality is increased by using scanning and camera device, and collected information is stored in the database.

Feature Extraction: Further, after assessing the quality of raw data, a template is formed by subjecting raw data to ‘signal enhancement algorithm’ to improve its quality and then this data is processed and set of traits are extracted to represent the unique features, this is stored in a database and known as a template.

Matching Module: In this module, templates found or recovered are compared with stored templates, and matching score between two is given, and based on the matching score identity of a person is established.

System Database: This database is a storage system of biometrics, templates extracted from raw data along with some biographic information of the person is stored here permanently on that database.

Identification and Verification Modes of Biometric Analysis

Identification and Verification modes of biometric analysis help in solving the crime or coming to a conclusion in any decision.

In identification mode, the system makes one too many comparisons to identify the extracted data from the crime scene matches any of the stored data, whereas, in verification mode, one to one comparison is made, when the comparison between certain identity that matches takes place.

Working of Biometrics

Biometrics is solely an automated system to establish the identity of a person based on various physical and behavioral attributes of an individual. It acquires these attributes and when at the time of the crime or establishing an identity of an individual, it extracts salient feature set from the data and helps to compare acquired feature set from a crime scene to that of a stored feature set in database. This procedure helps in real-time identification of an individual with a database of a complete set of information, and hence provide a forensic scientist with a proper result of comparison.

Fingerprint Biometrics

Fingerprint is one of the most ancient means used to identify suspect during criminal investigations, because of its unique and robust nature. 

A fingerprint is formed from the pattern of ridges and valleys on surface of a fingertip. This print left at the crime scene, is used by the fingerprint technician who scans it and uses the algorithms of biometric system and marks all the minute characteristics in the print. Then they are matched in the stored fingerprint database. IAFIS and AFIS are two such systems that aid in criminal investigation and response in real-time.

DNA Biometrics

Deoxyribose nucleic acid is the most vital and at same time common evidence found at the crime scene which can be extracted from ’n’ number of bodily materials, such as hair, teeth pulp, bone, mucus, semen, and blood.

DNA biometrics uses genetic profiling also known as genetic fingerprinting, in this process following steps are taken:

1. DNA is extracted from the sample.

2. DNA is segmented into VNTRs.

3. These VNTRs are compared with stored database.

CODIS – Combined DNA Index System, this system is launched by FBI IN 1990, which is used to identify suspects by matching DNA profiles, CODIS is assisting laboratories at all level to solve crimes.

Iris Biometrics

Different iris patterns unique to individuals are made into digital templates and then compared against stored templates.

The UK government started the iris program which enables travelers to enter a country via several British airports using only automatic iris recognition for identification.

Voice Biometrics

Identification of speaker from unique characteristics of his/her voice gave rise to voice biometrics. There are various tools such as AGNITIO’s Voice ID technology designed for police to perform speaker identification, also ‘VoiceGrid Nation’ this is a system that contains advance algorithms to match voices.

These tools allow building a huge database of known criminals or persons on watch list so that it can take less time to identify the speaker and this technique has already been deployed in Mexico.

Boston Marathon Bombing Case Study

One example of forensic biometrics being utilized was in the aftermath of the Boston Marathon bombing of 2013. At the behest of the FBI, tips were pouring in with photographs and videos. Each was meticulously analyzed, and face identifications were compared to databases of terrorists or persons of interest. However, the system at the time failed to recognize the terrorists and provide identification, the Tsamaev brothers, even though government agencies had their up-to-date photographs on hand and one was involved in a terrorism-related investigation, all of which was digitally recorded. However, the biometric system at the time failed to recognize them clear in several of the photographs due to low resolution of photos and one of the terrorists wearing sunglasses. It is a challenge that modern systems to this day struggle with, but with Next Generation Information System implementation, there is potential for improvement (Saferstein and Roy). As technology improves, both visual capturing (including on commercial devices) and digital analytics algorithms meant to search and compare databases, the biometric system can potentially experience greater success.

One noted benefit of the new technology and biometrics databases is that it allows law enforcement to revisit cold cases. Recently, a 30-year-old murder case was resolved using the FBI’s Integrated Automated Fingerprint Identification System (IAFIS). A 61-year old senior was brutally stabbed in 1978 in his apartment. After which the perpetrator stole the victim’s car and escaped. Police was able to collect evidence, including latent fingerprints and palmprints in the apartment, and later, in the found care, but no new leads could be found despite fingerprints being manually compared to local and state files. In 2008, after receiving an inquiry on the case, an office at the Omaha Police Department ran the latent fingerprints through the IAFIS. After investigating possible matches, she came up with a positive identification of a known felon currently serving prison time. An investigation was opened, placing the perpetrator in the vicinity of the crime, and subsequent DNA testing proved to be a match, allowing for prosecution of the dangerous criminal placing him in prison for life (“30-Year-Old-Murder Solved”). Repeatedly, the FBI notes that cold or contested cases can significantly benefit from modern biometric systems as technology allows for a much wider and efficient search of databases than ever before.

Conclusion

In modern society, the ability to identify individuals in real time, both reliably and effectively, is the foundation to various applications of biometrics in a highly networked world.

Forensic biometrics seeks to take advantage of this, both in real-time and post event collection of evidence at a crime scene.

The biometric modalities are algorithmically analyzed in complex databases which allow for rapid identification of individuals and offenders. As technology and networks continue to improve and become prominent in society, biometric data will become central to various applications, including forensic investigations.