WO2007049900A1

WO2007049900A1 - Flicker detecting device

Info

Publication number: WO2007049900A1
Application number: PCT/KR2006/004343
Authority: WO
Inventors: Sung Su Lee
Original assignee: Pixelplus Co., Ltd
Priority date: 2005-10-24
Filing date: 2006-10-24
Publication date: 2007-05-03
Also published as: KR100715932B1; KR20070044258A

Abstract

A flicker detecting device for detecting flicker noise in a CMOS (complementary metal-oxide-silicon) image sensor is provided. A flicker detecting pixel unit includes a 1 X 1 single pixel and reads and compares signals with respect to time change or in brightness of a fluorescent lamp from a fixed flicker detecting pixel by performing image sensing for a predetermined charge integration time. An analog-to-digital converter converts an analog output of the flicker detecting pixel unit into a digital value. A memory temporarily stores sampled odd digital data output from the analog-to-digital converter. An odd/even comparator compares sampled even digital data output from the analog-to-digital converter with the sampled odd digital data output of the memory and outputs comparison result. A shift register shifts a resultant value based on the comparison result of the odd/even comparator. A comparator determines whether there is a constant alternating current pattern or flicker is detected by comparing the resultant values of the shift register.

Description

FLICKER DETECTING DEVICE

Technical Field

[1] The present invention relates to a CMOS (complementary metal-oxide-silicon) image sensor, and more particularly, to a device and a method for detecting flicker noise in a CMOS image sensor. Background Art

[2] Generally an image sensor is a semiconductor device converting an optical image into an electrical signal. As the types of the image sensor, there are a CCD (charge coupled device) and a CMOS (complementary metal-oxide-silicon) image sensor. The CCD is a device in which MOS (metal-oxide-silicon) capacitors which charge carriers are stored in and move to/from are disposed very closely with each other. The CMOS image sensor is a device using switching technology in which MOS transistors are formed corresponding to the number of pixels by adopting CMOS technology utilizing a control circuit and a signal processing circuit as peripheral circuits, and outputs are detected one by one by the MOS transistors.

[3] The Image sensor may be used in various application fields. And the image sensor can be used in devices such as digital still cameras, medical cameras, and cellular phones. In most of the application fields, the image sensor having no flicker noise regardless of a light source is required.

[4] The light sources used in image capture of the CMOS image sensor can be classified into sunlight and artificial light sources. The artificial light sources can be sub-classified into fluorescent lamps and incandescent lamps. A frequency of an alternate current power source used for the fluorescent lamps is different for each country. However, in most countries, an alternate current power source having a frequency of 50 Hz or 60 Hz frequency is used as illustrated in FIGS. 1 and 2, and accordingly the brightness frequency of the fluorescent lamps is 100 Hz or 120 Hz.

[5] FIGS 3 and 4 are diagrams illustrating a method of exposure control using an electrical rolling shutter according to conventional technology.

[6] In the conventional method, as illustrated in FlG. 3, exposure time is controlled by moving a pre-reset line and a read-out line downward. All pixels belong to the pre- reset line are reset while data of pixels belong to the read-out line is output to an external circuit. The exposure time of pixels belong to a line is from time when the pre-reset pointer passes the line, that is time when all the pixels belong to the line are reset, to time when the read-out pointer designates the line, that is time when the data of the line is read out. Generally, the method of controlling exposure in the CMOS image sensor is as explained above.

[7] As illustrated in FlG. 4, a problem occurs when an image is captured by the image sensor using the electrical rolling shutter technology under the fluorescent lamp. Referring to lines A and B in the FlG. 4, line A is reset at time point t0 and read-out at time point tl while line B is reset at time point t2 and read-out at time point t3. Here, the exposure time is tl - 10 = t3 - 12.

[8] However, the amount of incident light onto line A is different from the amount of incident light onto line B, so resultant amplitudes of outputs from pixels at lines A and B are different with each other. In other words, although pixels at lines A and B are exposed for a same time, images having different brightness are output from lines A and B. As a result, flicker noise or flicker phenomenon occurs as illustrated in FlG. 5. Although the flicker noise does not occur when an image is captured by the image sensor under the sunlight or the incandescent lamp, when the same image is captured by the image sensor under the fluorescent lamp, brightness difference in a form of stripes may appear in the image. The occurrence of the brightness difference in the form of stripes in the image, when the image sensor is used under the fluorescent lamp, is called the flicker noise.

[9] General method to solve the problem of the flicker noise is to expose the image sensor for multiples of the brightness period T, since integrating a sine wave for one period always results in a constant value. However, when the frequency of light is not known whether it is 50 Hz, 60 Hz, or sunlight having a constant brightness level, the exposure time which should be multiples of the brightness period of the light cannot be determined. Accordingly, a method of detecting a frequency of light, whether it is 50 Hz or 60 Hz, exists is required.

[10] FIGS. 7 and 8 are diagrams illustrating sampled waveforms for detecting whether there is light having a specific period.

[11] As illustrated in FlG. 7, when there is light having periodical brightness, the existence of the light having the specific period can be determined. When the brightness period is T, sampling is performed at a period of T/2. When the sampling is performed as illustrated in FlG. 7, an odd sample is always greater than an even sample, and accordingly the existence of light having a specific cycle can be determined.

[12] However, it may not clearly determined whether there is relation between the even and odd samples in terms of amplitude as illustrated in FlG. 8, although the sampling is performed at a regular period of Ts = Tf / 2. That is a case when the brightness and the sampling time matches coincidentally. Disclosure of Invention Technical Problem

[13] In order to solve the aforementioned problems, an object of the present invention is to provide a flicker detecting device capable of detecting flicker noise in a CMOS image sensor for providing an optimized image regardless of existence of flicker by detecting the flicker using an additional flicker detection pixel. Technical Solution

[14] According to an aspect of the present invention, there is provided a flicker detecting device comprising: a flicker detecting pixel unit including a 1 X 1 single pixel and detecting a brightness frequency of incident light onto the single pixel by performing image sensing for a predetermined charge integration time; an analog-to-digital converter converting an analog output of the flicker detecting pixel unit into a digital value; a memory temporarily storing sampled odd digital data output from the analog- to-digital converter; an odd/even comparator comparing sampled even digital data output from the analog-to-digital converter with the sampled odd digital data output of the memory and outputting comparison result; a shift register shifting a resultant value based on the comparison result of the odd/even comparator; and a comparator determining whether there is a constant alternating current pattern or flicker is detected by comparing the resultant values of the shift register. Brief Description of the Drawings

[15] The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings in which:

[16] FIGS. 1 and 2 are diagrams of a waveform illustrating voltage and brightness of a commercial alternate current power source used for a fluorescent lamp with respect to time;

[17] FIGS 3 and 4 are diagrams illustrating a method of exposure control using an electrical rolling shutter according to conventional technology;

[18] FlG. 5 is a diagram illustrating a flicker phenomenon occurred when a method of exposure control according to the conventional technology is used;

[19] FlG. 6 is a diagram illustrating an adjusting process of exposure time to multiples of a fluorescent lamp period for removing the flicker phenomenon;

[20] FIGS. 7 and 8 are diagrams illustrating sampled waveforms for detecting whether there is light having a specific period;

[21] FIGS. 9 and 10 are diagrams illustrating a waveform having a specific sampling period and a waveform on which magnitudes and brightness of even/odd samples are represented according to an embodiment of the present invention to solve problems in the conventional technology; [22] FlG. 11 is a diagram illustrating a flicker detecting circuit according to an embodiment of the present invention; and

[23] FlG. 12 is an exemplary waveform illustrating observed windows having 16 samples according to an embodiment of the present invention. Best Mode for Carrying Out the Invention

[24] Hereinafter, the present will be described in detail with reference to accompanying drawings.

[25] FIGS. 9 and 10 are diagrams illustrating a waveform having a specific sampling period and a waveform on which magnitudes and brightness of even/odd samples are represented according to an embodiment of the present invention to solve the problems in the conventional technology.

[26] The sampling period according to the present invention is set to a value a little differently from Tf/2 to solve the problems in the conventional technology. Although it is illustrated for a case when Ts < Tf/2 in FlG. 9, the result is similar for a case when Ts > Tf/2, so only the case when Ts < Tf/2 will be explained.

[27] FlG. 10 is a diagram illustrating amplitudes and brightness of even/odd samples together for the waveform, and a beat having a frequency corresponding to difference from the sampling frequency, that is twice the brightness frequency can be seen.

[28] FlG. 11 is a diagram illustrating a flicker detecting circuit according to an embodiment of the present invention.

[29] Referring to FlG. 11, the flicker detecting circuit includes a flicker detecting pixel unit 1, an analog-to-digital converter 2, a flicker time controlling unit 3, a memory 4, an odd/even comparator 5, a shift register 6, and a comparator 7.

[30] Here, the flicker detecting pixel unit 1 is a kind of an additional pixel unit. The flicker detection unit 1 includes a 1 X 1 single pixel constituting one line, and the length of the line is the same as that of a line of a main pixel unit (not shown). In the embodiment, the flicker detecting pixel unit 1 includes a IX 1 single pixel. The flicker detecting pixel unit 1 performs image sensing for a predetermined charge integration time and detects a brightness period of incident light ontothe single pixel.

[31] The analog-to-digital converter 2 converts an analog output having an analog value of the flicker detecting pixel unit linto a digital value and outputs the converted digital value to the memory 4 and the odd/even comparator 5.

[32] The flicker time controlling unit 3 controls all the timings for controlling exposure for periodically reading data of the flicker detecting pixel unit 1, comparing the read data b with the sampled odd data output of the memory 4 by the odd/even comparator 5, or outputting the comparison result to the shift register 6 for shifting.

[33] The memory 4, as explained above, temporarily stores the sampled odd digital data, for example 8 bit data, output from the analog-to-digital converter 2. Preferably, the memory 4 may be implemented by a flip-flop. [34] The odd/even comparator 5 compares the sampled even digital data b output from the analog-to-digital converter 2 with the sampled odd data output of the memory 4 and outputs the comparison result to the shift register 6. [35] The odd/even comparator 5 tests whether the even digital data b and the odd data satisfy the following Conditions 1, 2, and 3 in the comparing process. [36] [Condition 1]

[37] even data > odd data + threshold

[38] [Condition 2]

[39] even data + threshold < odd data

[40] [Condition 3]

[41] The data does not satisfy Conditions 1 and 2.

[42] When the comparison result satisfies Condition 1, the odd/even comparator 5 inputs Tto the shift register 6. When the comparison result satisfies Condition 2, the odd/even comparator 5 inputs 'O'to the shift register 6. When the comparison result does not satisfy Conditions 1 and 2 simultaneously, inverted value of data c previously input to the shift register 6 is shifted to make the result of the comparator 7 negative. [43] The shift register 6 performs shifting the resultant value based on the comparison result of the odd/even comparator 5, as explained above. Preferably, the shift register 6 may be implemented by a shift register having a size of 1 bit X (Sample size of an observed window for comparing sampled even digital data with the sampled odd digital data) / 2. [44] The comparator 7 determines whether there is a constant alternating current pattern or flicker is detected by comparing the resultant values of the shift register 6. [45] The operations and effects of the flicker detecting circuit according to the embodiment of the present invention explained above will now be explained in reference to FlG. 11.

[46] At first, the flicker detecting pixel unit 1 performs image sensing for a predetermined charge integration time and detects a brightness frequency of incident light onto a single pixel. [47] The analog-to-digital converter 2 converts an analog value that is an analog output of the flicker detecting pixel unit 1 into a digital value and outputs the digital value to the memory 4 and the odd/even comparator 5. [48] The memory 4 temporarily stores sampled odd digital data, for example 8 bit data, output from the analog-to-digital converter 2. [49] The odd/even comparator 5 compares the sampled even digital data b output from the analog-to-digital converter 2 with the sampled odd data output of the memory 4 and outputs the comparison result to the shift register 6.

[50] The shift register 6 shifts a resultant value based on the comparison result of the odd/even comparator 5.

[51] The comparator 7 determines whether there is a constant alternating current pattern or flicker is detected by comparing the resultant values of the shift register 6.

[52] For example, values in the shift register 6 are all 'O's or all Ts, it is determined that a constant alternating current pattern is detected, and when 50 Hz or 60 Hz detection signals appears predetermined times or more, it is determined that flicker is detected.

[53] The flicker time controlling unit 3 repeats trying to detect 50 Hz or 60 Hz signals by turns.

[54] FIG. 12 is an exemplary waveform illustrating observed windows having 16 samples according to an embodiment of the present invention. When all the magnitudes of even and odd samples satisfy a same comparison condition simultaneously within a window, that is when all odd samples are a threshold value or more greater (or smaller) than the following even samples, the existence of an alternating current light is confirmed.

[55] In other words, since all even samples are greater than sums of the previous odd samples and the threshold within Observed window 1, it is determined that the flicker is detected.

[56] On the other hand, since not all pairs of (odd sample, even sample) in Observatory window 2 do not satisfy the comparison condition explained above, it cannot be determined that the flicker is detected.

[57] Preferably, the size of observed windows may be smaller than a half period of a beat.

[58] In a conventional image sensor, there is a problem in determining the existence of the flicker by determining whether stripes appears in a captured image. In other words, when it is determined that there is the flicker, flicker cancellation is operated so that the flicker cannot be seen in the image any more. Accordingly, although the flicker condition disappears, it cannot be determined whether it is caused by change in light or the operation of the flicker cancellation.

[59] However, the present invention is capable of continuously providing an optimized image based on the existence of the flicker regardless whether the flicker appears in the image, since an additional pixel is used for detecting the flicker.

[60] While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Industrial Applicability

[61] Accordingly, the present invention is capable of detecting flicker in a CMOS image sensor to provide an optimized image regardless of the existence of the flicker by using an additional pixel for detecting the flicker.

[62] In addition, the present invention is capable of consistently providing an optimized image based on the existence of the flicker regardless whether the flicker appears in the image, since an additional pixel is used for detecting the flicker.

Claims

[1] A flicker detecting device comprising: a flicker detecting pixel unit including a 1 X 1 single pixel and detecting a brightness frequency of incident light onto the single pixel by performing image sensing for a predetermined charge integration time; an analog-to-digital converter converting an analog output of the flicker detecting pixel unit into a digital value; a memory temporarily storing sampled odd digital data output from the analog- to-digital converter; an odd/even comparator comparing sampled even digital data output from the analog-to-digital converter with the sampled odd digital data output of the memory and outputting comparison result; a shift register shifting a resultant value based on the comparison result of the odd/even comparator; and a comparator determining whether there is a constant alternating current pattern or flicker is detected by comparing the resultant values of the shift register.

[2] The device of claim 1 further comprising a flicker time controlling unit controlling all the timings for controlling exposure for periodically reading data of the flicker detecting pixel unit, the odd/even comparator's comparing the read data with the sampled odd data output of the memory, or outputting the comparison result to the shift register for shifting.

[3] The device of claim 1 or 2 wherein the memory is implemented by a flip-flop.

[4] The device of claim 1 or 2 wherein the shift register is implemented by a shift register having a size of 1 bit X (Sample size of an observed window for comparing sampled even digital data with the sampled odd digital data)/ 2.