Innovation in the CCTV images to improve accuracy of facial recognition systems to identify criminal suspects

Published Date | 2018 September 25

The researchers from University of Lincoln and York in UK have discovered a method that digitally combines multiple CCTV images into a single enhanced image, which will help in accurate face recognition.

United Kingdom: The researchers have found that a person and a computer system are better at identifying a face when it views an average image that includes combination of multiple pixelated images, as compared to the original poor-quality image. Thus, they created a series of pictures with the use of “face averaging” technique, that digitally combines multiple images into a single enhanced image by removing other variants like head angles or any king of lighting to focus only on the features and identity of the person captured in the image. This discovery will have a positive impact on the residential security market because CCTV cameras are one of the major devices used in home security.

According to Dr Kay Ritchie, who led the research on “face averaging” technique, all CCTV systems have the luxury of high-quality cameras, meaning that face identifications are often being made from poor quality images. She further added that, “there are lots of different techniques that people can use as investigative tools to improve low-quality images, such as manipulating brightness.” Thus, the discovery of this “face averaging” method has the potential to help with suspect identification even from low-quality CCTV footage which include images from multiple different cameras, for instance, in case of tracking a suspect along a particular route. As per the study, when participants were asked to compare a high-quality image with a low-quality pixelated image and with the image created using the “face averaging” method, and to identify whether both the images depicted the same person or two different people.

The results revealed that face recognition accuracy was significantly high in case of the comparison between high-quality image and image created via “face recognition” technique. Furthermore, when the study was conducted using computerized systems, it also revealed higher levels of accuracy in identifying a person from average images created by the method. Thus, the results revealed by the research will have positive implications for law enforcement and security systems, where low quality, pixelated images are often the only pictures of suspects available to recognize the suspect.

According to BlueWeave Consulting, in the era of technological advancements that comprises Internet of Things (IoT), use of artificial intelligence, adoption of cloud-based security technologies, the demand for advanced security solutions have also increased manifold. This demand includes security systems like security cameras or CCTV cameras, along with alert systems and wireless locks. Thus, there is accelerated demand for the home security market, across the globe. Among all the products used for residential or home security, security cameras have the largest market as a result of the growth of security equipment industry, backed by the increasing digitalization. Security cameras can support in developing a safe and secure home environment via high-quality audio/visual vigilance. Thus, the discovery of this “face averaging” method, that can provide accurate face recognition, will strengthen the market demand for security systems including security cameras.

According to a recently published report of BlueWeave Consulting, on “Global Residential Security Market by Product (Security cameras, Smart locks, sensors, security alarms, Fire sprinklers and Extinguishers), by Solution (Audio-Visual Surveillance, Control & Management, Home Integrated Security, Alert System, Intercom System, Fire Protection) by Home Type (Independent and Apartments) Size and Forecast to 2014-2024”, the global residential security market is expected to rise significantly over a CAGR of 8%. The market is driven by the factors like advance security solutions, being offered by home builders, traditional security companies, and by contractors, for the protection of new home. In addition to this, the adoption of IoT technology offers remote monitoring of home and other property to users anywhere anytime, as per their convenience. Thus, the residential security market is growing over the years with surging investment and innovations in this market.