![]() ![]() Occupancy / amount of objects in a fieldīesides the configuration for live alarming, some video analytics offer to search for recorded alarms, or to change the configuration for a full forensic search.Typical and useful features for intrusion detection are Other challenging video analytics environments including shaking camera, grass / tress moving in the wind, snow, waves.Therefore the distance at which they are detected can be less. For an object moving towards the camera, the perceived motion and change in video is much less then for an object crossing the field of view. A camouflaged object is much harder to detect than one with a good contrast to the background. At night, artificial illumination is typically needed to detect anything at all. Therefore the same object (size) should be used to compare detection distances of different video analytics. A larger object will automatically have more resolution than a small one, and will thus be detected further away. When comparing different video analytics, the same or at least similar fields of view should be used for all of them. The field of view on the other hand is given by the camera sensor size, the camera lens focal length and possible lens distortions, and the relation of the camera to the surveyed scene. Thus it also depends on the field of view of the camera. the number of false alarms per hour / day / month.ĭetection distance measures the area in which an object / alarm can be reliably detected. An often-used measure of robustness for intrusion detection is the false alarm rate, given by the amount of false alarms over time, e.g.Note that some video analytics focus strongly on reducing false alerts as much as possible, while others focus on ensuring that every intruder will be detected, or on finding a good trade-off between sensitivity and false alarm robustness. A real progress can only be achieved if both sensitivity and false alarm robustness can be kept and improved. Exchanging a focus on sensitivity or robustness for the other might make a solution workable for a specific task, but it will not result in a better performance per se. ![]() For example, a video analytics that provides large detection distances needs to be more sensitive to be able to detect objects with few pixel only, and thus has more potential to detect false objects than a video analytics that has a reduced detection range and only detects objects covered by many pixels to start with. There is usually a trade-off between the sensitivity of a video analytics algorithm ensuring the detection of all objects / alarms and its false alarm robustness, as a higher sensitivity often means more false alarms, and a higher false alarm robustness often results in less sensitivity. While a single intruder in three month is already much, video analytics can easily generate a multitude of alarms per day. The ratio of true alarms to false alarms is typically very unbalanced. Any missed alarms, on the other hand, mean the video analytics did not fulfil their task at all and intruders could enter the premises unhindered. If too many false alerts occur, then operators have been known to shut down the video analytics system completely, as they were otherwise no longer able to fulfil their monitoring tasks. In case of intrusion detection, false alerts are very time consuming and annoying and should therefore be minimized as much as possible. False positive: Object / alarm detected though there was none.īoth false alarms as well as missed alarms have to be considered in the evaluation of robustness.True positive: Object / alarm detected correctly.Robustness can be determined by counting the following three cases: In this section, the main criteria for video analytics evaluation are presented. ![]() 1 How to measure video analytics performance? ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |