Effectively using CCTV to fight crimeSunday, July 02, 2017
BY KEDRIAN JAMES
Due to the spiraling crime rate, the Minister of National Security Robert Montague has been exploring numerous avenues to reduce it.
One such avenue is the use of technology, specifically Closed Circuit Television (CCTV) cameras for surveillance. In my view this is a step in the right direction. This view is supported by the wide use of CCTVs in countries such as France and the United Kingdom for similar purposes. However, such a technology should not simply be adapted and used as a passive surveillance tool. By this I mean it is simply put in place for surveillance and only when an incident occurs the recordings are replayed to aid in solving a crime.
There is more to CCTVs than just surveillance. With the use of more advanced computer science concepts such as Artificial Intelligence, Big Data Analytics and Computer Vision, the Jamaica Constabulary Force will be better able to harness the full power of CCTVs while reducing the manual labour required to actively monitor a large network of CCTVs.
Similar to how some companies have resorted to using automated telephone customer service for customer queries, the same can be done with CCTVs. This automation of CCTVs does not necessarily mean that persons hired to monitor these CCTVs will lose their jobs. This means that the burden on these individuals to monitor these systems will be less, and at the same time officers who would normally be required to monitor these systems can be reassigned to other areas. So the big question here is what is the proposed solution?
Video analytics uses artificial intelligence, computer vision and big data analytics to make video surveillance smarter and more effective. By adopting this system the JCF will not only gain the advantage of automating the process of CCTV monitoring, there is also the the added advantage of generating automated notifications about suspicious activities, as well as quickly generating more detailed reports about CCTV footage. These reports can later be used for post-incident analysis as well as fed into other systems to predict trends and crime patterns. The accuracy of these systems have been improving over the years as CCTV technology can now track movements, recognise vehicles, and the faces of people. Therefore, accuracy would not be an excuse to reject the adoption of such a system. What are the options for implementing such a system?
Implementing Video Analytics
One option that the Government could take if it decides to adopt such as system is to purchase the solution off the shelf from a provider. This option can be quite expensive as the cost of maintenance and licensing would have to be factored in. A better approach would be to start investing more in information technology research at the tertiary level by providing grant funding for students to build out these solutions and continually adding improvements. A model that the Government could adapt is that of the National Science Foundation (NSF) which funds research conducted by America's colleges and universities. According to the NSF, “We are the funding source for approximately 24 per cent of all federally supported basic research conducted by America's colleges and universities.” Such an initiative would also benefit the students by enabling them to gain more hands on experience and soft skills to better place them in the job market.
We must remember that improvements in technology are done at a fast pace. As such, the technology implemented today might become obsolete tomorrow. Therefore, let us not be a dinosaur in this area. Technology is the way to go in fighting crime, but can only reap success if it is used effectively.
Kedrian James lecturer in Cybersecurity at the School of Computing and Information Technology, Faculty of Engineering and Computing, The University of Technology, Jamaica.
Now you can read the Jamaica Observer ePaper anytime, anywhere. The Jamaica Observer ePaper is available to you at home or at work, and is the same edition as the printed copy available at https://bit.ly/epaper-login