Video processing system and method

Patent metadata

Field	Value
Publication number	US-9948999-B2
Application number	US-201414782466-A
Country	US
Kind code	B2
Filing date	Apr 7, 2014
Priority date	Apr 5, 2013
Publication date	Apr 17, 2018
Grant date	Apr 17, 2018

Abstract

The video processing system provides for automatic selection of video sequences showing goals in a video stream from cameras directed at a playing field. Classification is applied to image content from different foreground positions in an image. The set of classes comprising a first and second class corresponding players of different teams. Medians or averages of the image positions of the local image content assigned to the first and second class are determined. A condition is tested whether the results are on mutually opposite sides of a reference position and a distance between collected positions or distances between the reference position and the first and second collected positions exceed a threshold value. When the test shows that the condition is met, a preceding video sequence is selected, for example for display.

First claim

The invention claimed is:

1. An automated method of selecting a trigger image from series of images using a video processing system, the method comprising:
capturing image streams from a plurality of cameras directed at a same playing field, for each image in a series of images from the image stream from a first camera of said plurality, processing the image using the video processing system, wherein the processing comprises:
classifying local image content from different foreground positions in the image each to one of a set of classes, based on a property of the local image content, the set of classes comprising a first and second class;
computing, by the video processing system, a first median or an average or a second first median or an average, of all different foreground positions of the local image content assigned to the first and second class in the image respectively;
using, by the video processing system, the first median or average and the second median or average as a first collected position and a second collected position for the first and second class, respectively;
detecting whether a condition is met that (a) the first and second collected positions obtained from said computing are on mutually opposite sides of a reference position and (b) either a distance between the first and second collected positions obtained from said computing, or distances between the reference position and each of the first and second collected positions obtained from said computing, exceed a threshold value;
selecting the image as the trigger image dependent on whether said condition is met, and
in response to said selecting of the image as the trigger image, extracting, via the video processing system, a video clip containing only video data captured in a time interval preceding a time point at which the trigger image was captured, the time interval ending before a time point at which the trigger image was captured, the video clip containing a part of the image stream from a second camera of the plurality of cameras in said time interval.

2. A method according to claim 1 , wherein the selecting the image as the trigger image by the video processing system comprises detecting whether said condition is met for at least a predetermined fraction of the images for image time points in a time interval including the trigger image.

3. A method according claim 1 , further comprising determining an amount of motion related to the image using the video processing system, detecting whether a further condition is met that (c) the amount of motion is below a first threshold and/or a decrease in the amount of motion compared to a previous image exceeds a second threshold using the video processing system, and selecting the image as the trigger image using the video processing system dependent on whether said further condition is also met.

4. A method according to claim 1 , further comprising determining using the video processing system whether a first sequence of the first and second collected positions along a predetermined direction in the trigger image is a reverse of a second sequence along said predetermined direction of the first and second collected positions from a previous trigger image, and suppressing or qualifying the trigger image using the video processing system dependent on a result of said determining.

5. A method according to claim 1 , further comprising determining, using the video processing system, amounts of motion in first and second parts of the images for time points preceding the time point of the trigger image, the first and second parts showing different parts of a playing field, detecting via the video processing system whether the amount of motion in the first part exceeds the amount of motion in the second part or vice versa and qualifying the trigger image dependent on a result of said detection.

6. A method according to claim 1 , comprising generating a combined image stream from the image streams from the plurality of cameras, by switching between cameras from the plurality based on a content of the image streams, the video clip being subsequently extracted from the combined image stream.

7. A method according to claim 1 , comprising rendering the video clip on a video display, making the video clip available for display, and/or transmitting the video clip in response to selection of the image as the trigger image.

8. A method according to claim 1 , comprising selectively retaining the video clip in a video storage system in response to selection of the image as the trigger image.

9. A method according to claim 1 , comprising segmenting the image and deriving said image content from foreground regions identified by said segmenting using the video processing system.

10. A method according to claim 1 , comprising generating a combined image stream from the image streams from the plurality of cameras by switching between cameras from the plurality based on a content of the image streams, the method comprising a camera selection process to switch between the cameras, wherein the camera selection process comprises segmenting images from image streams from each of the cameras, determining image positions of the foreground segments from the images of the image streams from each of the cameras and making switches between the camera based on the image positions, wherein the segmentation performed in of said camera selection process is used as said segmenting and determining the image positions for the trigger image from the image stream for the first camera.

11. A method according to claim 1 , comprising playing the video clip and/or transmitting the video clip.

12. A video capture system, comprising a plurality of cameras, a video storage system, and a video processing system, the video processing system comprising:
an input for receiving a series of images from a first camera of said plurality of cameras;
a classifier for classifying local image content from different foreground positions in each of the images, the local image content from each different foreground position being assigned to one of a set of predefined classes, based on a property of the local image content, the set of predefined classes comprising a first and second class;
a position collector, configured to compute a first median or an average or a second first median or an average, of all different foreground positions of the local image content assigned to the first and second class in the image respectively, and to use the first median or average and the second median or average as a first collected position and a second collected position for the first and second class, respectively; and
a detector configured to detect whether a condition is met that the first and second collected positions are on mutually opposite sides of a reference position and a distance between the first and second collected positions or distances between the reference position and each of the first and second collected positions exceed a threshold value,
the detector being configured to issue a signal to the video storage system, upon detection the condition is met for a current image, to play a video clip, transmit the video clip and/or transmit messages to signal that the video clip is available,
the video storage system being configured to include in the video clip only video data captured in a time interval preceding a time point at which the trigger image was captured, the time interval ending before a time point at which the trigger image was captured, the video clip containing a part of an image stream from a second camera captured in said time interval.

13. A video capture system according to claim

Assignees

• TNO

Inventors

• Havekes Anton
• Schavemaker Johannes Gerardus Maria

Classifications

• H04N23/90

  Arrangement of cameras or camera modules, e.g. multiple cameras in TV studios or sports stadiums · CPC title

• G06T7/11

  Region-based segmentation · CPC title

• H04N21/8549 Primary

  Creating video summaries, e.g. movie trailer {(retrieval in video databases by using presentations in form of a video summary G06F16/739)} · CPC title

• G06T7/20

  Analysis of motion (motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51) · CPC title

• H04N7/181

  for receiving images from a plurality of remote sources · CPC title