Data processing: measuring – calibrating – or testing – Calibration or correction system – Sensor or transducer
Reexamination Certificate
1999-02-05
2001-10-16
Hoff, Marc S. (Department: 2857)
Data processing: measuring, calibrating, or testing
Calibration or correction system
Sensor or transducer
C702S085000, C702S124000, C702S183000
Reexamination Certificate
active
06304826
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention generally relates to image sensing systems and more particularly relates to a self-calibration method and circuit architecture of linear image sensors that can be used in scanners, facsimile, photocopy machines and other image reproduction systems.
2. Description of the Related Art
There are many applications that need an imaging system to convert a target to an electronic format that can be subsequently analyzed, printed, distributed or archived. The electronic format is generally a digital image of the target. A typical example of the imaging system is a scanner and the target is a sheet of paper from a book or an article. Through the scanner, an electronic or digital image of the paper is generated and subsequently may be analyzed, computed, or transmitted through the Internet.
An imaging system generally includes an image sensing module that converts a target optically into an image. The key element in the sensing module that converts the target optically to the image is an image sensor comprising an array of photodetectors responsive to light impinged upon the image sensor. Each of the photodetectors produces an electronic signal representing the intensity of light reflected from the target. The electronic signals from all the photodetectors are readout as a video signal that is then digitized through an analog-to-digital converter to produce a digital signal or an image of the target.
FIG. 1A
illustrates a configuration system
100
that has been used for the past tens of years. A scanning document
110
that can be a page from an article or book is scanned in by an image sensing system
111
that can be a scanner, such as SCANJET 4100CSE Color Scanner from Hewlett Packard. The output of the scanner is typically a digital image
114
of scanning document
110
. Scanner
111
includes an image sensor
112
and an analog-to-digital converter
115
. Image sensor
112
generates images
117
that are typically digitized to gray scale or color images of 8-bit precision. Binalization process
116
receives and converts image
114
to binary image
118
that is a preferred form for data analysis and understanding in data process
120
. Binalization process
116
is typically implemented in a separate circuit or a software application. The separate circuit may be implemented in a post-processing circuit coupled to A/D converter
115
and the software application may be embedded in a scanner driver or provided in a commercial image editing software, such as Adobe PhotoShop, running in a host computer
119
.
FIG. 1B
depicts a contact image sensor (CIS) system that can be used in image sensor
112
of FIG.
1
A. Scanning document
110
is illuminated by an illumination source
121
. Reflected light from scanning document
110
is collected and focused by a full-width rod-lens system
122
. The CIS system allows one-to-one scanning of the document because rod lens
122
and an image sensor chip
124
are of the same width as (or greater width than) scanning document
110
.
FIG. 1C
is a functional block diagram of image sensor
112
, along with
FIG. 1D
showing some detail of the construction of image sensor array
126
. To be specific, a plurality of individual sensor chips
130
are butted end-to-end on a single substrate. Each of the individual sensor chips comprises a plurality of photodetectors
128
arranged in a row. In operation, image sensor array
126
is triggered by a start pulse to the first-in-sequence individual sensor chip
130
which serially activates the photodetectors on the first individual sensor chip
130
. After the signal from the last photodetector element of the first individual sensor chip
130
is read, an end-of-scan pulse is generated so that the next sensor chip in sequence is triggered.
The number of individual sensor chips chosen is dependent upon the desired width of scanning. Sensor array
126
also comprises necessary circuits to serially activate the individual chips and to readout signals generated from photodetectors. The strength of the signals is directly proportionate to the reflected light from the scanning document. To preserve the contents in the scanning document, most CIS systems produce signals that are subsequently digitized to 8 or 12 bit data by a following analog-to digital (A/D) converter.
In many imaging applications, such as check verification at checkout counters in a retail store and document archival, the primary interest is to extract texture information from captured images, for example, for optical character recognition (OCR). To be applicable for such process, the images are preferably in binary format, namely the texture information in black and the background in white or vice versa. In other words, the digitized signals from the A/D converter must be binalized.
The photodetectors in sensor array
126
are, however, subject to several inherent shortcomings that may cause sensor array
126
to produce errors that are hardly correctable in binary data. One of the shortcomings is that the gain from the photodetectors is not uniform from one photodetector to another within a chip or within the array. For example, the base of an NPN photodetector is formed by ion implantation. There is typically a ±5% non-uniformity across a wafer subject to ion implantation. This non-uniformity results in a current gain variation of as much as ±30% across the wafer. The non-uniformity of the gain yields a non-uniformity of the photo response of the same magnitude. Another shortcoming that may adversely affect the performance of sensor array
126
is inherent noises from some or all of the photodetectors or the non-uniformity of the wafer. Signals generated by the photodetectors could be distorted by the noises, which could cause misrepresentation of the contents in scanning objects. There is therefore a great need for an image sensor that produces signals, ideally, independent from the noises or at least with minimized noise effects.
SUMMARY OF THE INVENTION
The present invention has been made in consideration of the above described problems and needs and has particular applications to image sensors that are widely used in image systems, such as scanners, digital cameras and computer vision systems. The present invention discloses a self-calibration method and circuit architecture for image sensors and compensates signals from the image sensors for any distortions caused by inherent deficiencies in the image sensors. Image sensing modules employing the present invention will produce self-calibrated image signals without requiring the commonly used follow-up signal processing schemes.
According to one embodiment, the present invention is a method for self-calibration of an image sensor comprising a plurality of photodetectors, each producing an electronic signal when the image sensor is activated, the method comprising
generating respectively first test electronic signals and second test electronic signals; said first test electronic signals and said second test electronic signals representing respectively a first test target and a second test target;
determining pairs of gain and offset from said first test electronic signals and said second test electronic signals; wherein each of said pairs of gain and offset corresponds to one of said photodetectors;
storing said pairs of gain and offset in a memory; and
adjusting, in a circuit, electronic signals from said photodetectors in accordance to said pairs of gain and offset in said memory, wherein said circuit is directly connected to said memory.
According to another embodiment, the present invention is a circuit architecture for self-calibration of an image sensor, said circuit architecture comprising
an amplifier coupled to said image sensor, said image sensor producing first electronic test signals and second electronic test signals when said image sensor is activated to image a first test target and a second test target;
an analyzing circuit coupled to said amplifier and receiving said first and
Hoff Marc S.
Syscan Inc.
Tsai Carol S
Zheng Joe
LandOfFree
Self-calibration method and circuit architecture of image... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Self-calibration method and circuit architecture of image..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Self-calibration method and circuit architecture of image... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2593414