IQanalytics Vandal, Motion and Intrusion Detection User Guide 1
Contents 1 2 3 4 5 6 7 Overview Installation IQanalytics Configuration Advanced Pages Camera Placement & Guidelines Recommended Settings Event Output Format 2 3 4 5 8 9 11 14
1 Overview IQanalytics software provides an advanced solution that performs intelligent video analysis and fully automates video monitoring. It automatically tracks and identifies objects, analyzes motion and extracts video intelligence from the camera. These analytics can output real-time events and object data for video data mining or storage in a database. IQanalytics is focused on automating video analysis and security alerts, thus eliminating the need for manual work and huge monitoring costs.
2 Installation If the IQanalytics Application package is not pre-installed on your camera, manual installation will be required. Before installation, please ensure the IQanalytics.iqa application file is available. 1. 2. 3. 4. 5. 6. 7. 8. 9. Connect IQeye camera onto the local network To help detect camera on local network, use IQeye Auto run CD that is provided with the camera Click “Scan for IQeye cameras on my network” to find IP address of the camera. Type camera IP address into your browser.
3 IQanalytics Configuration The configuration of IQanalytics parameters can be adjusted using the web user interface. The IQanalytics configuration has the following sections (detailed descriptions are provided in subsequent sections): 1. 2. 3. 4.
• Object Minimum height: specifies the minimum allowable height of the object to be detected (in terms of percentage) • Object Maximum width: specifies the maximum allowable width of the object to be detected (in terms of percentage) • Object Maximum height: specifies the maximum allowable height of the object to be detected (in terms of percentage) Settings for people, vehicle and other type of objects are similarly setup.
A. Enable: This will enable Camera Tamper Detection. Ignore Area Settings Enabling the ‘ignore area’ allows the engine to ignore processing in areas that don’t require it. This may include the ceiling, moving light sources, environmental conditions, or areas where no activity is expected. This allows processing to be reduced and lowers false alarms. A. Enable: will enable the ‘ignore area’. B. Brush: allows area selection painting to use a larger 3x3 area.
F. Sudden Change Learning: Enable this feature to automatically relearn the background if the entire scene changes by more than the specified threshold. G. Sudden Change Sensitivity: The percentage of the scene that needs to change before the sudden change learning is triggered. H. Default Settings: Resets all parameters back to default settings. I. Relearn: Manually relearn the background. 4 Advanced Pages A few hidden pages are also provided for testing and demo purposes.
5 Camera Placement & Guidelines This section provides recommended camera angles and positions. Camera Positioning Camera positioning is critical for capturing good data for video analytics. Care should be taken to avoid a) object size being set too large, and b) occlusions. Typically, cameras should be installed with one of three views: 1. Top-Down view (vertical ceiling mount) 2. Perspective or Angled view (wall or corner mount looking down) – Useful for detecting general intrusion. 3.
9. Ensure that the size of the objects that are to be tracked in the scene are not too small. Object sizes have to be at least 5-10% of the scene for consistent detection. If the objects are only a few pixels (people or vehicles), it will be difficult to detect them consistently and will cause false alarms on most systems. 10. Ensure that the object size is not too large – more than 40% of the scene. The camera should not be too close to the objects being tracked. 11.
2. Camera Tamper Detector For tamper detection, the camera can be installed in any position; top-down view, side view or front view. The maximum size of objects entering the scene should not exceed 50-60% of scene size. If the camera is used during the night, lighting reflections should be minimized to avoid false detections. Light should be uniformly distributed in the scene (avoid shadows). In addition, ensure no light sources are directed at the camera.
Adjusting Advanced Settings Some scenes require extra or advanced settings like continuous learning, shadow suppression etc. Continuous Learning This feature allows the engine to learn the scene as it changes gradually. Learning happens in every frame and gradual changes in background are captured. This feature is typically used for scenes such as: 1. Outdoor scenes with shadows 2. Indoor scenes with changing lighting conditions 3.
7 Event Output Format When an event is generated, IQanalytics will provide event metadata and object metadata associated with that event. This metadata will be included in the JPEG headers or H264 Headers. Please contact IQeye on details for parsing this data from the stream. The event information is provided in the following xml structure. Aside from the camera tamper event, every other event will be accompanied by object data that gives the ID, Location, and Size of the object triggering the event.
