IIIT Hyderabad researchers come up with solution to store humongous CCTV cameras footage

An average CCTV camera that has a 1080 pixel resolution generates 72 MB of data per hour and 1,000 such cameras generate 72GB data per hour, which is about 2 TB in a day.

Hyderabad: The International Institute of Information Technology (IIIT) – Hyderabad researchers have come up with a solution that will not just efficiently store humongous amounts of the CCTV cameras footage but also extract and analyze footage in the real-time.

In a study titled ‘A Cloud-Fog Architecture for Video Analytics on Large Scale Camera Networks Using Semantic Scene Analysis’ that was presented at CCGrid 2023 – the 23rd IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing, IIIT-H researchers proposed a video analytics framework that checks storage and extract relevant information in the real-time. The team focused on public CCTV cameras installed on the road networks.

---

According to the researchers, an average CCTV camera that has a 1080 pixel resolution generates 72 MB of data per hour and 1,000 such cameras generate 72GB data per hour, which is about 2 TB in a day.

“With 50,000 cameras, we can see that data generated is about 100 TB in a day which is enormous,” says Kunal Jain, 5th year dual degree student in Computer Science and the primary author of the research paper.

One of their motivations was to come up with a solution to save on network bandwidth. To achieve that, the researchers focused on things primarily – semantic scene analysis and cloud-fog architecture, which refers to the way machines understand visual scenes – through object recognition – and establish relationships between objects by generating a textual description of the same. This, according to researchers, is useful when retrieving images as the textual query search throws up the appropriately captioned image.

“In our system, all visual information is stored in textual format known as Scene Description Records (SDR). For instance, in the road setting, if there’s an image of two black cars at an intersection with one of them taking a right turn because of a free right, a query for a black car at an intersection will retrieve the image,” Kunal says.

For quickly processing the information, the researchers used a fog node which is a small processing unit that processes information given and forwards it to the nearest data centre.