Television – Camera – system and detail – Optics
Reexamination Certificate
1999-07-16
2002-10-15
Vu, Ngoc-Yen (Department: 2612)
Television
Camera, system and detail
Optics
C348S296000, C348S364000
Reexamination Certificate
active
06466267
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to auto-focusing and, more particularly, to auto-focusing by use of a solid-state imaging device.
Auto-focusing is initiated by projecting an image through a lens onto a detection device which is made up of imaging elements. The lens is, then, focussed according to a comparison of image signals generated by the imaging elements. Since auto-focusing is performed on the basis of the detected image, it is important to accurately detect the image projected onto the detection device.
The detection device for detecting the image may be embodied, for example, as a Charge Coupled Device (CCD) linear sensor
100
shown in
FIG. 11
which is a solid-state imaging device. Fundamentally, the CCD linear sensor includes light receiving units
101
, i.e., the imaging elements, which receive the projected image and generate a pixel output signal VOUt composed of a series of voltage signals representing pixels of the image. The exposure time, that is, the time in which the CCD linear sensor is exposed to light, is determined by the level of a peak value PH
out
for the pixel voltages. Thus, the pixel output signal V
out
is employed to generate an auto-focusing signal and the signal PH
out
is employed to control the exposure time so that an output level of the pixel output signal V
out
is appropriate for auto-focusing.
The light receiving units
101
of the CCD linear sensor that receive the image comprise a sensor array
102
. It will be appreciated that the sensor array may be linear, i.e., one-dimensional, because auto-focusing does not necessarily require detecting signals for an entire, i.e., two-dimensional, image. That is, the CCD linear sensor need only receive a single line of the image and it is sufficient that the sensor array is linear. Each light receiving unit in the sensor array converts incident light into an electric charge and accumulates the electric charge during the exposure time. The amount of charge in each light receiving unit corresponds to the amount of incident light accumulated over the exposure time and these accumulated charges, therefore, represent the image along the linear sensor array.
A read-out gate controlled by a read-out gate pulse &phgr;ROG initiates the reading of the charges. The charges are read out by transferring each charge from the sensor array of light receiving units; and the transfer of charges is timed by transfer pulses &phgr;H
1
, &phgr;H
2
, which are essentially clocks that trigger the light receiving units, the read-out gate and the charge transfer register to transfer the accumulated charges. The shifted charges are transmitted serially, via charge transfer register
104
, to a charge\voltage converting unit
105
. The charge\voltage converting unit converts each accumulated charge into a corresponding voltage and stores these voltages in a buffer
106
. The buffer holds these voltages for transmission to the output terminal
107
as the output signal CCD
out
from the CCD linear sensor.
A buffer
111
stores and holds the output signal CCD
out
at an output thereof as the pixel signals V
out
. The pixel signals V
out
are a series of voltage signals representing the accumulated charge in each of the light receiving units, or pixels. Auto-focusing is achieved, for example, by comparing the signal levels of each pixel voltage in the pixel output signal V
out
. This comparison cannot be made, however, when the light receiving units receive too much light. When the light incident on the light receiving units becomes too great, for example, the accumulated charges reach their maximum and the light receiving elements become saturated. As a result, it is not possible to derive a contrast therefrom and auto-focusing is not possible.
It shall be noticed that the amount of accumulated charge for each light receiving unit varies with the exposure time. To resolve the saturation problem, therefore, the exposure time may be controlled to limit the amount of accumulated charge in each light receiving unit. The exposure time is controlled by, for example, controlling a shutter speed of the CCD linear sensor. One way to obtain the exposure time is to monitor the peak value for a given series of pixel voltages and modify the exposure time for the next cycle if the peak value is outside an acceptable range.
The circuit of
FIG. 11
provides a peak detecting unit
114
which detects and holds the peak value using a diode
112
and a capacitor
113
. A buffer
115
stores the peak value for transmission to a buffered output PH
out
. As shown in
FIG. 12
, the peak value PH
out
follows the peak values for the pixel voltages V
out
, thus yielding the overall peak value for all of the pixel voltages.
The peak detecting unit
114
of
FIG. 11
may be constructed as the MOS transistor circuit shown in FIG.
13
. The pixel signals CCD
out
of the CCD linear sensor (
100
,
FIG. 11
) are received as an input signal V
in
. The input pixel signals V
in
are sent to a source follower circuit
121
comprising an input stage of N-channel MOS transistors Q
51
and Q
52
. The input stage functions as a buffer, generates the signal V
out
and forwards the same to a peak detecting unit
122
. The peak detecting unit is comprised of a diode Q
53
and a capacitor C; the diode Q
53
is formed by a P-channel MOS transistor with its gate connected to its source. The peak detecting unit
122
detects the peak value PH
out
of the pixel signals V
out
and forwards this peak value to an output stage
123
. The output stage is another source follower circuit and is composed of N-channel MOS transistors Q
54
and Q
55
which buffer the peak value PH
out
.
The devices of
FIGS. 11 and 13
are disadvantageous because they cannot determine the correct exposure time accurately. When the difference in light intensity-between a target object to be focused and surrounding objects is great, for example, the contrast is too high to detect and, as a result, the correct exposure time cannot be determined. To explain, when the target object has a low signal level and the surrounding objects have a high signal level, the peak detecting unit will detect the high signal levels of the surrounding objects and the exposure time will be greatly reduced to decrease the exposure to the high signal levels. Since the signal level of the target object is already low, greatly reducing the exposure time will decrease the target object signal level below detectable levels. As a result, a detectable contrast between the target object and the surrounding objects cannot be ascertained and the correct exposure time cannot be determined. Consequently, accurate auto-focusing cannot be achieved with the devices of
FIGS. 11 and 13
.
Another disadvantage of the devices of
FIGS. 11 and 13
is that auto-focusing cannot be performed accurately for a target object. The devices of
FIGS. 11 and 13
do not detect the peak value for only the target object, but detect the peak value for all of the objects in the image projected on the sensor array
102
. Therefore, it is not possible to perform accurate auto-focusing for only the target object in the image.
OBJECTS AND SUMMARY OF THE INVENTION
An object of the invention, therefore, is to provide an adaptive peak value detection method and apparatus that adaptively detects peak values.
A further object of the invention is to provide an adaptive peak value detection method and apparatus which eliminates the problem of high contrast in the detected image.
Another object of the invention is to provide an adaptive peak value detection method and apparatus that improves auto-focusing for the target object.
In accordance with the above objectives, the present invention provides an adaptive peak value detection apparatus and method.
A first embodiment of the present invention determines the peak value for a peak hold section while ignoring other sections of the input signal, thus preventing a high contrast from occurring. In addition, the peak hold section is selected to allow objects in the image to be adapti
Maki Yasuhito
Yoshida Shin-ya
Frommer William S.
Frommer & Lawrence & Haug LLP
Polito Bruno
Vu Ngoc-Yen
LandOfFree
System for performing selective detection of image signal... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System for performing selective detection of image signal..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System for performing selective detection of image signal... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2928857