Image analysis – Applications – Target tracking or detecting
Reexamination Certificate
1996-10-30
2002-08-13
Johns, Andrew W. (Department: 2153)
Image analysis
Applications
Target tracking or detecting
C382S107000, C382S123000, C382S270000, C348S086000, C348S142000, C348S152000, C348S169000, C348S122000, C348S174000
Reexamination Certificate
active
06434254
ABSTRACT:
BACKGROUND OF THE DISCLOSURE
Many computer vision systems for automatic surveillance and monitoring seek to detect and segment transitory objects that appear temporarily in the system's field of view. Examples include traffic monitoring applications that count vehicles and automatic surveillance systems for security. These systems often require different object detection and segmentation methods depending on the ambient conditions. An example of such a system is disclosed in U.S. patent application Ser. No. 08/372,924 filed Jan. 17, 1995, the disclosure of which is incorporated herein by reference.
To be effective, an image based surveillance or monitoring system must delineate or segment individual objects that appear in the field of view of an imaging device, e.g., video camera that produces one or more images of a scene within the field of view. Often the object segmentation task is accomplished by examining the difference between a current image and a “reference” image that contains only the static background of the scene. The reference image can be thought of as a representation of the scene as it would appear if no transitory objects were in view. However, such simple image differencing techniques are often insufficient for accurate segmentation, especially in the presence of common illumination artifacts. Two such artifact types are shadows cast by objects in or near the scene, and reflections from lights mounted on objects within the scene. For example, in the application domain of video vehicle detection, it is common to have difficulty delineating objects accurately and/or to detect false objects when vehicles are casting shadows or when bright reflections from headlights or taillights are present.
Previous efforts have attempted to avoid the problems of false object detection due to shadows and headlight reflections by establishing parameters of object detection in such a way that these illumination artifacts do not trigger an object detector. Unfortunately, it is very difficult to find simple absolute thresholds that achieve this goal and do not miss objects for all imaging conditions (e.g., day
ight, bright/hazy days, dusk, rain and the like). As a result, object detection is not as sensitive as desired and the system fails to detect a significant number of objects.
Therefore, a need exists in the art for an improved method and apparatus for detecting and tracking objects within a scene and, particularly, in a scene containing shadows and headlight reflections.
SUMMARY OF THE INVENTION
The invention is a method and apparatus for detecting objects from a sequence of images of a scene containing an object by using two distinct methods for object detection. One is suited for well-lit scenes (e.g., daytime), while the other is suitable for poorly-lit scenes (e.g., nighttime) where the objects have lights mounted on them. Further, the method and apparatus of the invention are adaptive to image statistics regarding the objects being detected, and can be programmed to filter out “weak” detections that lie below a certain percentile in an observed statistical distribution of image measures. The specific percentile is determined based on whether the scene has been determined to be well- or poorly-lit and whether the scene contains shadows or not.
REFERENCES:
patent: 4959714 (1990-09-01), Lo et al.
patent: 5218414 (1993-06-01), Kajiwara
patent: 5243418 (1993-09-01), Kuno et al.
patent: 5267329 (1993-11-01), Ulich et al.
patent: 5606376 (1997-02-01), Shinohara
M. Kilger, “A Shadow Handler in a Video-based Real-time Traffic Monitoring System”, IEEE Workshop on Application of Computer Vision, pp. 11-18, 1992.
D. Koller, J. Weber, J. Malik, “Robust Multiple Car Tracking with Occlusion Reasoning”, European Conference on Computer Vision, 1994, pp. 189-196.
Copy of International Search Report dated Mar. 4, 1997, from corresponding international application PCT/US96/17504.
Burke William J.
Johns Andrew W.
Le Hieu
Sarnoff Corporation
LandOfFree
Method and apparatus for image-based object detection and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for image-based object detection and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for image-based object detection and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2943319