European Commission logo
English English
CORDIS - EU research results

FOREnsic evidence gathering autonomous seNSOR

Article Category

Article available in the following languages:

Set-and-forget surveillance cameras wake themselves up and call for help

Despite their advantages, digital security cameras are inconvenient. A new surveillance model reports on crime by itself.

Society icon Society
Security icon Security

Digital surveillance cameras have come a long way in a short time. They now combine very-high–quality images with autonomous operation and convenient sizes. Therefore, such devices are widely used in law enforcement. Nevertheless, video surveillance remains costly, power hungry and complicated. Current systems are used mainly for recording and have very little capability for real-time scene interpretation. The EU-funded FORENSOR project has developed a new sensor system that addresses these problems. The sensor is small and concealable; its low-power requirements free it from electrical infrastructure. The device can operate for months. Hence, the sensor is also capable of a degree of independent operation. Smart-vision chip “Our system was developed for fast deployment in rural or remote areas,” says Dr Petros Daras, project coordinator. “For example, it might be set up on a coast to watch for drug smuggling.” The system’s low power requirements make this possible. The device does not need to be plugged in, so it can be set and left. The low power consumption is partly achieved through intelligent operation of an entirely new smart-vision chip. Most of the time, the device will remain in an ultra-low–power hibernation mode. In this mode, the chip monitors routine movement, such as ocean waves or waving tree branches, for which the device does not fully activate. The sensor was designed to achieve a very low rate of incorrect detections in this mode. Independent interpretation When the chip detects movements of greater interest, the device wakes up into a higher power mode. Then a high-level algorithm classifies large objects. For example, the sensor can recognise objects such as cars, pedestrians and boats. The algorithm also links objects to actions such as car passing by, car stopped, boat arriving at coast, and human running. “The major challenge,” adds Dr Daras, “was to embed – in a low-power, low-memory system – technologies which are very demanding in terms of computational resources, including memory and processing power.” Depending on the suspicion level of the analysis, the device either returns to hibernation mode or it raises an alarm. The sensor’s third element is a sophisticated and secure wireless communication system. This also conserves power and does not operate unless required. The communication system additionally allows remote operation and management of recorded data. Further advantages of the FORENSOR system include the device being inexpensive and adaptable. It is also robust enough for extended outdoor operation, which helps to protect any evidence recorded. The sensor and project both reflect an intention to respect ethical and legal guidelines concerning the collection of personal data. Traditional cameras store everything they see. The FORENSOR system deletes most of what it sees, significantly reducing the amount of data to be stored and processed. Project researchers hope to make the system fully commercial once the necessary funding has been obtained. The consortium is investigating options. The team applied for a patent and has developed a commercial strategy. FORENSOR’s low cost and ease-of-use means that authorities will be able to install and operate more cameras. This should allow the detection of more crime, helping to keep Europe safer.


FORENSOR, sensor, cameras, surveillance, smart-vision chip, algorithm, low-power, communication system, security, law enforcement

Discover other articles in the same domain of application