Whether perimeter protection, traffic monitoring, people counting in retail stores, monitoring of logistic processes or home security – RIVA offers the perfect solution for many different vertical markets. The different analytic filters like people or vehicle counting, object tracking, access control, dwell time or direction detection allow a very precise object monitoring and analyzing. Furthermore, a self-learning algorithm ignores light changes as well as repetitive movements such as swaying trees or rippling water. In order to use all benefits of the video analytics, a few simple rules should be considered, when installing the camera and configuring VCA. Thus, the precision and reliability of the video analytics can be improved significantly.
At first placing the camera is very important for the whole system. A too low placement of the camera can have a negative influence on the video analytics. Also too long or short distances between camera and objects, that should be analyzed, need to be avoided. Distant objects might still be detected, but do not provide a sufficient size to be successfully classified. Appears contrary an object directly in front of the camera, the system has not enough time to detect it properly. Additionally, the choice of the lens is very important. For example the detection of a passing car from a distance of 10 m with a focal length of 25 mm does not achieve a satisfactory result. Furthermore, light conditions have a significant influence of the precision of the analytics. Better light conditions lead to better results. Especially extreme backlighting should be avoided; as well as excessive movements. The RIVA analytics is a learning system that adapts to environmental influences, rippling water, swaying trees and shadows. Nevertheless, these disturbing influences should be within reasonable limits. If a false alarm by excessive movements cannot be avoided, the most sensitive areas should be marked and ignored.
When the camera is installed correctly, the configuration of the video analytics can start. Again, a few rules should be kept in mind, because an incorrectly configured system can possibly ignore important things or create false alarms. First, the 3D calibration should be completed. It determines the sizes and is essential for the classification and several filters. After calibration the video analytics has all references for determination of sizes depending on distances. After the calibration the classification starts. That means object sizes, depending on their area and speed, are determined. Thus, the system differentiates animals, motorcycles, cars, trucks, pedestrians or pedestrian groups. Then the selection of filters follows. Various filters or filter combinations for any desired application are offered by RIVA. It is important to know exactly, what should be achieved with the analytics and which filters lead to the desired success. Only after a thorough examination and test run, correct working video analytics and alarming can be guaranteed.
In the next part you will get to know, which cameras and video analytics filters protect your home perfectly.