FIELD OF INVENTION
The invention relates to a method and system for automatically adjusting display
settings in digital television (DTV) based on the distance between the viewers(s)
and the television. In particular, the invention provides an intelligent module in a
DTV that can adjust viewing based settings such as color, contrast, tint and
brightness on the basis of physical distance between the television viewer and
the television system.
BACKGROUND OF THE INVENTION
US Patent number 6, 002, 386 discloses an automatic contrast control circuit for
a display device. The contrast control circuit includes a distance detector
provided on a front panel of the display device which detects the distance
between the front panel and the viewer for generating a representative distance
signal. A micro-computer outputs a contrast gain control signal based on the
distance signal from the distance detector on an on-screen-display (OSD)
indicative of an automatic contrast control function and an OSD function. The
contrast of the display device is thus controlled automatically, depending on the
distance between the viewer and the television for better eye protection. One
disadvantage of the system is that it does not use any face detection system.

SUMMARY OF THE INVENTION
The main object of the present invention is to provide a plug-in intelligent
module for digital television which can automatically adjust the display setting
options like color, contrast, tint and brightness. This and other objects of the
invention can be achieved with the help of visual sensors for capturing the
images or view in front of the DTV at periodic intervals and comparing-display
setting ratios with the current settings. This can then be gradually changed so
that the transition is smooth for human eyes.
The visual sensor can either be an infra red camera or a low resolution CCD
camera. The system uses image segmentation techniques to extract the user /
viewer in front of the television and tries to determine the distance between the
television screen and the viewer. Image processing filters are used for
performing digital image processing including object extraction, skin
segmentation, and face localization.
The system uses motion detection, skin segmentation, face and body localization
for locating the viewer. Then a distance calculation module comprising a look up
table is used to determine the distance of the viewer from the television screen

based on the ratios determined between the sizes of the viewer face with respect
to the captured frame. From the determined distance of the viewer from the
television screen viewing based settings like color, contrast, tint and brightness
can automatically be adjusted with the help of provided means, which may be in
the form of an intelligent module.
In a preferred embodiment the present invention provides a method for
automatically adjusting display settings in a digital television (DTV), comprising
the steps of: capturing low resolution images or view in front of a DTV at
periodic intervals; comparing the captured image with the frame in buffer;
segmenting the low resolution image with the aid of k means or watershed
based on computing power available in the DTV; locating viewer face(s) by skin
segmentation over the segmented image; determining the distance of the viewer
from the DTV screen with the help of a distance calculator module; and
automatically adjusting the viewing based settings like color, contrast, tint and
brightness in accordance with the determined distance.

BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWING
The invention can now be described in detail with the help of the figure of the
accompanying drawing in which
Figure 1 shows in flow chart form the method of the present invention.
Figure 2 shows a pseudo look-up table
DETAILED DESCRIPTION
The method of the present invention is illustrated in the flow-chart of Figure 1,
which is self explanatory.
The system captures images at periodic Intervals with the aid of visual sensors.
If the captured image is a first capture (i.e. buffer is empty) low resolution image
segmentation can be performed directly. Else the frame in buffer can be
compared with the new captured image. If motion detection is found the old

frame from the buffer can be deleted and segmentation step can be carried out.
Else, if no motion is found (i.e. new captured frame is same as old frame stored
in buffer, which means that the viewers have not moved considerably) capture
module goes to sleep mode.
The low resolution image is segmented using algorithms like K means or the
watershed based on the computing power available in the DTV. Skin
segmentation is performed over the segmented image and consequently face is
located. If multiple faces are found the system takes into consideration the
biggest face assuming that it is closest to the television.
The system then predicts the human body on the basis of the average human
face size ratio to the human body which is again compared to the face size with
respect to that of the frame.
The system then passes the body position to a distance calculator module and
the system then calculates the size of a rectangular sized window around the
body. The computed size is then compared to the size of the frame. The

distance calculator module uses a look-up table (Figure 2) which stores the
viewing displaying settings based on distance. And on the basis of the results
calculated the settings are retrieved from the look-up table. The originally
captured new image is stored in the buffer as old frame. The capture module
goes in a pre determined sleep mode and after its completion the control goes
back to image capture step.
The settings are gradually applied to the television such that the changes are not
abrupt for the human eye. The factory pre stored look up table has two main
fields viz., ratio and visual settings. The visual settings can be further divided
into color, contrast, brightness and other possible visual display settings. A
higher human face-captured frame ratio value in the look up table corresponds
to the viewer sitting at a large distance from the screen and a lower human face-
captured frame ratio value in the look-up table corresponds to a smaller
distance. The look up table is illustrated in Figure 2. The ratio field can also be
submitted with a depth calculation algorithm output value. If the sensor
supports automatic depth calculation then a modified look up table can
automatically be used without the need of segmentation.

WE CLAIM
1. A method for automatically adjusting display settings in a digital television
(DTV), comprising the steps:
- capturing low resolution images or view in front of a DTV at
periodic intervals;
- comparing the captured image with the frame in buffer;
- segmenting the low resolution image with the aid of k means or
watershed based on computing power available in the DTV;
- locating viewer face(s) by skin segmentation over the segmented
image;
- determining the distance of the viewer from the DTV screen with
the help of a distance calculator module; and
- automatically adjusting the viewing based settings like color,
contrast, tint and brightness in accordance with the determined
distance.
2. The method as claimed in claim 1, wherein the captured low resolution
image is segmented directly in the event of the buffer being empty.

3. The method as claimed in claim 1, wherein after comparing captured
image with the earlier image in buffer, if motion detection is found the old
frame in the buffer is deleted and segmentation step is carried out.
4. The method as claimed in claim 3, wherein after comparing captured
image with the earlier image in buffer, if motion detection is not found,
the capture module goes to a pre-determined sleep mode.
5. The method as claimed in claim 1, wherein the biggest face located by
skin segmentation step is taken into consideration, assuming it to be
closest to the DTV, where multiple faces are located.
6. The method as claimed in claim 5 wherein the human body is predicted
on the basis of average human face size ratio to the human body and
comparing the body to the face size with respect to that of the frame.

7. The method as claimed in claim 1, wherein determining the distance of
the viewer from the DTV Screen comprises the steps of:
- determining the size of a rectangular sized window around the body;
- comparing the determined size to the size of the frame; and
- determining the distance of the viewer from the DTV Screen with the
help of a distance calculator module.

8. The method as claimed in claim 7, wherein the viewing displaying settings
(based on the distance) stored in the look-up, table are retrieved on the
basis of results of the determined distance.
9. A system for automatically adjusting display settings in a digital television
(DTV) comprising:
- a visual sensor for capturing images or view in front of the DTV at
periodic intervals;

- processing filters for digital image processing, like object extraction,
skin segmentation and face localization;
- a distance calculator module for determining the distance of the
viewer from the television set; and
- means for automatically adjusting the settings like color, contrast,
tint and brightness according to the determined distance.
10.The system as claimed in claim 9, wherein said visual sensor is an infrared
camera.
11.The system as claimed in claim 9 wherein said visual sensor is a low
resolution CCD camera.
12.The system as claimed in claim 9, wherein said distance calculator module
comprises a look up table.

13.The system as claimed in claim 9, wherein said means for automatically
adjusting the viewing based settings is an intelligent moldule to adjust
color, contrast tint and brightness according to the determined distance.
14. A method for automatically adjusting display settings in a digital television
(DTV), substantially as herein described and illustrated in the
accompanying drawings.

The main object of the present invention is to provide a plug-in intelligent module for digital television which can automatically adjust the display setting options like color, contrast, tint and brightness. This and other objects of the invention can be achieved with the help of visual sensors for capturing the images or view in front of the DTV at periodic intervals and comparing-display setting ratios with the current settings. This can then be gradually changed so that the transition is smooth for human eyes. The visual sensor can either be an infra red camera or a low resolution CCD
camera. The system uses image segmentation techniques to extract the user / viewer in front of the television and tries to determine the distance between the television screen and the viewer. Image processing filters are used for
performing digital image processing including object extraction, skin segmentation, and face localization. The system uses motion detection, skin segmentation, face and body localization
for locating the viewer. Then a distance calculation module comprising a look up
table is used to determine the distance of the viewer from the television screen based on the ratios determined between the sizes of the viewer face with respect to the captured frame. From the determined distance of the viewer from the television screen viewing based settings like color, contrast, tint and brightness
can automatically be adjusted with the help of provided means, which may be in the form of an intelligent module. In a preferred embodiment the present invention provides a method for
automatically adjusting display settings in a digital television (DTV), comprising the steps of: capturing low resolution images or view in front of a DTV at periodic intervals; comparing the captured image with the frame in buffer;
segmenting the low resolution image with the aid of k means or watershed based on computing power available in the DTV; locating viewer face(s) by skin segmentation over the segmented image; determining the distance of the viewer from the DTV screen with the help of a distance calculator module; and automatically adjusting the viewing based settings like color, contrast, tint and brightness in accordance with the determined distance.