KR20080050067A - Method and apparatus for correcting of defective pixel - Google Patents

Abstract

A method for correcting a defective pixel and an apparatus thereof are provided to enable an image processor to detect the defective pixel using a href signal and a vsync signal inputted from an image sensor and to correct a correction target pixel, thereby improving productivity. A method for correcting a defective pixel comprises the following steps of: receiving pixel data corresponding to an optical subject, a vsync(Vertical synchronization) signal necessary for pixel data acquisition, and an effective data enable signal from an image sensor(910); detecting a fault signal using the vsync signal and the href signal inputted from the image sensor, generating the location information of a correction target pixel in correspondence to the fault signal, and outputting the generated location information to a defective pixel correcting unit(915); generating the type information of the correction target pixel using the location information of the correction target pixel and outputting the generated type information to the defective pixel correcting unit(920); determining whether the type information of the correction target pixel inputted from a correction target pixel detecting unit is single(925); and correcting a center pixel using a difference value between two pixels except the center pixel in the pixel block of a predetermined size adjacent to the correction target pixel(930).

Description

Defect pixel correction method and apparatus therefor {Method and apparatus for correcting of defective pixel}

1 illustrates the location of a defective pixel of a typical pixel array.

2 is a block diagram of an image processing apparatus according to an embodiment of the present invention.

3 is an internal functional block diagram of an image sensor according to an embodiment of the present invention.

4 is a diagram illustrating a clock signal output by a conventional image sensor corresponding to pixel data;

5 is a diagram illustrating a clock signal output by an image sensor according to an embodiment of the present invention corresponding to pixel data.

6 is an internal functional block diagram of an image processor according to an embodiment of the present invention.

FIG. 7 illustrates types of pixels to be corrected according to an embodiment of the present invention. FIG.

8 is a diagram illustrating a type of pixel to be corrected according to another embodiment of the present invention.

9 is a flowchart illustrating a method of correcting a defective pixel by an image processor according to an embodiment of the present invention.

<Explanation of symbols for the main parts of the drawings>

210: image sensor

220: image processor

410: correction target pixel detection unit

420: defective pixel correction unit

430: control unit

TECHNICAL FIELD The present invention relates to an image processing apparatus, and more particularly, to a method and an apparatus for easily correcting a defective pixel by using a position address of the defective pixel.

A digital camera converts a light signal of a subject into a digital signal and processes it. One of the most necessary things is an image sensor.

Charged Coupled Devices (CCDs) and Complementary Metal Oxide Semiconductors (CMOS) containing unit pixels are mainly used.

CCD and CMOS captures an image by accumulating light incident from a subject at each pixel as an electrical signal (charge) through photoelectric conversion means such as a photodiode and converting it into a digital signal. During the manufacturing process, defects may exist in a specific pixel due to foreign matter and other instability in the manufacturing process, and fixed pattern noise may be generated by the defective pixel.

There are three types of defective pixels: Stuck High, Stuck Low, and Abnormal Response, where double stuck high defect pixels always have a high luminance value for a specific illuminance. As the pixel to be shown, noise caused by stuck high defect pixels is called white noise.

For example, if the luminance value of a pixel is in the range of 0 to 255, and if the luminance value is 25, if the luminance value is 25, the stuck high defect pixel will always display a high luminance value, for example, 255. do.

Meanwhile, the stuck low defect pixel is a pixel that always exhibits a low luminance value at a specific illuminance, and the noise caused by the stuck low defect pixel is called black noise.

For example, even when the luminance value of 100 or 200 is represented by the normal function pixel within the luminance value range of the pixel, the low-throw defect pixel is output as low as about 5.

On the other hand, the abnormal response pixel is a pixel indicating a change value relative to the normal pixel. For example, when it is normal to indicate a luminance value of A, the abnormal response pixel responds with a value of 1.25ⅹA instead of A. .

As described above, since various types of defective pixels may exist in the manufacturing process of the image sensor, the image processing apparatus generally performs a process of detecting and appropriately correcting the defective pixels.

When there are defective pixels on the pixel array as shown in FIG. 1, conventionally, the image processor detects a defective pixel using data input from an image sensor, records position information of the corresponding defective pixel in a memory, and corrects the defective pixel. Was performed.

As described above, in order to detect the defective pixel in the image processor, a test environment for detecting the defective pixel by using illuminance and a background image, etc. must be separately provided, and an image sensor must be controlled.

In addition, there is a problem that the product production process is complicated because the image processor has to perform a separate operation for detecting a defective pixel separately.

Accordingly, an object of the present invention to solve the above-mentioned problem is that the image processor uses a valid data enable signal and a vertical synchronizing signal (i.e., clock signal) input from an image sensor to detect a defective pixel (i.e., a pixel to be corrected). The present invention provides a defective pixel correction method and apparatus capable of detecting and performing correction on a corresponding correction target pixel.

Another object of the present invention is to provide a defective pixel correction method and apparatus for detecting and correcting a defective pixel without receiving a separate control signal from an image sensor for detecting the defective pixel.

Another object of the present invention is to provide a defect pixel correction method and apparatus for improving product mass productivity because an image processor does not perform a separate operation for detecting a defective pixel of a pixel array of an image sensor.

Still another object of the present invention is to provide a method and apparatus for correcting a defective pixel, which enables an image processor to perform correct correction on a defective pixel without performing an operation for detecting a defective pixel.

Other objects of the present invention will be easily understood through the description of the following examples.

In order to achieve the above object, according to an aspect of the present invention, there is provided an image processor and an image processing apparatus capable of correcting defective pixels using a vertical synchronization signal and a valid data enable signal.

According to an embodiment of the present invention, a pixel data corresponding to a pixel array, a valid data enable signal in which a fault signal is inserted corresponding to a defective pixel of the pixel array, and a vertical synchronizing signal are inputted from an image sensor. An image processor for generating position information and correcting a pixel to be corrected may be provided.

A correction target pixel detector configured to receive the pixel data, the valid data enable signal, and the vertical synchronization signal, and generate position information of a correction target pixel using the fault signal; And a defective pixel corrector configured to generate a type of the corrected pixel using position information of the corrected pixel, and correct the corrected pixel using a predetermined correction algorithm according to the generated type.

When the fault signal is detected in the valid data enable signal, the correction object pixel detector may generate position information of the correction object pixel corresponding to the fault signal.

The valid data enable signal has a plurality of valid periods, the fault signal is a modified clock signal within the valid period, and the correction target pixel detector is activated by the vertical synchronization signal to rise a rising edge of the valid period. an y-axis information counter for increasing the y-axis coordinate for each edge of either an edge or a falling edge; And an x-axis information counter that increases an x-axis coordinate for each clock signal in the valid period, wherein the pixel to be corrected is counted by the y-axis information counter and the x-axis information counter when the fault signal is detected. Based on the y-axis coordinates and the x-axis coordinates, position information of the correction target pixel may be generated.

The defective pixel correcting unit determines a presence of another corrected pixel in the pixel block after setting a pixel block having a predetermined size using the position information of the corrected pixel as the center pixel. The type of the correction target pixel may be set to a cluster.

When the type of the correction target pixel is a cluster, the defective pixel correction unit may correct the correction target pixel by calculating an average value or a median value of the pixels of the pixel block.

The defective pixel corrector may determine whether another corrected pixel exists in the pixel block, and if the other corrected pixel does not exist, set the type of the corrected pixel to single.

When the type of the correction target pixel is single, the defective pixel correction unit may correct the correction target pixel using a difference value between pixels except for the center pixel in the pixel block.

According to another embodiment of the present invention, a valid data enable signal in which a fault signal is inserted, a vertical sync signal, and pixel data corresponding to the pixel array are output, corresponding to a defective pixel of a pixel array in which a plurality of unit pixels are arranged. An image sensor; And an image processor configured to generate position information of a pixel to be corrected using the valid data enable signal and the vertical synchronization signal to correct the pixel to be corrected.

The image processor may include: a correction target pixel detector configured to receive the pixel data, the valid data enable signal, and the vertical synchronization signal, and generate position information of the correction target pixel using the fault signal; And a defective pixel correction unit generating a type of the correction target pixel by using the position information of the correction target pixel and correcting the correction target pixel by using a correction algorithm predetermined according to the generated type.

When the fault signal is detected in the valid data enable signal, the correction object pixel detector may generate position information of the correction object pixel corresponding to the fault signal.

The valid data enable signal has a plurality of valid periods, the fault signal is a modified clock signal within the valid period, and the correction target pixel detector is activated by the vertical synchronization signal to rise a rising edge of the valid period. an y-axis information counter for increasing the y-axis coordinate for each edge of either an edge or a falling edge; And an x-axis information counter that increases an x-axis coordinate for each clock signal in the valid period, wherein the pixel to be corrected is counted by the y-axis information counter and the x-axis information counter when the fault signal is detected. Based on the y-axis coordinates and the x-axis coordinates, position information of the correction target pixel may be generated.

According to another aspect of the present invention, a method for correcting a defective pixel by an image processor connected to an image sensor includes: inserting pixel data corresponding to a pixel array and a fault signal corresponding to a defective pixel of the pixel array from the image sensor; Receiving a valid data enable signal and a vertical synchronization signal; Detecting the fault signal using the valid data enable signal and the vertical synchronization signal and generating position information of the pixel to be corrected corresponding to the fault signal, wherein the pixel to be corrected is the defective pixel; Determining a type of the pixel to be corrected using position information of the pixel to be corrected; And a defective pixel correction method of correcting the correction target pixel using a predetermined correction algorithm according to the determined type.

The determining of the type of the pixel to be corrected by using the position information of the pixel to be corrected may include setting a pixel block having a predetermined size in which the pixel to be corrected is set as the center pixel by using the position information of the pixel to be corrected. Determining whether another pixel to be corrected in the pixel block exists; And if the other pixel to be corrected in the pixel block does not exist, sets the type of the pixel to be corrected to single, and sets the type of the pixel to be corrected to a cluster when one or more other pixels to be corrected are present in the pixel block. It may include the step.

If it is determined that the type of the pixel to be corrected is a cluster, the method may further include correcting the pixel to be corrected by calculating an average value or a median value of the pixels of the pixel block.

If it is determined that the type of the pixel to be corrected is single, further comprising: setting a pixel block of a predetermined size adjacent to the pixel to be corrected, and correcting the pixel to be corrected using a difference value between the pixels in the pixel block. can do.

As the 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. In describing the drawings, similar reference numerals are used for similar elements.

Hereinafter, with reference to the accompanying drawings, it will be described in detail a preferred embodiment of the present invention. Hereinafter, the same reference numerals are used for the same components in the drawings, and duplicate descriptions of the same components are omitted. In addition, in describing the present invention, when it is determined that the detailed description of the related known technology may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted.

2 is a block diagram of an image processing apparatus according to an embodiment of the present invention, FIG. 3 is an internal functional block diagram of an image sensor according to an embodiment of the present invention, and FIG. 5 is a diagram illustrating a clock signal correspondingly output, FIG. 5 is a diagram illustrating a clock signal output corresponding to pixel data by an image sensor according to an embodiment of the present invention, and FIG. 6 is an embodiment of the present invention. 7 is a block diagram illustrating an internal functional block of an image processor according to an embodiment of the present invention. FIG. 7 is a diagram illustrating a type of a pixel to be corrected according to an embodiment of the present invention. The illustrated figure.

Hereinafter, when the image sensor 210 stores position information on a defective pixel of a pixel array in a memory during a yield test performed by a manufacturer at the time of manufacturing the image sensor 210, the image sensor 210 stores the image processor. It is assumed that when the pixel data is output to 220, the stored defective pixel position information is output together, and the image processor 220 performs correction on the defective pixel using the defective pixel position information input from the image sensor 210. This will be described.

As shown in FIG. 2, the image processing apparatus 200 according to the present invention includes an image sensor 210 and an image processor 220.

The image sensor 210 converts a signal of an optical subject incident through the lens into an electrical signal (hereinafter, a signal converted according to the subject will be referred to as "pixel data" for convenience of understanding and explanation. ) Outputs to the image processor 220.

In addition, the image sensor 210 includes a pixel array in which a plurality of unit pixels are arranged along a row or a column, such as a CCD or a CMOS sensor, and each unit pixel is an electrical signal of an object. It includes a photoelectric conversion element (for example, a photodiode) to convert to output.

In addition, the image sensor 210 detects various types of defective pixels existing in the manufacturing process of the image sensor 210 and non-volatile information on the position information of the corresponding defective pixels (hereinafter referred to as "defect pixel location information"). It is stored in the memory (not shown).

In addition, whenever the image sensor 210 outputs pixel data to the image processor 220, the image sensor 210 may change the clock signal corresponding to the defective pixel position information and output the same to the image processor 220. A method of changing and outputting a clock signal corresponding to the position information of the defective pixel by the image sensor 210 will be described in detail with reference to the accompanying drawings.

Referring to FIG. 3, the image sensor 210 includes a pixel array 310, a driving circuit 315, a timing generator 320, a gain unit 325, a clamping unit 330, an ADC 335, and defective pixel information. It includes an output unit 340 and the clock generator 345.

The pixel array 310 includes a plurality of unit pixels including a photodiode having a photoelectric conversion function. The pixel array 310 may include 330,000 pixels (640 × 480) for VGA and 1.3 million pixels (1280 × 1024) for SVGA. Can be.

The pixel array 310 sequentially outputs pixel data through pulses output from the driving circuit 315, where the driving circuit 315 outputs pixel driving pulses according to the pulses output from the timing generator 320. You can enter with).

The driving circuit 315 may include a row decoder designating a row address of the pixel array 310 and a column decoder designating a specific column address among the designated row addresses, thereby providing a predetermined exposure time. The pixel data accumulated in each unit pixel may be sequentially output in the pixel line order.

The gain unit 325 amplifies the output pixel data to a predetermined value or more, and the clamping unit 330 proposes a magnitude of the output signal so that the circuit does not operate unstable by the output signal. Do this.

The pixel data output through the gain unit 325 and the clamping unit 330 is converted from an analog signal to a digital signal through the ADC 335, and the converted digital signal is output to the image processor 220. In the present specification, a digital signal output through the ADC 335 will be referred to as "pixel data" for convenience of understanding and description.

The defective pixel information output unit 340 reads the defective pixel position information previously stored in the internal memory (not shown) of the image sensor 210 at each point in time when the image sensor 210 captures a subject and outputs corresponding pixel data. Outputs to the image processor 220.

The image sensor 210 is located at a corresponding defective pixel from a defective pixel detection unit (not shown), which is a separate device for testing a defective pixel corresponding to the pixel array in the yield inspection of the image sensor 210 after the manufacture of the image sensor 210. Information may be received and stored in internal memory.

Therefore, the defective pixel information output unit 340 may read the defective pixel position information stored in the internal memory every time the image sensor 210 outputs pixel data corresponding to the subject, and output the read defective pixel position information to the image processor 220.

In the present specification, the image sensor 210 focuses on outputting the defective pixel position information of the defective pixel corresponding to the pixel array to the image processor 220 at each time of outputting the pixel data. Naturally, a separate detector (not shown) capable of detecting a defective pixel generated corresponding to the pixel array due to the use of the image sensor 210 may be included therein.

Of course, when the detection unit for detecting a defective pixel corresponding to the pixel array is included in the image sensor 210 as described above, position information corresponding to the defective pixel detected by the corresponding detection unit may be stored in the internal memory. Naturally, the position information may be output together with each time point of outputting the pixel data.

 Hereinafter, a defective pixel is referred to herein as a "defective pixel". That is, the defective pixel may be a pixel in which a defect is generated during the manufacturing process of the image sensor 210, or may be a pixel in which a defect occurs in a later use process.

The method of detecting a defective pixel of the image sensor 210 varies according to a manufacturer who manufactures the image sensor, and the method of detecting a defective pixel is obvious to those skilled in the art, and thus a separate description thereof will be omitted.

The clock generator 345 may receive a vertical synchronization signal vsync and a valid data enable signal href for the operation of the image processor 220 to obtain pixel data input from the image sensor 210. ) The vertical synchronization signal vsync may be a synchronization signal for displaying an effective section of pixel data output to the image processor 220 through the image sensor 210.

The valid data enable signal href may be a clock signal for acquiring pixel data output from the image sensor 210 by the image processor 220. For example, the image processor 220 may have a vertical sync signal vsync in a second state (eg, low), and a valid data enable signal href in a first state (eg, high). (high), each pixel data can be obtained.

Referring to FIG. 5, the clock generation unit 345 according to the present invention changes and outputs a clock 520 according to the defective pixel position information output by the defective pixel information output unit 340. In addition, although not shown in FIG. 3, the pixel data output through the image sensor 210 may include a clock signal generated by the clock generator 345 (eg, a vertical synchronization signal vsync and a valid data enable signal). (href) may further include a synchronization unit (not shown) for synchronization. Since the pixel data is synchronized and outputted by the image sensor 210 by the synchronizer and the image processor 220 acquires the pixel data corresponding to the clock signal, it is obvious to those skilled in the art, so a separate description of the synchronization Will be omitted.

As a result, the image processor 220 causes the valid data enable signal href having a predetermined period 510, 515 to be predetermined according to the pixel data output through the image sensor 210 within the predetermined period. By detecting a sudden change, the position information of the defective pixel can be generated. This will be described in detail with reference to the accompanying drawings below.

4 illustrates a vertical synchronization signal vsync and a valid data enable signal href output by the clock generator 345 to acquire pixel data according to a conventional method. It can be seen that the conventional valid data enable signal href is output at a constant period corresponding to the data output through the image sensor 210.

Referring to FIG. 4, the vertical synchronization signal vsync may be a signal indicating the start of each image frame corresponding to pixel data output through the image sensor 210. In this specification, a description will be focused on generating position information of a pixel to be corrected using the vertical sync signal vsync and the valid data enable signal href0, but the horizontal sync signal hsync and the valid data enable signal according to an implementation method. Naturally, the position information of the correction target pixel may be generated using (vref).

Hereinafter, it will be described on the assumption that the valid data enable signal is activated after the vertical synchronization signal vsync becomes the second state (a falling edge). Of course, the valid data enable signal may be implemented after the vertical synchronization signal vsync becomes the rising edge.

The valid data enable signal href has a plurality of valid periods, and after the vertical synchronization signal vsync becomes the second state, the x-axis information counter is updated within the valid period and is updated to the updated count of the corresponding x-axis information counter. X-axis coordinates can be generated.

In addition, the y-axis information counter may be updated at each time a valid period in the valid data enable signal href starts, thereby generating y-axis coordinates by the updated count of the corresponding y-axis information counter.

Therefore, the pixel to be corrected detector 610 may acquire the position information of the pixel data using the counts updated by the y-axis information counter and the x-axis information counter, respectively. This will be described in detail with reference to the accompanying drawings below.

As described above, the image sensor 210 detects a defective pixel corresponding to the pixel array, stores the detected defect pixel position information in a memory, and outputs pixel data, and then outputs the valid data corresponding to the defective pixel position information. The clock of the enable signal href may be changed and output. As described above, the image sensor 210 outputs pixel data output to the image processor 220 by a clock signal generated by the clock generator 345 (for example, a vertical synchronization signal vsync and valid data). The method may further include synchronization for synchronization with the enable signal href.

Hereinafter, a clock changed corresponding to defective pixel position information within a predetermined period of the valid data enable signal href is referred to as a "fault signal" for convenience of understanding and description. This will be described in detail with reference to the accompanying drawings below.

Referring back to FIG. 2, the image processor 220 may include a clock signal (eg, a vertical sync signal vsync, a valid data enable signal href, etc.) including pixel data and a fault signal from the image sensor 210. ) Detects the fault signal, generates position information of the correction target pixel, and corrects and outputs the correction target pixel.

Referring to FIG. 6, the image processor 220 according to the present invention includes a pixel to be corrected detector 610, a defective pixel corrector 620, and a controller 630.

The correction target pixel detection unit 610 receives the vertical synchronization signal vsync, the valid data enable signal href, and the pixel data from the image sensor 210 and is included in the valid period of the valid data enable signal href. A function of detecting a fault signal and generating position information of a correction target pixel and outputting the position information to the defective pixel correction unit 620.

The correction target pixel detector 610 receives a vertical sync signal vsync and a valid data enable signal href as shown in FIG. 5.

For convenience of understanding and explanation, the period in which the valid data enable signal href is in the first state (eg, high) will be referred to as the "valid periods 510 and 515." In the present specification, a description is made on the assumption that the clock is detected by rising edges 510a and 510b for obtaining valid data, and may be implemented to be detected by a falling edge according to an implementation method. Of course.

In addition, the fault signal may be referred to as a clock signal changed within a valid period within the valid data enable signal href after the vertical synchronization signal vsync becomes a second state (for example, a falling edge). do.

라고 가정하자. 여기서,

는 한 클럭을 지칭하며, n은 이미지 센서(210)가 출력하는 한 행(row)의 픽셀의 개수일 수 있다. 유효 주기가

이면, 폴트 신호(도 5의 520 참조)는

일 수 있다. If the valid period (see 510, 515 of Figure 5) is for example

Suppose here,

Denotes one clock, and n may indicate the number of pixels of one row output by the image sensor 210. The effective period

, The fault signal (see 520 in FIG. 5)

Can be.

The correction target pixel detection unit 610 is activated by the vertical synchronization signal vsync, and increases the y-axis information counter every time the valid period becomes a second state (for example, a rising edge). And an x-axis information counter (not shown) that increments the x-axis information for each clock signal within an effective period (not shown). Since a method of operating a counter internally according to the operation of a clock is obvious to those skilled in the art, a separate description thereof will be omitted.

Accordingly, the pixel to be corrected detector 610 may generate position information of the pixel data input from the image sensor 210 by using information counted by the y-axis information counter and the x-axis information counter, respectively.

보다 작으면, 해당 시점을 폴트 신호로 감지할 수 있다. 그리고, 보정 대상 픽셀 검출부(610)는 해당 폴트 신호가 감지된 시점의 x축 정보 카운터 및 y축 정보 카운터에 의해 각각 카운트된 정보(즉, x축 좌표 및 y축 좌표)를 이용하여 보정 대상 픽셀의 위치 정보를 생성할 수 있다.Accordingly, the correction target pixel detector 610 detects a point in time at which the falling edge occurs after the first state is reached within the effective period, and thus, the point in time at which the corresponding falling edge is reached.

If smaller, the corresponding time point may be detected as a fault signal. The pixel to be corrected 610 detects a pixel to be corrected using information counted by the x-axis information counter and the y-axis information counter at the time when the corresponding fault signal is detected (that is, the x-axis coordinate and the y-axis coordinate). Location information may be generated.

Of course, when the fault signal is detected within the effective period, the correction target pixel detector 610 may control the y-axis information counter to not increase the y-axis coordinates, but to increase the x-axis coordinates only.

In addition, the correction target pixel detector 610 may generate the type information of the corresponding correction target pixel by using the generated position information of the correction target pixel and output the generated information to the defective pixel correction unit 420.

Referring to FIGS. 7 and 8, the correction target pixel detection unit 610 sets a pixel block having a predetermined size centering on the correction target pixel to determine whether other correction target pixels exist in the pixel block. Can generate type information.

In the present specification, "single" will be defined as an address corresponding to another pixel to be corrected in the X-axis or Y-axis direction adjacent to the address corresponding to the pixel to be corrected.

Referring to FIG. 7, when there is no other correction target pixel in the pixel block based on the correction target pixel 710, the type information of the correction target pixel may be set to single.

In addition, "cluster" will be defined as one or more addresses corresponding to the pixel to be corrected in the X-axis or Y-axis direction adjacent to the address corresponding to the pixel to be corrected.

Referring to FIG. 8, unlike the single unit, when another correction target pixel 815 exists in a predetermined pixel block, type information of the correction target pixel may be recognized as a cluster. 6 and 7 assume that the size of the pixel block is 3, but it can be set to any size according to the implementation method.

As such, the pixel to be corrected detector 410 may generate position information and type information of the pixel to be corrected and output the same to the defective pixel corrector 420.

The defective pixel corrector 620 receives the position information and the type information of the pixel to be corrected from the pixel to be corrected 610 and corrects the pixel to be corrected according to a predetermined correction algorithm corresponding to the type information of the pixel to be corrected. The corrected pixel data is output to the controller 630.

If the type of the pixel to be corrected is single, the defective pixel corrector 620 sets the pixel to be corrected as the center pixel, and then sets a pixel block of a predetermined size to determine adjacent pixels except for the center pixel (ie, the pixel to be corrected). By calculating the difference value of these, correction may be performed on the center pixel using adjacent pixels having the minimum difference value. That is, the difference between the two pixels may be calculated based on the center pixel, and the average value of the two pixels having the minimum calculated difference may be calculated and replaced with the center pixel.

For example, assume that the center pixel is a pixel to be corrected within a 3x3 pixel block. In this case, the defective pixel corrector 620 calculates a difference between eight pixels except for the correction target pixel. Here, the calculated difference value may be an absolute value. The defective pixel corrector 620 may correct the value by replacing the center pixel with an average value of two pixels having the smallest calculated difference values.

However, if the type of the pixel to be corrected is a cluster, the defective pixel corrector 620 calculates an average value or a median value of pixels in a region corresponding to a cluster unit (hereinafter, referred to as a "defective pixel region" for convenience). Correction may be performed on the correction target pack using the calculated average value or the median value.

The controller 630 controls the internal components of the image processor 220 according to the present invention (eg, the pixel to be corrected detector 610, the defective pixel corrector 620, etc.).

In addition, the controller 630 may receive the corrected pixel data from the defective pixel corrector 620 and output the corrected pixel data to a post-processing processor (not shown) or store it in a storage unit (not shown).

9 is a flowchart illustrating a method of correcting a defective pixel by an image processor according to an exemplary embodiment of the present invention. Hereinafter, the image processor 220 receives a fault signal corresponding to the position information of the defective pixel from the image sensor 210, detects the corresponding fault signal, detects the corrected pixel, and corrects the corrected pixel. Let's explain. Here, as described above, the fault signal is a clock signal inserted into the valid period of the valid data enable signal by the image sensor 210 corresponding to the position information of the defective pixel, and the image processor 220 detects the corresponding fault signal. After generating position information of the correction target pixel according to the fault signal, the correction may be performed on the correction target pixel. Here, the method of detecting the defective pixel by the image sensor 210 is different for each manufacturer that manufactures the image sensor and the method is obvious to those skilled in the art, so a separate description thereof will be omitted.

In operation 910, the pixel to be corrected 610 detects pixel data corresponding to an optical object, a vertical synchronization signal (vsync) necessary for acquiring pixel data, and effective data from the image sensor 210 under the control of the controller 630. It receives an enable signal href.

In operation 915, the pixel to be corrected 610 detects a fault signal using the vertical synchronization signal vsync and the valid data enable signal href input from the image sensor 210 under the control of the controller 630. ) Is detected, the position information of the pixel to be corrected is generated according to the fault signal, and output to the defective pixel correcting unit 620.

For example, the correction target pixel detection unit 610 includes a y-axis information counter and an x-axis information counter, and is activated by the vertical synchronization signal vsyn so that the valid period in the valid data enable signal href is in a first state. The y-axis information counter increments the y-axis coordinate at each of the time points 510a and 510b. The x-axis information counter is operated whenever the valid period becomes the first state, and increases the x-axis coordinate for every clock signal in the valid period.

As described above, the correction target pixel detector 610 detects a fault signal within an effective period, and updates the x-axis coordinate and the y-axis coordinate updated by the x-axis information counter and the y-axis information counter at the time when the fault signal is detected. By using the position information of the pixel to be corrected can be generated. The correction target pixel detector 610 may control the y-axis information counter not to increase the y-axis information when the fault signal is detected within the valid period.

In operation 920, the correction target pixel detector 610 generates the type information of the correction target pixel using the position information of the correction pixel under the control of the controller 630, and outputs the type information of the correction pixel to the defective pixel corrector 620. Here, when the type information of the correction target pixel is generated, the correction target pixel detector 610 may control to output the type information of the correction target pixel and the position information of the correction target pixel to the defective pixel correction unit 620.

For example, the pixel to be corrected detector 610 sets a pixel block having a predetermined size centering on the pixel to be corrected using the position information of the pixel to be corrected (see FIGS. 7 and 8). The type of the correction target pixel may be generated by determining whether the correction target pixel exists.

As described above, if one or more other correction target pixels are present in the pixel block, the correction target pixel detector 610 may determine the type of the correction target pixel as a cluster. However, if there is no other pixel to be corrected in the pixel block, the pixel to be corrected detector 610 may set the type of the pixel to be corrected to single.

In operation 925, the defective pixel corrector 620 determines whether the type information of the correction target pixel input from the correction target pixel detector 610 is single under the control of the controller 630.

If it is determined that the type of the pixel to be corrected is single, the defective pixel corrector 620 may control a pixel block of a predetermined size adjacent to the pixel to be corrected (for example, 3 × 3) under the control of the controller 630 in step 930. ), Where the correction target pixel may be set to the center pixel of the pixel block, the correction may be performed on the center pixel (ie, the correction target pixel) by using a difference value between two pixels except the center pixel. .

For example, the defective pixel corrector 620 may correct the center pixel (that is, the pixel to be corrected) by calculating an average value of two pixels having a minimum difference value between two pixels except the center pixel in the pixel block. .

However, if it is determined that the type of the pixel to be corrected is a cluster, in step 935, the defective pixel corrector 620 sets a pixel block having a predetermined size using the position information of the pixel to be corrected, and averages or intermediates the pixels in the pixel block. The value may be calculated to perform correction on the correction target pixel.

In the present invention, for convenience of description, it is described that the pixel data is input to the image processor in units of rows. However, the above description is applicable to the case where the pixel data is input in units of columns.

As described above, by providing the method and apparatus for correcting the defective pixel according to the present invention, the image processor uses the valid data enable signal and the vertical synchronizing signal (i.e., the clock signal) inputted from the image sensor (i.e., the clock signal). , The correction target pixel) can be detected and the correction can be performed on the correction target pixel.

In addition, the present invention has the effect that the image processor can detect and correct the defective pixel without receiving a separate control signal from the image sensor for detecting the defective pixel.

In addition, the present invention has the effect that the product processor can be improved because the image processor does not perform a separate operation for detecting the defective pixels of the pixel array of the image sensor.

In addition, the present invention has the effect that the image processor can perform the correct correction for the defective pixel without performing an operation for detecting the defective pixel.

Although the above has been described with reference to a preferred embodiment of the present invention, those skilled in the art to which the present invention pertains without departing from the spirit and scope of the present invention as set forth in the claims below It will be appreciated that modifications and variations can be made.

Claims (16)

The pixel data corresponding to the pixel array, the valid data enable signal in which the fault signal is inserted corresponding to the defective pixel of the pixel array, and the vertical synchronization signal are inputted from the image sensor to generate position information of the pixel to be corrected. Image processor to calibrate. The method of claim 1, A correction target pixel detector configured to receive the pixel data, the valid data enable signal, and the vertical synchronization signal, and generate position information of the correction target pixel and type information of the correction target pixel using the fault signal; And And a defective pixel correcting unit configured to correct the corrected pixel using a predetermined correction algorithm according to type information of the corrected pixel using the position information of the corrected pixel. The method of claim 2, And when the fault signal is detected in the valid data enable signal, the correcting pixel detection unit generates position information of the correcting pixel according to the fault signal. The method of claim 3, wherein The valid data enable signal has a plurality of valid periods, the fault signal is a modified clock signal within the valid period, The correction target pixel detection unit, A y-axis information counter which is activated by the vertical synchronization signal and increases the y-axis coordinate for each edge of either the rising edge or the falling edge of the valid period; And An x-axis information counter for increasing an x-axis coordinate for each clock signal in the valid period, The correcting pixel detection unit generates position information of the correcting pixel using the y-axis coordinates and the x-axis coordinates counted by the y-axis information counter and the x-axis information counter when the fault signal is detected. Characterized by an image processor. The method of claim 2, The pixel to be corrected is configured to determine whether another pixel to be corrected exists in the pixel block after setting a pixel block having a predetermined size using the position information of the pixel to be corrected as the center pixel. In this case, the image processor, characterized in that for setting the type of the correction target pixel to the cluster. The method of claim 5, And the defective pixel correcting unit corrects the corrected pixel by calculating an average value or a median value of the pixels of the pixel block including the corrected pixel when the type of the corrected pixel is a cluster. The method of claim 5, And the correcting pixel detection unit determines whether another correcting pixel exists in the pixel block, and sets the type of the correcting pixel to single if no other correcting pixel exists. The method of claim 7, wherein And the defective pixel corrector corrects the corrected pixel using a difference value between pixels except the center pixel in the pixel block. An image sensor configured to output a valid data enable signal, a vertical synchronization signal, and pixel data corresponding to the pixel array, in which a fault signal is inserted, corresponding to a defective pixel of a pixel array in which a plurality of unit pixels are disposed; And And an image processor configured to generate position information of a pixel to be corrected using the valid data enable signal and the vertical synchronization signal to correct the pixel to be corrected. The method of claim 9, The image processor, A correction target pixel detector configured to receive the pixel data, the valid data enable signal, and the vertical synchronization signal, and generate position information of the correction target pixel and type information of the correction target pixel using the fault signal; And And a defective pixel corrector configured to correct the corrected pixel using a predetermined correction algorithm according to type information of the corrected pixel using position information of the corrected pixel. The method of claim 10, And when the fault signal is detected in the valid data enable signal, the correcting pixel detection unit generates position information of the correcting pixel according to the fault signal. The method of claim 11, The valid data enable signal has a plurality of valid periods, the fault signal is a modified clock signal within the valid period, The correction target pixel detection unit, A y-axis information counter which is activated by the vertical synchronization signal and increases the y-axis coordinate for each edge of either the rising edge or the falling edge of the valid period; And An x-axis information counter for increasing an x-axis coordinate for each clock signal in the valid period, The correcting pixel detection unit generates position information of the correcting pixel using the y-axis coordinates and the x-axis coordinates counted by the y-axis information counter and the x-axis information counter when the fault signal is detected. An image processing apparatus. An image processor connected to an image sensor corrects defective pixels, Receiving a valid data enable signal and a vertical synchronization signal in which pixel data corresponding to a pixel array and a fault signal corresponding to a defective pixel of the pixel array are inserted from the image sensor; Detecting the fault signal using the valid data enable signal and the vertical synchronization signal and generating position information of the pixel to be corrected corresponding to the fault signal, wherein the pixel to be corrected is the defective pixel; Determining a type of the pixel to be corrected using position information of the pixel to be corrected; And And correcting the pixel to be corrected using a predetermined correction algorithm according to the determined type. The method of claim 13, Determining the type of the correction target pixel using the position information of the correction target pixel, Determining whether there is another correction target pixel in the pixel block by setting a pixel block having a predetermined size in which the correction target pixel is set as a center pixel using the position information of the correction target pixel; And If there is no other pixel to be corrected in the pixel block, the type of the pixel to be corrected is set to single, and if there is at least one other pixel to be corrected in the pixel block, the type of the pixel to be corrected is set to a cluster. And a defective pixel correction method comprising the steps of: The method of claim 14, If it is determined that the type of the pixel to be corrected is a cluster, correcting the pixel to be corrected by calculating an average value or a median value of the pixels of the pixel block. The method of claim 14, If it is determined that the type of the pixel to be corrected is single, further comprising: setting a pixel block of a predetermined size adjacent to the pixel to be corrected, and correcting the pixel to be corrected using a difference value between the pixels in the pixel block. Defect pixel correction method.

Images