Biometric technology has contributed significantly to propelling civilisation forward, making possible a degree of accuracy rate in identification and authentication needed for society to rest its confidence in the inner workings of law enforcement, banking, healthcare, digital transactions, and elections, among other endeavors.
The most common modalities of biometric identification and authentication had so far centered around fingerprint, iris, and face. Yet the impetus to take things just a little further has prompted researchers to investigate other less obtrusive modalities.
What is gait recognition biometrics?
Human gait recognition is one such unobtrusive biometric modality that is gaining attention.
Unlike biometrics using fingerprint, face, vein pattern, or iris scans which measure physical biometric traits, gait recognition is a behavioral biometric modality that identifies people based on their gait dynamics and motion features.
How does gait recognition biometrics work?
While general walking movements appear the same to the naked eye, there are minute variances in timing
Gait recognition technology is hinged on the idea that individuals have a unique and nuanced way of walking that can be recorded, measured, and analysed by gait recognition software for human identification and authentication purposes. While general walking movements appear the same to the naked eye, there are minute variances in timing and magnitude from one person to another.
Such gait sequences and variations are unique, making human gait patterns strong biometric identifiers to differentiate individuals' gait characteristics and motion features.
Gait analysis
Gait analysis considers gait shape and gait dynamics such as step length, step width, speed, and cycle time along with kinematic factors such as rotations of the joints in the hip, knee, and ankle, mean joint angles of the hip, knee, and ankle and thigh, trunk and foot angles.
Other factors accounted for are the relationship between step length and the height of an individual.
Gait recognition techniques
In contrast to well-known biometric devices such as fingerprint scanners and facial recognition terminals, the three main types of gait recognition techniques are those based on the automatic analysis of video imagery from video cameras or wearable sensors, the second method uses radar, and the third uses sensors embedded in smartphones and other gadgets.
- This first approach involves artificial intelligence analysis of video samples of a subject’s walk and the swing of joints and angles. A mathematical model of the individual motion is created and is subsequently compared against other samples to determine identity.
- The second method uses a radar system, which records the gait patterns which are then compared to other samples to perform human identification.
- The third utilises accelerometers and gyroscopes that can be used to record the user’s walking style.
The benefits of gait recognition biometrics
Already, studies are being made to test the suitability of gait recognition in actual use cases such as hardening smartphones against identity theft using accelerometers and gyroscopes.
Continuous data generation
This method has shown an ability to rapidly identify users with a high degree of confidence and prevent data leaking
One big benefit of this is that the user does not need to perform any extra action as the needed data is continuously generated while the person walks normally. This unobtrusiveness of gait-based authentication makes it a viable and powerful way to identify and authenticate individuals, especially when paired with other biometric modalities for persistent and continuous multimodal protection for smartphone mobile devices.
Indeed, this method has already shown an ability to rapidly identify users with a high degree of confidence and prevent sensitive data from being leaked.
Pressure-based gait-authentication systems
Another interesting use case is where pressure-based gait-authentication systems are installed on the floor to capture natural footsteps which are then used to distinguish individual gait patterns. This biometric modality allows for physical access control that is completely natural and unobtrusive.
Unlike most biometric traits which require formal user registration and require users to actively present themselves for inspection, gait samples for registration can be acquired silently and unobtrusively making the whole process painless for both staff and those seeking to get into the facilities. With everything being done inconspicuously, the method would be unknown to would-be imposters, thereby deterring attacks.
Conclusion
Biometric identity and authentication have come a long way in engendering countless innovations and conveniences into modern life and are still ceaselessly evolving.
With new modalities such as gait recognition rapidly developing and set to be ready for prime time not too long from now, identity and authentication continue to be at the forefront of technologies that work to improve societies.
- Network / IP
- Biometrics
- Digital video surveillance
- Electronic security systems
- Application security
- Physical security
- Industrial security
- Hospital security
- Commercial security
- Healthcare security
- Private sector security
- Public sector security
- Security management
- Security devices
- Security installation
- Security cameras
- Covert cameras
- Security camera systems
- Security monitoring system
- Security access systems
- Radio frequency Identification
- Voice recognition systems
- Facial recognition systems
- Iris recognition systems
- Network monitoring
- Video analytics
- Electronic access control
- Identity management
- Fingerprint recognition
- Fingerprint reader
- Fingerprint scanner
- Public security
- Industrial security systems
- Industrial surveillance
- Human identification system
- Covert Surveillance
- Testing & Approvals
- Corporate Security
- Central Monitoring
- Digital Video Recorders
- Warning Devices
- Incident Management
- Artificial intelligence (AI)
- Smart home
- Mergers & Acquisitions
- Smart Cities
- Machine Learning
