KR20120046577A - Method for detecting dead pixel and apparatus for performing the same - Google Patents

Abstract

PURPOSE: A bad pixel detection method and an apparatus performing the same are provided to increase compression rate of a dead pixel and detection probability of the dead pixel. CONSTITUTION: A bad pixel detecting unit does not focus a photographed image by control of an AF(Auto Focus) control function(105). The apparatus sets up collection time of an image sensor as predetermined first time(107). The apparatus determines a hot pixel based on pixel value difference configuring a photographed image frame(109). The apparatus stores location information of hot pixel(111).

Description

Bad pixel detection method and apparatus for performing the same {METHOD FOR DETECTING DEAD PIXEL AND APPARATUS FOR PERFORMING THE SAME}

The present invention relates to detection of bad pixels of an image sensor, and more particularly, to a bad pixel detection method applicable to an image sensor having a large number of pixels and an apparatus for performing the same.

An image sensor refers to an optical conversion device in which a plurality of pixels are arranged in a two-dimensional structure, and each pixel of the image sensor has an electric signal of a predetermined level according to the intensity of light incident through the optical system. And converts the converted electrical signal into a digital signal capable of signal processing and outputs the converted digital signal.

The image sensor may include pixels that do not operate normally due to various causes during the manufacturing process. Since the pixels do not respond to light normally, the image output from the image sensor containing abnormal pixels may be different from the actual distortion. The displayed image is displayed.

As described above, a pixel that does not normally react to light may appear brighter or darker than the actual one, and a pixel that stands out due to a large difference in brightness from surrounding pixels is called a bad pixel (or a dead pixel).

A conventional dead pixel detection method is to transfer Bayer data, which is provided from an image sensor to an image signal processor, to n line memories required for dead pixel cancellation (DPC). The defective pixels were determined by applying various bad pixel detection algorithms to each RGB pixel corresponding to the kernel using an n × n kernel, and then corrected.

However, in the conventional dead pixel detection method, a single dead pixel can be detected through the use of a simple algorithm. However, when a plurality of dead pixels are adjacent to each other or a large number of dead pixels exist in the n × n kernel, the dead dead pixels are detected. Since it can be recognized as an edge or a general pixel rather than a pixel, there is a disadvantage of reducing the detection probability of the dead pixel and the correction rate of the dead pixel.

In addition, in order to compensate for the above-mentioned disadvantages, when storing dead pixel position information using an One Time Programmable EPROM (OTP) or detecting a dead pixel using a kernel larger than an n × n kernel, Although it is possible to increase the correction rate, the chip size is greatly increased, which is difficult to apply to a portable terminal.

The present invention provides a bad pixel detection method capable of improving the detection accuracy and correction rate of bad pixels.

The present invention is to provide a bad pixel detection device that can improve the detection accuracy and correction rate of a bad pixel.

Technical problems of the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the following description.

In accordance with an aspect of the present invention, there is provided a method for detecting a bad pixel, by controlling an auto focus adjustment function to prevent an image from being in focus, and photographing the image. Setting an integration time of the image sensor to a preset first time such that is under exposure, and pixels of each pixel constituting at least one out-of-focused image frame Determining a hot pixel based on the value difference and storing the determined position information of the hot pixel.

In addition, a bad pixel detection apparatus according to an aspect of the present invention for achieving another object of the present invention, the image sensor for providing an image corresponding to the brightness of the incident light, and the image sensor when the image sensor is activated After controlling to prevent focusing, setting the condensing time of the image sensor to a preset time, determining a dead pixel based on a difference of pixel values of each pixel constituting an image frame provided from the image sensor, and then determining A bad pixel processing unit for storing the position information of the dead pixel, an automatic exposure adjusting unit for setting the condensing time of the image sensor to a predetermined time in accordance with the control of the bad pixel processing unit, and according to the control of the bad pixel processing unit Autofocus adjustment to adjust the position of the focusing lens so that the captured image is out of focus And a memory that stores the location information of the dead pixel.

According to the method and apparatus for detecting a bad pixel of the present invention, it is possible to remarkably improve the probability of detection of a dead pixel and the correction rate of the dead pixel while using the device resources to a minimum.

1 is a flowchart illustrating a method of detecting a bad pixel according to an exemplary embodiment of the present invention, which illustrates a hot pixel detection process.
2 is a flowchart illustrating a method of detecting a bad pixel according to an exemplary embodiment of the present invention, which illustrates a process of detecting a cold pixel.
3 is a flowchart illustrating a bad pixel detection method according to another exemplary embodiment of the present invention.
4 is a block diagram illustrating a configuration of a bad pixel detection apparatus according to an exemplary embodiment.

As the present invention allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the written description.

However, this is not intended to limit the present invention to specific embodiments, it should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component. And / or < / RTI > includes any combination of a plurality of related listed items or any of a plurality of related listed items.

When a component is referred to as being "connected" or "connected" to another component, it may be directly connected to or connected to that other component, but it may be understood that other components may be present in between. Should be. On the other hand, when a component is said to be "directly connected" or "directly connected" to another component, it should be understood that there is no other component in between.

The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting of the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, the terms "comprise" or "have" are intended to indicate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, and one or more other features. It is to be understood that the present invention does not exclude the possibility of the presence or the addition of numbers, steps, operations, components, components, or a combination thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the relevant art and are to be interpreted in an ideal or overly formal sense unless explicitly defined in the present application Do not.

Hereinafter, with reference to the accompanying drawings, it will be described in detail a preferred embodiment of the present invention. In describing the present invention, in order to facilitate the overall understanding, the same reference numerals are used for the same elements in the drawings, and redundant description of the same elements is omitted.

Hereinafter, the overall concept of the bad pixel detection method according to an embodiment of the present invention will be described.

High-pixel System on Chip (SoC) -based CMOS image sensors essentially use auto focus (AF) to focus the image.

The auto focus method uses a stepper motor to adjust the position of the focusing lens included in the lens module while continuously determining a corresponding contrast, and a contrast detection method that determines that focus is achieved when the contrast is maximized. (contrast detection system) and by deliberately creating a distortion in the lens and correcting it using the distortion inside the chip.

Among the above-described auto focusing methods, the method of automatically adjusting the focus by using the distortion of the lens has a large implemented logic size and a high development cost. Therefore, a contrast detection method that is simple to control and low in development cost is mainly used.

The method and apparatus for detecting a bad pixel according to an embodiment of the present invention use a dead pixel using an autofocus control function of a contrast detection method applied to an SoC-based image sensor having a large number of pixels (for example, 3 million or more pixels). When the image sensor is initially driven, the focusing lens is adjusted by adjusting the focusing lens while maintaining an integration time (integration time or illuminance time) for adjusting the amount of light incident on the image sensor. Out of focus to soften the edges of the image as much as possible (or to minimize the contrast), and to find pixels that have a pixel value higher or lower than the adjacent pixel value for a frame or frames The result is determined as a dead pixel, and the position information of the determined dead pixel is stored in a memory (SRAM). The dead pixel may be classified into a hot pixel having a large high pixel value and a cold pixel having a low pixel value.

When the dead pixel is corrected, the dead pixel is accurately corrected by using the position information of the dead pixel. Therefore, it is possible to significantly improve the detection probability of the dead pixel and the correction rate of the dead pixel while minimizing the resource of the device.

1 is a flowchart illustrating a method of detecting a bad pixel according to an exemplary embodiment of the present invention, and illustrates a process of detecting a hot pixel.

Referring to FIG. 1, in the bad pixel detection method according to an embodiment of the present invention, a bad pixel detection device (hereinafter, abbreviated as 'device') first supplies power to an image sensor to activate the image sensor (step 101). ), The Auto Exposure (AE) function is stopped (step 103). In general, when the image sensor is driven, the device does not know the brightness of the external light source. Therefore, the device obtains brightness information for one frame or a plurality of frames and applies the obtained brightness information to the automatic exposure algorithm to obtain the optimum exposure value or optimal value. Since the exposure value is determined, in the embodiment of the present invention, the above automatic exposure function is stopped to detect the dead pixel.

The device then controls the auto focus function to smooth out the edges included in the image by out of focus (step 105). Here, the device controls the driving of the step motor to move the position of the focusing lens out of focus, thereby minimizing contrast and smoothing the edges included in the image.

The reason for smoothing the edges in step 105 is to make hot pixel detection more accurate by making the level difference between the normal pixel and the hot pixel larger. That is, in the conventional dead pixel detection method, when one hot pixel exists among the pixels input to the n × n kernel or several non-adjacent areas are easily detected, the hot pixels are continuously positioned. When present or dense, it is difficult to detect such a hot pixel as an edge. In order to solve the above problems, one embodiment of the present invention controls the auto focusing function so that the focus is not in focus, thereby smoothing the edges, thereby increasing the pixel value difference between the normal pixel and the hot pixel so that the continuous or dense It also makes it possible to accurately detect hot pixels present at locations.

Referring again to FIG. 1, the device quickly sets the integration time of the image sensor to a preset first time, so that a small amount of light is focused on the image sensor (step 107). That is, in step 107, the focusing time is set to the first time so that the image is in an under exposure state. Pixels of the image sensor that focus a small amount of light through the adjustment of the focusing time as described above generally have a low luminance pixel value. In contrast, a hot pixel has a higher value than a normal pixel. Pixels can be detected.

The device then detects a hot pixel in one frame or a plurality of frames (step 109) and stores position information of the detected hot pixel (step 111).

In operation 109, the device may detect the hot pixel through various methods. For example, the device may determine a pixel having a pixel value higher than a preset pixel reference value as a hot pixel, and determine a pixel having a pixel value larger than a predetermined value larger than an average value of all pixels included in the frame as a hot pixel. You may. Alternatively, a pixel having a pixel value greater than or equal to a predetermined value than a pixel value of an adjacent pixel may be determined as a hot pixel. In operation 111, the device may store the detected hot pixel information in the SRAM.

After performing the above-described steps 101 to 111 to detect the hot pixel and store the detected position information of the hot pixel, the device corrects the hot pixel using the position information of the hot pixel (step 113). Here, the device generates a pattern by assigning a mask value to the hot pixel position of the n × n kernel using the stored hot pixel position information, and generates the pattern n × n to perform normal processing of the frame. Hot pixel correction may be performed by applying to pixel data provided through the kernel.

2 is a flowchart illustrating a method of detecting a bad pixel according to an exemplary embodiment of the present invention, and illustrates a process of detecting a cold pixel.

Referring to FIG. 2, in the cold pixel detection method according to an embodiment of the present invention, the device first powers up the image sensor to activate the image sensor (step 201), and then stops the automatic exposure (AE) function ( Step 203).

The device also controls the auto focus (AF) adjustment function to smooth out the edges included in the image by out of focus (step 205). Here, the device controls the driving of the step motor to move the position of the focusing lens out of focus, thereby minimizing contrast and smoothing the edges included in the image.

Thereafter, the device obtains illuminance information for one frame or a plurality of frames (step 207). Here, the illumination information may be obtained through various known methods, for example, the illumination information may be obtained by using an operation time of the photodiode.

Thereafter, the apparatus compares the acquired illuminance value with a preset reference value (step 209), and when the acquired illuminance value is not smaller than the preset reference value, the integration time of the image sensor is set to the preset second time. The slow setting allows a large amount of light to be focused on the image sensor (step 211).

In other words, in step 211, the focusing time is set to the second time so that the image is in an over exposure state. Since the pixels of the image sensor condensing a large amount of light sources through the adjustment of the condensing time as described above are saturated in the low light condition, the pixels differ greatly from the cold pixels.

Thereafter, the device detects a cold pixel in one frame or a plurality of frames (step 213) and stores position information of the detected cold pixel (step 215). Here, the detected cold pixel information may be stored in the SRAM.

After detecting the cold pixel through the above steps 201 to 215 and storing the detected cold pixel position information, the apparatus corrects the cold pixel using the cold pixel position information (step 217). Here, the device generates a pattern by assigning a mask value to the cold pixel position of the n × n kernel using the stored cold pixel position information, and generates the pattern n × n to perform normal processing of the frame. Cold pixel correction may be performed by applying to pixel data provided through a kernel.

In addition, when the illuminance value acquired in step 209 is smaller than the preset reference value, the apparatus determines the low illuminance and does not perform steps 211 to 217. The maximum condensing time in a low light environment causes amplification of amplified noises, which affects the determination of a cold pixel. Therefore, in the embodiment of the present invention, a cold pixel detection process is not performed in a low light environment lower than a preset reference light intensity.

3 is a flowchart illustrating a method of detecting a bad pixel according to another exemplary embodiment of the present invention, and illustrates a case of selectively detecting a hot pixel and a cold pixel illustrated in FIGS. 1 and 2.

Referring to FIG. 3, first, power is supplied to an image sensor to activate the image sensor (step 301), and then the auto exposure (AE) function is stopped (step 303).

The device then controls the auto focus function to smooth out the edges included in the image by out of focus (step 305). It is determined whether the dead pixel to be detected is a hot pixel or a cold pixel (step 307). Here, the detected dead pixel type may be selected by a user or an inspector.

If it is determined in step 307 that the hot pixel is detected, the device quickly sets the focusing time of the image sensor to a preset first time, so that a small amount of light is focused (i.e. underexposed) on the image sensor (step 309, hot pixels are detected in one frame or a plurality of frames (step 311), and the position information of the detected hot pixels is stored (step 321).

Alternatively, if it is determined in step 307 that the cold pixel is detected, the apparatus acquires illuminance information for one frame or a plurality of frames (step 313), and compares the obtained illuminance value with a preset reference value (step 315). When the acquired illuminance value is not smaller than the preset reference value (that is, when it is not low illuminance), the light condensing time of the image sensor is set slowly to a second predetermined time, so that a large amount of light is condensed on the image sensor (ie, Overexposure) (step 317).

Thereafter, the device detects a cold pixel in one frame or a plurality of frames (step 319), and stores position information of the detected cold pixel (step 321).

As described above, after detecting the hot pixel or the cold pixel, and after the detected position information of the hot or cold pixel is stored, the device uses the position information of the hot or cold pixel to correspond to the position information of the hot or cold pixel. The dead pixel is corrected (step 323).

4 is a block diagram illustrating a configuration of a bad pixel detection apparatus according to an exemplary embodiment.

Referring to FIG. 4, a bad pixel detection apparatus according to an exemplary embodiment may include an image sensor 410, a bad pixel processor 420, an automatic exposure adjuster 430, an autofocus adjuster 440, and a memory 450. It may include.

The image sensor 410 may be configured as, for example, a complementary metal oxide (CMOS) image sensor, and converts and outputs an electrical signal corresponding to the brightness of incident light into a digital signal. Here, the image sensor 410 may be configured as a Bayer pattern, and output Bayer image data corresponding to the brightness of the incident light.

When the image sensor 410 is activated, the bad pixel processing unit 420 controls the auto exposure adjustment unit 430 to pause the auto exposure function, and controls the auto focus adjustment unit 440 to focus the position of the focusing lens. By moving to the wrong position, the contrast is minimized to smooth the edges included in the captured image.

In addition, when the bad pixel processor 420 detects a hot pixel, the bad pixel processor 420 controls the automatic exposure adjustment unit 430 to quickly set the condensing time of the image sensor 410 to a preset first time. After the exposure state, the hot pixel is detected in one frame or a plurality of frames, and the position information of the detected hot pixel is stored in the memory 450.

Alternatively, the bad pixel processing unit 420 acquires an illumination value using one or a plurality of frames provided from the image sensor 410 when detecting a cold pixel, and an image sensor when the obtained illumination value is larger than a preset reference value. Slowly setting the condensing time of 410 to a second predetermined time so that the captured image is in an over exposure state, and then detecting a cold pixel in one frame or a plurality of frames, and detecting the detected cold pixel. Location information is stored in the memory 460.

As described above, after the detection and the position information of the hot pixel or the cold pixel are stored in the memory 450, the bad pixel processor 420 uses the position information of the hot pixel or the cold pixel to store a dead pixel (hot pixel or cold pixel). Calibrate

In detail, the bad pixel processing unit 420 generates a pattern by allocating a mask value to a hot pixel location of an n × n kernel using location information of hot or cold pixels stored in the memory 450. The pattern may be applied to pixel data provided through an n × n kernel from a line memory (not shown) to perform normal processing of a frame, thereby performing correction of a dead pixel (hot pixel or cold pixel).

The automatic exposure control unit 430 automatically adjusts the exposure according to the illuminance when processing a normal frame, and stops the automatic exposure function according to the control of the bad pixel processing unit 420 when detecting a dead pixel, and the hot pixel. The detection time is set to the first time, and the detection time is set to the second time when the cold pixel is detected.

The auto focus adjusting unit 440 continuously adjusts the focus lens by changing the position of the focus lens while processing a normal frame and continuously focusing the lens by positioning the focus lens to a position where the contrast is maximized. When the dead pixel is detected, the position of the focusing lens is positioned at the end of the focusable lens that is movable according to the control of the bad pixel processing unit 420, thereby minimizing the contrast to minimize the edge included in the image. Soften. Here, the focusing lens (not shown) may be moved by a step motor or an ultrasonic motor, and the auto focus adjusting unit 440 may control the rotation of the step motor or the ultrasonic motor.

The memory 450 may be configured of, for example, static random access memory (SRAM), and stores position information of a dead pixel (hot pixel or cold pixel) provided from the bad pixel processing unit 420.

Although described with reference to the embodiments above, those skilled in the art will understand that the present invention can be variously modified and changed without departing from the spirit and scope of the invention as set forth in the claims below. Could be.

410: image sensor 420: bad pixel processing unit
430: auto exposure adjustment unit 440: auto focus adjustment unit
450: memory

Claims (12)

In the method for detecting a bad pixel of the image sensor,
Controlling an auto focus adjustment function so that the captured image is out of focus;
Setting an integration time of the image sensor to a preset first time such that the captured image is in an under exposure state;
Determining a hot pixel based on a pixel value difference of each pixel constituting at least one out-of-focused image frame; And
And storing the determined position information of the hot pixel. The method of claim 1, wherein the bad pixel detection method is performed.
And stopping the execution of the automatic exposure adjustment function before setting the light condensing time of the image sensor to a first predetermined time. The method of claim 1, wherein the controlling of the auto focusing function so that the captured image is out of focus includes:
And controlling the autofocus function so that the contrast of the picked-up image is minimized so as to soften an edge included in the picked-up image. The method of claim 1, wherein the bad pixel detection method is performed.
Generating a correction pattern by allocating a mask value to a hot pixel position of a kernel having a predetermined size by using the stored position information of the hot pixel; And
And correcting the hot pixel by applying the generated correction pattern to pixel data provided through a kernel having the predetermined size. In the method for detecting a bad pixel of the image sensor,
Controlling an auto focus adjustment function so that the captured image is out of focus;
Obtaining illuminance information based on at least one image frame;
Setting the integration time of the image sensor to a preset second time such that the captured image is in an over exposure state when the obtained illuminance information is equal to or greater than a preset illuminance reference value;
Determining a cold pixel based on a pixel value difference of each pixel constituting at least one out of focus image frame; And
And storing the determined position information of the cold pixel. The method of claim 5, wherein the bad pixel detection method is performed.
And when the illuminance information is smaller than the illuminance reference value, detecting no bad pixels. The method of claim 5, wherein the bad pixel detection method is performed.
And stopping the performing of the automatic exposure adjustment function prior to setting the condensing time of the image sensor to a second predetermined time. The method of claim 5, wherein the bad pixel detection method is performed.
Generating a correction pattern by allocating a mask value to a hot pixel position of a kernel having a predetermined size by using the stored position information of the cold pixel; And
And correcting the cold pixel by applying the generated correction pattern to pixel data provided through a kernel having the predetermined size. An image sensor providing an image corresponding to the brightness of the incident light;
When the image sensor is activated, the photographed image is controlled to be out of focus, the focusing time of the image sensor is set to a preset time, and then the difference in pixel value of each pixel constituting the image frame provided from the image sensor is controlled. A bad pixel processor configured to determine dead pixels based on the dead pixels, and store position information of the determined dead pixels;
An automatic exposure adjusting unit for setting a condensing time of the image sensor to a preset time according to control of the bad pixel processing unit;
An autofocus adjusting unit for adjusting a position of a focusing lens such that the captured image is not focused according to the control of the bad pixel processing unit; And
And a memory configured to store position information of the dead pixel. The method of claim 9, wherein the bad pixel processing unit
A mask value is generated by assigning a mask value to a dead pixel position of a kernel having a predetermined size by using the position information of the dead pixel, and applying the generated correction pattern to pixel data provided through the kernel having the predetermined size. And correcting the dead pixel. The method of claim 9, wherein the bad pixel processing unit
In the case of determining a hot pixel among the dead pixels, the condensing time is set to a preset first time so that the captured image is under exposed, and in the case of determining a cold pixel, the condensing time is preset. And setting the second time so as to cause the captured image to be over exposed. 10. The method of claim 9, wherein the automatic focus adjustment unit
According to the control of the bad pixel processing unit, the position of the focusing lens is positioned at the extreme end of the movable lens can be moved to minimize the contrast to smooth the edges included in the captured image, characterized in that the device .

Images