Communications: electrical – Condition responsive indicating system – Specific condition
Reexamination Certificate
2001-02-26
2004-11-02
Pham, Toan N. (Department: 2632)
Communications: electrical
Condition responsive indicating system
Specific condition
C340S565000, C348S169000, C382S103000
Reexamination Certificate
active
06812835
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to intruding object monitoring techniques, and more particularly to an intruding object monitoring method and an intruding object monitoring system capable of monitoring a plurality of intruding objects with a plurality of monitoring apparatus.
In a recent intruding object monitoring system, some intruding object monitoring apparatus using an image pickup device such as a camera as an image input means does not rely upon a manual monitoring method by a guard man, but automatically detects an intruding object in a monitor area, controls a pan/tilt head of the camera to always capture the intruding object in the view field of the camera in accordance with the detected position of the intruding object, and changes the camera direction and zoom setting to issue a predetermined notice or alarm, to record the image of the intruding object, and the like.
A monitoring apparatus, particularly an apparatus using two cameras, is disclosed in JP-A-5-334572 and JP-A-6-153051.
SUMMARY OF THE INVENTION
In the description of the specification, “tracking” and “following” are defined as in the following.
“Tracking”: A process of monitoring by detecting or measuring the positional change of an object on an image taken in the view field of an image pickup device such as a monitor camera.
“Following”: A process of monitoring an object by controlling the pan/tilt head of an image pickup device (that is, controlling of the direction of the camera optical axis).
First, in order to describe problems associated with a conventional intruding object monitoring system, a method of monitoring a tank yard with two types of TV cameras such as shown in
FIG. 8
will be described.
Referring to
FIG. 8
, a first camera
802
is an image input means for sequentially picking up an image in the whole monitor area
804
at predetermined times. A first monitoring means including this first camera
802
detects an intruding object
801
from a plurality of frame images picked up at different times, and supplies the detection information of the intruding object
801
to a second camera
803
. The second camera
803
is an image input means for automatically performing following of the intruding object
801
and sequentially taking the images thereof at predetermined times, in accordance with the detection information of the intruding object
801
supplied from the first camera
802
.
Referring again to
FIG. 8
, by using a plurality of frame images taken with the first camera
802
at different times, differences between luminance values of each pixel are calculated and an area having a large difference is detected as the intruding object. When the intruding object
801
is detected, the first camera
802
controls the pan/tilt head of the second camera
803
in accordance with the detected position of the intruding object. In this manner, the intruding object
801
can be captured in the view field
805
of the second camera
803
.
FIG. 9
is a diagram illustrating the principle of an intruding object detecting method to be used by the first camera
802
. This method called a frame subtraction method has been used widely.
Referring to
FIG. 9
, a subtractor
909
calculates the difference between luminance values of each pixel in an input image
901
taken at time t
0
−
1
and an input image
902
taken at time t
0
, and outputs a difference image
904
between the input images
901
and
902
. Similarly, a subtractor
910
calculates the difference between luminance values of each pixel in the input image
902
taken at time t
0
and an input image
903
taken at time t
0
+
1
and outputs a difference image
905
between the input images
902
and
903
.
Next, a binarizing unit
911
generates a binarized image
906
by setting “0” if the luminance value of each pixel in the difference image
904
is smaller than a predetermined threshold value Th, and, for example, “255” (assuming eight bits of the luminance value of each pixel) if the pixel luminance value is equal to or larger than the threshold value Th. Similarly, a binarizing unit
912
generates a binarized image
907
from the difference image
905
by the method described above.
Next, a logical product unit
913
calculates a logical product between the luminance values of each pixel in the binarized images
906
and
907
and outputs a logical product image
908
. With the above-described operations, human-like objects
914
,
915
and
916
in the input images
901
,
902
and
903
are calculated as difference areas
917
and
918
by the subtractors
909
and
910
. The binarizing units
911
and
912
derives images
919
and
920
as clusters of luminance values “
255
”. The logical product unit
913
detects an image
921
as a cluster of luminance values “
255
” of each pixel in both the images
919
and
920
. In this manner, the image
921
is detected as an intruding object.
Instead of the frame subtraction method, other methods may also be used if they can detect an intruding object in a monitor area.
As described above, when the first camera
802
detects an intruding object
801
, the first camera
802
supplies control signals to the second camera
803
in accordance with the size and position of the intruding object
801
detected with the first camera
801
. In response to the control signals, the direction (pan/tilt head direction) of a monitor view field
805
of the second camera
803
can be controlled and the intruding object
801
can be captured with the second camera
803
.
As described above, the second camera
803
starts operating when an intruding object is detected from images taken with the first camera
802
, and the zoom lens and taking direction (pan/tilt head direction) of the second camera
803
are controlled in accordance with the size and position of the intruding object detected from the images taken with the first camera
802
.
A control amount of the zoom ratio of the second camera
803
is set in the following manner. For example, if the vertical image size of the first camera
802
is 240 pixels and the vertical image size of the detected intruding object is 30 pixels, then the zoom lens control amount of the second camera
803
is set to an eightfold (=240/30) of the zoom lens control amount of the first camera
802
.
By setting the control amounts of the zoom lenses of the first and second cameras
802
and
803
in the above manner, the intruding object detected with the first camera
802
can be displayed over the whole screen of the second camera
803
.
A method of calculating a control amount of the pan/tilt head of the second camera
803
will be described with reference to FIG.
10
.
FIG. 10
illustrates how a second camera
1003
performs following of an intruding object
1001
detected in a view field
1004
of a first camera
1002
. For the purposes of simplicity, in
FIG. 10
, the first and second cameras
1002
and
1003
are set at the same position, and the center direction (optical axis direction) of the view field of the first camera
1002
is made coincident with a reference view field direction (pan/tilt head control angles: pan 0° and tilt 0°) of the second camera
1003
, paying attention only to the x-axis direction.
In
FIG. 10
, 2 W represents the number of pixels (unit: pix) in the horizontal (x-axis) direction of an input image in the whole view field (monitor area)
1004
of the first camera
1002
, &Dgr;x represents a displacement (unit: pix) in the horizontal direction of a detected intruding object from the view field center of the first camera
1002
, &thgr;w represents a half monitor angle (unit:°) of the first camera
1002
, and &thgr;x represents a pan control angle (unit:°) of the pan/tilt angle head of the second camera
1003
. A half monitor angle &thgr;w of the first camera
1002
is given by:
θ
w
=
tan
-
1
(
1
2
×
h
f
)
Eq. (1)
where h (unit: mm) represents the horizontal length of the image pickup device of the first camera, and f (unit: mm) represents a focal leng
Ito Wataru
Ueda Hirotada
Antonelli Terry Stout & Kraus LLP
Hitachi Kokusai Electric Inc.
Pham Toan N.
LandOfFree
Intruding object monitoring method and intruding object... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Intruding object monitoring method and intruding object..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Intruding object monitoring method and intruding object... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3337848