摘要:
A method for detection and/or tracking of objects in motion 16 in a scene under surveillance 15, in which besides the objects in motion 16, interfering objects and/or interfering regions—both hereinafter called interfering factors 17, 23—can occur, is proposed, in which the scene under surveillance 15, a plurality of regions are defined that are divided up into various region classes; and a first region class D1, 5 includes sensitive regions, in which no and/or only insignificant interfering factors 17, 23 are located and/or are to be expected; and for detection and/or tracking of the objects in motion 16 in the sensitive regions, a sensitive content analysis is performed, and a second region class D2, 6 includes semi-sensitive regions 19, in which interfering factors 17, 23 are located and/or are to be expected, and for detection and/or tracking of the objects in motion 16 in the semi-sensitive regions 19, a semi-sensitive content analysis is performed, which is limited and/or modified compared to the sensitive content analysis.
摘要:
The invention relates to video surveillance systems that are used, for example, for surveying public places, stations, streets, industrial estates, buildings or similar. Said video surveillance systems comprise one or more surveillance cameras that are oriented towards surveillance scenes and transfer image data streams, in the form of image sequences, to an evaluation center. The invention also relates to a device (1) for identifying and/or classifying a movement pattern in an image sequence of a surveillance scene comprising a plurality of moving objects, an interface (2) for recording the image sequence, a calculation module (5) for determining an optical flow field (10) in the surveillance scene by evaluating the image sequence and an identification module (6) that is designed in terms of programming and/or circuitry such that the optical field and/or partial areas thereof are compared to one or more patterns in order to identify the movement pattern in the image sequence.
摘要:
In daily life, people are often forced to join a queue in order, for example, to pay at a checkout or to be dealt with at an airport, etc. Because of the various forms of a queue, these are not usually recorded automatically, but are analyzed manually. For example, if a long queue is formed at a supermarket, as a result of which the predicted waiting time for the customers rises above a threshold value, this situation can be identified by the checkout personnel, and a further checkout can be opened. A device 1 is proposed for identification of a queue 2 of objects 10 in a monitoring area, having an interface 6 which can be connected to an image source 7, with the interface 6 being designed to observe at least one monitoring image 3 of the monitoring area of the image source, wherein the monitoring image 3 shows a scene background of the monitoring area with possible objects 10, having an evaluation device 5 which is designed to identify the queue 2 of the objects 10 in the at least one monitoring image, wherein the evaluation device 5 has an object detector module 8 which is designed to detect a plurality of objects 10 on the basis of the monitoring image 3, wherein the plurality of the detected objects 10 forms the basis for identification of the queue 2 of the objects 10, wherein the object detector module 8 is designed to identify the objects 8 in the monitoring image with the scene background and/or wherein the object detector module 8 has content-sensitive detectors 9 for detection of the objects 10.
摘要:
Video surveillance systems are used for monitoring such areas as streets, railroad stations, industrial plants, buildings, or squares with video cameras, and on the basis of the sequences of images recorded with the video cameras, discovering anomalies in the monitored areas. A method for detecting image regions that are conspicuous in terms of the movement in them in a sequence of images of a scene and a corresponding apparatus and computer program are proposed, in which a current optical flow field in the sequence of images is calculated that includes motion vectors of image regions of the scene; an optical main flow field is determined that includes main flow directions of image regions of the scene; and a counterflow field with motion vectors of image regions that are conspicuous in terms of the movement in them is formed from image regions of the scene whose motion vectors are oriented differently from the main flow directions.
摘要:
In addition to the clear advantages of video monitoring systems for securing monitoring regions and optionally for following suspicious objects, there is the requirement to secure the private environments of people in the regions being monitored. A masking module (4) for a monitoring system (1) is disclosed for the above, wherein the monitoring system (1) has at least one monitoring camera (2) designed and/or arranged for observing monitoring regions with moving objects (14, 16), comprising a selection device for selecting objects as selected objects (16), wherein the masking module (4) is designed to output the selected objects (16) or partial regions thereof subsequently together called selected objects (16) in a masked form, wherein the masking module (4) is designed to limit the masking of objects (16) to at least one selected physical partial region (15, 18) of the monitoring region.
摘要:
Video monitoring systems are commonly used for monitoring streets, junctions, squares, and also public buildings, museums, prisons. Video monitoring systems which output the results for the viewing areas/objects in the scene in metric units, such as lengths in meters and speed in meters per second, or the absolute and/or relative positioning in a camera network, require knowledge of camera parameters such as e.g. the position, orientation, and aperture angle of the camera(s). These parameters are usually manually measured at high cost during the installation process, for example special patterns (sample patterns) are used, or people walking backwards and forwards in front of the camera. A monitoring camera 1 is proposed, according to which the monitoring camera 1 is positioned and/or can be positioned for monitoring at a stationary monitoring position, with a calibration aid system 5, wherein the calibration aid system 5 is designed for determining the stationary monitoring position of the monitoring camera.
摘要:
In automatic assessment or surveillance with camera-based surveillance systems, it is of particular interest to detect moving surveillance objects, such as persons or motor vehicles, and track them over time so that if there are unusual trajectories or unusual behavior, an alarm can be tripped. An apparatus for image-based tracking of surveillance objects in front of a scene background in a surveillance scene is proposed, having an object segmentation module that is embodied for ascertaining moving object areas in front of the scene background on the basis of a comparison of one or more camera images with a reference image of the scene; having an administration module which is embodied for assigning surveillance objects to the moving object areas detecting fusion of two or more surveillance objects and marking them jointly as an object group; and having a feature detection module which is embodied in terms of programming and/or circuitry for determining local features for tracking the surveillance objects of the object group in the image areas defined by the moving object areas.
摘要:
A surveillance camera is positioned and positionable at a stationary surveillance position for monitoring. The surveillance camera has a calibration tool that is constructed or configured for ascertaining the stationary surveillance position of the surveillance camera.
摘要:
In daily life, people are often forced to join a queue in order, for example, to pay at a checkout or to be dealt with at an airport, etc. Because of the various forms of a queue, these are not usually recorded automatically, but are analyzed manually. For example, if a long queue is formed at a supermarket, as a result of which the predicted waiting time for the customers rises above a threshold value, this situation can be identified by the checkout personnel, and a further checkout can be opened. A device 1 is proposed for identification of a queue 2 of objects 10 in a monitoring area, having an interface 6 which can be connected to an image source 7, with the interface 6 being designed to observe at least one monitoring image 3 of the monitoring area of the image source, wherein the monitoring image 3 shows a scene background of the monitoring area with possible objects 10, having an evaluation device 5 which is designed to identify the queue 2 of the objects 10 in the at least one monitoring image, wherein the evaluation device 5 has an object detector module 8 which is designed to detect a plurality of objects 10 on the basis of the monitoring image 3, wherein the plurality of the detected objects 10 forms the basis for identification of the queue 2 of the objects 10, wherein the object detector module 8 is designed to identify the objects 8 in the monitoring image with the scene background and/or wherein the object detector module 8 has content-sensitive detectors 9 for detection of the objects 10.
摘要:
The invention relates to a monitoring system (1) for detecting and/or tracking at least one moving monitoring object (4) in a monitoring area (2), having a plurality of monitoring sensors (6) which are and/or can be spatially distributed in the monitoring area (2) and which are designed to detect the monitoring object (4) and to output detected object information, having a signature lock (8a, b, c, d) having at least one signature sensor (9) which is designed and/or arranged to detect the monitoring object (4) and to output object signature information, having an evaluation device (10) which is designed to create and/or add to an object signature (16) of the monitoring object (4) based on the object signature information and to detect and/or track the monitoring object (4) by comparing detected object information to the object signature (16) of the monitoring object (4), wherein the signature lock (8a, b, c, d) comprises at least two signature sensors (9) and is designed in such a way that the object signature information of the at least two signature sensors (9) can be associated with the monitoring object distinctly and/or unmistakably.