Guest blog: How the cloud can improve biometric identification systems
31 October 2014 13:59 GMT

Cloud-based biometric system reduce costs, but there are often security and connectivity issues.

By Tanvir Ahmed

Automated Fingerprint Identification Systems (AFIS) have already become a widely accepted platform in many law enforcement agencies for identifying criminals in one to many (1:N) fingerprint search environments against registered latent prints.

Modern law enforcement agencies are now implementing an advanced version of AFIS with multimodal capabilities (called Automated Biometric Identification Systems or ABIS) to improve their ability to track and identify national security threats.

ABIS and cloud ABIS

ABIS offers a more reliable and accurate multimodal biometric identification system, because it can combine any biometric modality, including fingerprint, finger vein, palm vein, palm print, iris, facial, voice and DNA. These systems expand well beyond traditional AFIS architectures that usually rely on a unimodal system, most often fingerprint recognition.

Cloud-based ABIS systems are the most recent and flexible update to biometric identification technology, and offer a wide range of advantages to advance the adoption of biometrics in different verticals to achieve different goals.

Cloud-based ABIS opens up new opportunities for governments to utilize the power of biometric identification management. Government agencies/departments that can use this platform include election commissions, customs and border protection, and national identity and health departments.

Biometrics in the cloud

Cloud-based biometric technology can utilize cloud services through a web-based user interface, which can either be a browser or a mobile application. The basic layout of any biometric identification system is more or less same for any platform or modality.

Although both the software part and the biometric database are moved to the cloud, there are a number of aspects which are specific to a cloud-based system - such as real-time and parallel processing capabilities - that can make it more appealing.

Figure 1: Biometrics on a cloud platform

Cloud computing is a highly active field of research and development. Cloud-based ABIS can be used on a local client (e.g. on the user’s computer) as well as on mobile devices to capture individual biometric traits.

However, there are both advantages and potential drawbacks of cloud-based ABIS systems:

In terms of benefits, access to the ABIS application will likely be faster than existing platforms. Hosting in the cloud doesn’t require scheduled maintenance or software updates, and there is typically a reduction of system complexities.

Drawbacks include the fact that cloud-based systems may not comply with local legislation. There are also legal, privacy, and data protection concerns linked to the storage of biometric data.

Cost comparison

Figure 2: Hypothetical cost comparison 

Automated Biometric Identification Systems previously cost millions of dollars to install and were originally only used by central governments. Moving ABIS systems to the cloud significantly reduces these costs.

Cost-wise, this new technology doesn't require any bulk database engines to store biometric data, which would reduce the initial setup and maintenance costs. Traditional servers capable of storing 10 million individual biometric traits usually cost hundreds of thousands of dollars, and also requires thousands of dollars for maintenance per year.


Cloud-based ABIS have an enormous potential market value to governments and law enforcement agencies all around the world. ABIS deployments in cloud-based environments could cut deployment cost more than 35% compared to traditional server based systems, making their deployment more feasible to governments and large organizations. 

However, cloud-based ABIS also face challenges. For example, the reality of possible downtime makes users dependent on the reliability of Internet connections. Secondly, there are security issues. Storing biometric templates and subject records in the cloud may trigger privacy concerns and data protection issues, all representing major challenges for the development process.

Tanvir Ahmed is a digital marketing analyst currently working at M2SYS. He has published articles on biometric identity management technology.

Related articles

Cloud-based face and speaker recognition APIs now on Okta