Thursday , 22 March 2018

FACE FORWARD: The Future of Facial Recognition

Digital solutions could radically change the face of road travel in India, making it a safe, smart and comfortable experience. TrafficInfraTech speaks to a few Facial Technology solutions provider to understand the technology.

Smart roads’ or ‘Smart highways’ has managed to gain ground in the developed world that are intended to be both interactive and largely self-powering. The concept holds tremendous potential for India considering its road network being the second largest in the world and continues to be the most important means of transport. This large road asset can be leveraged for a lot more than it currently is, implementing technology innovations that yield significant improvements in the driving experience.

Today, we see futuristic concepts like hyperloops and biometrics but with these seismic shifts comes the demand for infrastructural changes to support the new reality and maximize productivity. Reportedly, the Indian biometric market value is expected to grow from $823.46 million in 2014 to $2.06 billion in 2020 at an estimated CAGR of 16.47% from 2015 to 2020. India has started to witness biometric devices being deployed in law enforcement, local security, and travel & immigration sectors.

How does the technology work?

In facial recognition, the software scans specific critical points of the face, such as retina, nose, lips and ears. These are points that cannot change in an instance, and thus ensure that changes to the face over a period of time do not impact the recognising capabilities. Once these points are identified, a proportionate map of the face, which is unique for a person, is generated. This map is identical to the one used in fingerprint identification, where the value is calculated based on the differences in shapes of the rings of the finger.

Based on this map, a complex algorithm is generated with a specific value, unique for each face. The first such algorithm generated is registered in the device. The next time a face comes in front of it, the device checks if it matches with the registered face and unlocks if it is a match.

Explaining the special features of their Facial Technology solutions, Ashish P Dhakan, MD & CEO, Prama Hikvision said, “Deep Learning has swept through the IT industry, bringing benefits and better classifications to a number of applications. Inspired by the way the human brain works, the technology uses a layered learning process to enable the computer to classify, store and access data, which it can then refer to for learning. This means it can use a whole image to recognise, rather than relying on separate elements of that image. This is a cumulative process – the more elements it has to draw on, the better the classification – thus, the better the ‘learning’.

The benefits of this technology for face recognition and image classification make it hugely valuable in the field of security. It touches on every aspect of the security industry – from facial and vehicle detection for behaviour analysis. This, in turn, starts to change the focus of security from being reactive to being able to predict problems before they happen.

There is a need for strict norms for passenger safety and logistics compliance. Another challenge is delay in implementing transport related security projects. Infrastructure projects are beset by time and cost overruns emanating from delay in execution, resulting in disputes between private partners and government agencies.

– Ashish P Dhakan

Hikvision’s Deep In view IP camera range and DeepInmind NVR range work together to provide all the power and benefits of Deep Learning. While the cameras provide the smart ‘eyes’ of the system, the NVR represents the analysis and storage capabilities of the brain. The products help to tackle security on two fronts – recognition, monitoring and counting of people and recognition & detection of vehicles. This uses Deep Learning technology at its most effective – for its ability to classify and recognise thousands of ‘features’. The DeepinView deep learning cameras self-learn the number plate information within a scene and recognize a larger amount of number plates than conventional ANPR systems, without the often cumbersome camera positioning adjustments. DeepinView Traffic monitoring analytics, when applied in populous areas and on highways, capture traffic violations such as running red-lights, wrong-way driving, illegal parking, and illegal U-turns, contributing to decreased traffic congestion and placing the public and vehicle passengers at lower risk of accidents.

HIKVision has successfully implemented various customized solutions for different projects in India. Dhakan says, “Hikvision solutions for Transport sector are getting good response from the end-users. Some of the significant Railway projects are implemented by us. In the Public Transport segment, we have successfully implemented various projects for many State and City Transport Authorities”. Explaining some of the challenges in integrating these cameras with other intelligent traffic systems, he adds, “The transport sector in India is fragmented

AllGoVision has the one of the most advanced Artificial intelligence enabled Facial Recognition solutions. The use of Deep Learning for Facial Recognition brings it much closer to human perception. Layered filters consider the most minute details of faces and thus the parameters considered for matching faces are greatly increased in number, thereby increasing the level of accuracy. Camouflaged faces, fugitives, criminals, etc. can be easily identified and the limitations in the earlier technologies can be removed to have a much-advanced technology. Ashwin Amarapur, Director & CEO, AllGoVision Technologies Pvt Ltd, explains, “We have optimized Facial Recognition to be robust to the field challenges. This enables the Facial Recognition to be used for complex usecases for Traffic Management. Facial Recognition can be used at the entry/ exit gates to capture and identify the facial details of drivers. Similarly, Facial Recognition can even be installed at Toll Booths on Highways. Easy pass-through for employees at the entry/ exit of office buildings/ campuses can be enabled through Facial Recognition. Also, this can be used for monitoring Taxis and other vehicles as we can integrate License Plate Recognition along with Facial Recognition. A camera with IR can see through the windscreen. Facial Recognition can be done on highways also for suspect search.”