Method and apparatus for image processing
Inventors
JONES, Simon Mark Chave • HUTCHINSON, Nicholas Dunkley
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-10885349-B2
Publication Date
2021-01-05
Expiration Date
Abstract
A method and apparatus for detecting fine movement of a subject in video images, and for distinguishing over noise and other image artefacts. The video images are processed to detect movement of image features through the sequence and to calculate how spatially distributed those moving features are across the image. The movement tracks of the features may be subject to principal component analysis and a spatial dispersion measure calculated by the product of the distance between tracked image features and the contributions of those image features to the most significant principal components. If the spatial dispersion measure is high then this is indicative of feature movement being dispersed widely across the image, whereas if it is low, it is indicative of the main feature movements being concentrated in one part of the image, and thus more likely to represent subject movement than noise.
Core Innovation
A method of determining whether a video image of a scene contains movement of a subject within the scene acquires a sequence of image frames forming the video image. Movement of a plurality of image features is detected through the sequence of image frames to form a corresponding plurality of movement signals, and the movement signals are analyzed to find related movement signals that are likely to relate to the same movement source.
A spatial dispersion measure is calculated for the related movement signals, the spatial dispersion measure representing the spatial dispersion in the image of the related movement signals. A lowest of the calculated spatial dispersion measures is compared with a predetermined first threshold, and when the lowest of the calculated spatial dispersion measures is lower than the predetermined first threshold, the video image is determined to contain subject movement.
The analysis framework distinguishes fine subject movement from noise and artifacts by relating movement signals likely coming from a common movement source, optionally conditioning the movement signals, and combining spatial dispersion information with related-signal structure derived from blind separation or clustering/correlation. A spatial dispersion measure is computed in a way that combines feature-pair distance and principal-component scores, and fine subject movement is classified when the minimum spatial dispersion across components or clusters falls below an empirically tuned threshold.
Claims Coverage
The independent claim set described in the partial content includes one independent method claim and one independent apparatus claim. Across these independent claims, the core coverage comprises five main inventive features.
Detecting movement of image features to form movement signals
Detecting movement of a plurality of image features through the sequence of image frames to form a corresponding plurality of movement signals.
Analyzing and relating movement signals to a common movement source
Analyzing the movement signals to find related movement signals, these being signals analyzed as likely to relate to the same movement source.
Calculating a spatial dispersion measure for related movement signals
Calculating a spatial dispersion measure for the related movement signals, the spatial dispersion measure representing the spatial dispersion in the image of the related movement signals.
Threshold decision using the lowest spatial dispersion measure
Comparing the lowest of the calculated spatial dispersion measures with a predetermined first threshold, and if it is lower than the predetermined first threshold determining that the video image as containing subject movement.
Room monitoring apparatus with visible or audible indication
An apparatus that monitors a subject in a room by capturing room video, automatically processes the video to determine whether a video image contains movement of a subject in the scene, and outputs a visible or audible status/alert based on that determination.
The independent method claim is centered on detecting feature movement signals, relating those signals to the same movement source, computing spatial dispersion for the related signals, and declaring subject movement when the lowest spatial dispersion measure is below a predetermined first threshold. The independent apparatus claim implements the same determination for room monitoring with a visible or audible alert/output.
Stated Advantages
Distinguishes fine subject movement from noise/artifacts by analyzing related movement signals likely from the same movement source and using a spatial dispersion measure with thresholding.
Enables welfare/security monitoring in a room monitoring system by determining whether a subject is moving and producing an output.
Documented Applications
Room monitoring (welfare/security monitoring context) using a video camera and a video signal processor/data processor to determine whether a subject is moving in the scene and to output a visible or audible alert.
Interested in licensing this patent?