Detection of environmental conditions in a sequence of images

• US 8,045,761 B2
• Filed: 05/30/2006
• Issued: 10/25/2011
• Est. Priority Date: 05/30/2006
• Status: Active Grant

• Current Assignee: Intelliview Technologies Inc.
• Original Assignee: Intelliview Technologies Inc.
• Inventors: Cai, Jun, Shehata, Mohammed, Radmanesh, Ahmad, Pervez, Muzamil Shah, Burr, Tyson, Badawy, Wael
• Primary Examiner(s): Desire; Gregory M
• Assistant Examiner(s): Allison; Andrae S
• Application Number:
  US11/444,170
• Publication Number:
  US 20070280504A1
• Time in Patent Office: 1,974 Days
• Field of Search: 382/104, 382/100, 382156-162, 382/294, 382/254, 362465-466, 340/600, 340/601, 250/208.1, 348/122, 348/143
• US Class Current: 382/104
• CPC Class Codes: G06T 2207/30236 Traffic on road, railway or crossing
• CPC Class Codes: G06T 7/254 involving subtraction of images
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G06V 20/40 in video content extracting overlay text G06V20/62; video retrieval G06F16/70; processing of video elementary streams in video servers H04N21/234; processing of video elementary streams in video clients H04N21/44

Abstract

A method for determining the presence and location of static shadows and other ambient conditions (such as glare, snow, rain, etc.) in a series of time-successive images is provided. Each image comprises a series of image elements locatable on a plane, with each element being associated with a color defined by three chromatic elements. Furthermore, each image is partitioned into a set of elements, with each element comprising one or more pixels. According to the process of the present method, the ambient conditions are detected using a mixture of processes which utilize the chromatic elements, luminance qualities and temporal characteristics of the series of images.

Claims

1. A method for shadow detection in a series of image frames, the method comprising repeating the following steps:

acquiring a reference frame containing data elements, a current frame of data elements, and a third frame of data elements;

comparing corresponding elements of the current frame and the third frame to detect elements that correspond to motion;

comparing corresponding elements of the current frame and the reference frame, eliminating from comparison elements detected to correspond to motion; and

using the comparison of corresponding elements of the current frame and the reference frame to create a static shadow map indicating which elements of the current frame correspond to shadows of objects not appearing in the current frame;

in which acquiring the reference frame, the current frame and the third frame is carried out with a digital video camera and the comparing steps are carried out in a computing system.

2. The method of claim 1 in which comparing corresponding elements further comprises creating an initial set of elements which are candidates for being shadows.

3. The method of claim 1 in which the reference frame is generated by a series of successive frames that contain no shadows.

4. The method of claim 3 in which the reference frame is generated immediately after sunrise.

5. The method of claim 3 further comprising generating the reference frame by averaging a number of frames.

6. The method of claim 1 in which comparing corresponding elements comprises:

computing a difference between the current frame and the reference frame; and

determining if the difference lies between two thresholds.

7. The method of claim 1 in which each data point of the reference frame and each data point of the current frame has color components.

8. The method of claim 1 further comprising filtering by determining elements that do not correspond to shadows.

9. The method of claim 1 carried out as part of a method of video surveillance.

10. The method of claim 9 in which the method of video surveillance is video surveillance of a roadway.

11. A non-transitory computer-readable medium having stored thereon computer programmable code which implements the method of claim 1.

12. The method of claim 1 in which comparing corresponding elements of the current frame and the reference frame is not performed when the current frame corresponds to a time after sunset and prior to sunrise.

13. A method for shadow detection in a series of image frames, the method comprising repeating the following steps:

acquiring a reference frame containing data elements, a current frame of data elements, and a third frame of data elements;

comparing corresponding elements of the current frame and the third frame to detect elements that correspond to motion;

comparing corresponding elements of the current frame and the reference frame, eliminating from comparison elements detected to correspond to motion; and

using the comparison of corresponding elements of the current frame and the reference frame to create a static shadow map indicating which elements of the current frame correspond to shadows of objects not appearing in the current frame;

in which acquiring the reference frame, the current frame and the third frame is carried out with one or more of a digital video camera, radar image, or ultra sound device and the comparing steps are carried out in a computing system.