Progress made to information and communication technologies opens up the possibility for more smart city integration across the world. Proper infrastructure is needed for this to become a reality including data processing and high performing surveillance cameras. An important first step for human activity recognition (HAR) is establishing systems capable of face detection from distance while maintaining function in adverse weather. Human activity recognition used in smart cities can help curve crime rates and be used for purchasing and establishment check-ins. Cities that have already integrated these systems have seen decreases in crime rates and at the height of the COVID-19 pandemic assisted in infection tracking to control spread. Surveillance cameras equipped with SWIR (Short-Wave Infrared) image sensors can not only be used for safety and crime prevention but can actively be used for traffic monitoring, accident and pedestrian detection, and fume detection.
The use of facial recognition systems in public spaces is vital to ensuring the safety of a city’s inhabitants. Consistent monitoring can help prevent acts of crime and violence by immediately alerting authorities when an act is being committed and properly identifying perpetrators. Now that smart cities have more access to data servers and processing, AI methods are more accessible and feasible to assist in real life situations. Issues affecting human activity recognition systems to be properly utilized in smart cities is their performance in long distances and low resolutions especially in adverse conditions (haze, fog, etc.), lowering recognition accuracy.
Creating a robust and cohesive surveillance system requires image sensors capable of detection in visual impairing settings to properly identify citizens, distinguish fumes, object identification, and monitor traffic as it ebbs and flows throughout the day. SWIR wavelengths can penetrate through differing weather types because of their insusceptibility to scattering to provide a clearer image, granting surveillance systems the ability to identify objects such as a crossing pedestrian or vehicles. Along with its penetrability, longer wavelengths can distinguish fumes that may look similar to the naked eye due to absorption and reflectiveness, aiding in identifying hazardous leaks.
SeeDevice’s QMOS™ image sensor provides the perfect solution for smart cities to be integrated with high performing surveillance systems to provide much-needed data. Equipped with SWIR sensing capabilities coupled with fast integration time, high dynamic range, and low costs, the QMOS™ sensor can be easily retrofitted into existing camera systems to meet budget restrictions. Once integrated, surveillance systems can easily provide imaging data for smart cities to identify and monitor traffic conditions, auto accidents, traveling pedestrians, criminal activity, and fume leaks. As society continues to advance, the future smart cities of the world require enhanced solutions aimed at improving the safety and health of its citizens.
