Motorola Solutions, Inc.

Avigilon Corporation

Sergey Veselkov

Aleksey Lipchin

Maksim Bolonkin

Keshav T Seshadri

Chia Ying Lee

Xiao Xiao

Aral Sarrafi

Jeffrey Hunter

Cole Bush

Yang Liu

Examples provide a video security system including an electronic processor configured to detect an anomaly by obtaining video data captured using an image sensor and determining a number of people detected in each of a plurality of frames of the video data. For each moment t, the electronic processor determines an average number of people n over a time interval (t-Δt, t), where Δtis greater than or equal to the moment t. For each non-overlapping time interval Δt, the electronic processor updates a histogram based on the average number of people for the time interval Δt. In response to determining, using the histogram, that the probability ρ of the average number of people n over the time interval (t-Δt, t) is less than a threshold probability ρ*, the electronic processor generates an alert indicating that an anomalous crowd size is detected.

Crowd anomaly detection

Examples provide an imaging system including an electronic processor configured to obtain video data of a scene including a region of interest including a plurality of frames, determine a total number of detected humans in the region of interest of a current frame of the plurality of frames, determine a number of running humans in the region of interest of the current frame, and determine a probability associated with the number of running humans in the region of interest of the current frame and the total number of detected humans in the region of interest of the current frame based on accumulated statistics for the region of interest. In response to determining that the probability is less than a threshold probability, the electronic processor generates an alert indicating a stampede is detected in the region of interest.

Stampede detection

A method, system and computer program product for divided processing in providing object detection focus is disclosed. The system includes a first video analytics determinator configured to provide, by infrequent employment thereof within the system, a partial object detection focus in first information that informs as to where moving objects are likely and unlikely to appear within video frames in respect of video. The system also includes a second video analytics determinator configured to: informatively employ the first information to determine a refined object detection focus; and provide the refined tracking and/or object detection focus in second information.

Method, system and computer program product for divided processing in providing object detection focus

Systems and method for camera calibration. In one example, the camera includes an electronic processor configured to obtain a bounding box in a captured image, the bounding box bounding a person in the image and having a lowermost vertical coordinate and an uppermost vertical coordinate. The electronic processor determines an initial estimation of values of a set of camera parameters Θ and a current estimation of values of the set of camera parameters Θ based on the initial estimation. The electronic processor iteratively updates the current estimation of values of the set of camera parameters Θ by determining feet coordinates within the bounding box, determining a vertical head coordinate, and updating the current estimation of values of the set of camera parameters Θ by reducing a difference between the vertical head coordinate and the uppermost vertical coordinate of the bounding box.

Automatic camera calibration

Methods, systems, and techniques for automatic camera calibration. One or more calibration images are captured using a camera. The calibration images depict one or more bounding boxes, and each of the bounding boxes bounds a person. For each of the bounding boxes, the person is modeled using a rectangle or a parallelepiped, and a projection of the rectangle or parallelepiped is determined. A mapping that maps foot vertices of the projection to head vertices of the projection is determined, and using the foot vertices and the mapping, estimates of the head vertices and distances between the head vertices and the estimates of the head vertices are determined. The camera is calibrated by iteratively updating, using an objective function, the camera parameters so as to reduce those distances.

Growing community of inventors

Ashland, MA, United States of America

Sergey Veselkov