Cernium Corporation

Checkvideo LLC

Ross & Baruzzini, Inc.

Maurice V Garoutte

Fengtang Wang

Michael T Fitzgerald

Jesse J Chounard, Ii

Donald K Ross

Rick D Thomason

David W McGhee

The present invention involves a system for automatically screening closed circuit television (CCTV) cameras for large and small scale security systems, as used for example in parking garages. The system includes six primary software elements, each of which performs a unique function within the operation of the security system to provide intelligent camera selection for operators, resulting in a marked decrease of operator fatigue in a CCTV system. Real-time image analysis of video data is performed wherein a single pass of a video frame produces a terrain map which contains parameters indicating the content of the video. Based on the parameters of the terrain map, the system is able to make decisions about which camera an operator should view based on the presence and activity of vehicles and pedestrians, furthermore, discriminating vehicle traffic from pedestrian traffic. The system is compatible with existing CCTV (closed circuit television) systems and is comprised of modular elements to facilitate integration and upgrades.

System for automated screening of security cameras

Some embodiments include incorporating detailed snapshots of targeted areas of interest into a video or digitized recording of an overall scene. In some embodiments, a single fixed high-resolution scene camera or a fixed standard resolution scene camera of analog or IP connected type is used. The scene camera can be co-located with at least one pan-tilt-zoom (PTZ) camera or used in a mode where less than full resolution of the scene camera is used for video but full resolution snapshots of areas are captured by the same camera. The area of interest is selected by the object-tracking feature of an intelligent video system, operated without human intervention, by electronically-implemented identification of the existence of area-specific subjects or targets of interest in the scene image field.

Directed attention digital video recordation

The present invention describes systems and methods of incorporating detailed snapshots of targeted areas of interest into a video or digitized recording of the overall scene including metadata to link the snapshots to the time and location in the overall scene from which the snapshot was acquired. A single fixed high-resolution scene camera or a fixed standard resolution scene camera of analog or IP connected type is used, co-located with at least one pan-tilt-zoom (PTZ) camera or by using the same scene camera in a mode where less than full resolution of the scene camera is used for video but snapshots of areas are captured by the same camera where higher resolution thereof is used for snapshots than for video. The area of interest is selected by the object-tracking feature of an intelligent video system, operated without human intervention, by electronically-implemented identification of the existence of area-specific subjects or targets of interest in the scene image field.

Method and system for image data analysis, by making use of a Terrain Map Summary Element (TMSE) data structure, in which a Terrain Map data structure provides processed primitive data about each 2&#xd7;2 kernel of pixels from an original image. The new TMSE is a hierarchical multiple-level or tiered structure of such Terrain Map elements, in which each successive Terrain Map level summarizes, for example, 16 lower level elements or member. By providing such a hierarchical nature of the map elements, richness is enhanced by each higher element because every element of each successively higher level map summarizes the data from elements of the lower level map. For example, in a five-level Terrain Map Summary Element structure according to the disclosure, each element or member of the fifth level contains information summarizing data about 256K pixels.

Terrain map summary elements

An object selective video analysis and recordation system in which one or more video cameras provide video output to be recorded in a useful form on recording media with reduction of the amount of the recording media, with preservation of intelligence content of the output. Video output is both background video and object video, as images of objects appearing against a background scene. Preset knowledge of symbolic categories of objects in the scene and analysis of object attributes is provided by the system. Spatial resolution and temporal resolution of objects in the scene are automatically varied in accordance with preset criteria based on predetermined interest in the object attributes while recording the background video and object video. A user of the system may query recorded video images by content, enabling the user to recall recorded data according to symbolic content, by requesting data with specified symbolic content.

Object selective video recording

Video segmentation methods and systems are described wherein scene images of subjects and background are captured by video and are automatically segmented according to subject content, as by system determination of the nature of subjects. Segmentation transformations of an original background and foreground images in the scenes are provided as a terrain map. Automatic segmentation threshold adaptation is based on measuring the segmentation noise while holding the number count of noise blobs constant, whereby to achieve noise-adaptive segmentation gain. Automatic threshold adaptation within multiple areas of the scene results in different possible thresholds in different areas based on the amount of segmentation noise in each area. Segmentation is improved accordingly.

Video analysis using segmentation gain by area

Video segmentation wherein scene images of subjects and background are captured by video and are automatically segmented according to subject content, as by system determination of the nature of subjects, but segmentation of the subjects can be compromised by visual noise. Segmentation transformations of an original background and foreground images in the scenes are provided as a terrain map, and the segmentation may also incorporate noise blobs. Automatic segmentation threshold adaptation is based on measuring the segmentation noise while holding the number count of noise blobs constant, whereby to achieve noise-adaptive segmentation gain. Automatic threshold adaptation within multiple areas of the scene results in different possible thresholds in different areas based on the amount of segmentation noise in each area. Segmentation is improved accordingly.

An exit guard system automatically detects movement of a subject along an exit path in an unauthorized direction. A first video camera is positioned for video imaging movement of a subject along the path, while a second video camera images the subject for identification. Multiple video motion detectors operate in conjunction with the first video camera to determine direction of movement of a subject along the path. A CPU operates with the video motion detectors under an exit guard program to provide a watching mode, during which frames of the second video camera are continuously grabbed and saved. Audible and/or visual alarm signaling devices are operated by the CPU in an alarm mode in response to traffic of a subject in the unauthorized direction. A video monitor displays modes of operation of the system, and control devices allow operator selection and control of modes of operation of the system. Remote access to the system for control, monitoring or other purposes is provided by data modem feature. Method of using such system for guarding movement of a subject along the path is disclosed.

Growing community of inventors

Dittmer, MO, United States of America

Maurice V Garoutte