Image sensor with scaler and image scaling method thereof

Abstract

An image sensor is wieldy used in many fields, including medical field and security device. Specifically, the image sensor is most widely used in digital camera and mobile phone. The digital camera and the mobile phone requires capture image of higher resolution and higher quality. However, a preview size of the mobile phone or digital camera requires a small size because of a display limitation. Therefore, a function of reducing an image size or magnifying a specific portion of the picture is essential in the image sensor. Accordingly, there is provided an image sensor with a scaler. The image sensor with the scaler can arbitrarily adjust a size of an image without any additional scaling chip.

Claims

1. A scaling system comprising: a scaling ratio generator configured to: generate a pre-scaling ratio according to a scaling ratio; and generate a post-scaling ratio according to the scaling ratio; a pre-scaler configured to generate pre-scaler image data by scaling original image data according to the pre-scaling ratio; and a post-scaler configured to: receive the original image data; receive the pre-scaler image data; and generate post-scaler image data by selectively scaling one of the original image data and the pre-scaler image data according to the post-scaling ratio.

2. The scaling system of claim 1, wherein the post-scaler is further configured to: generate the post-scaler image data by scaling the original image data according to a first post-scaling ratio if the scaling ratio is greater than a reference value; and generate the post-scaler image data by scaling the pre-scaler image data according to a second post-scaling ratio if the scaling ratio is less than the reference value.

3. The scaling system of claim 1, wherein the pre-scaler is further configured to selectively generate the pre-scaler image data if the scaling ratio is less than a reference value.

4. The scaling system of claim 1, wherein the post-scaler is further configured to: generate the post-scaler image data according to a full-color scaling if the original image data comprises full-color data; and generate the post-scaler image data according to a bayer mosaic scaling if the original image data comprises bayer color data.

5. The scaling system of claim 1, wherein the post-scaling ratio is proportional to the scaling ratio and proportional to the pre-scaling ratio.

6. The scaling system of claim 1, further comprising an output data synchronizer configured to generate synchronized scaler image data from the post-scaler image data.

7. A method of scaling image data, the method comprising: generating, with a scaling ratio generator, a pre-scaling ratio according to a scaling ratio; generating, with the scaling ratio generator, a post-scaling ratio according to the scaling ratio; scaling original image data according to the pre-scaling ratio to generate pre-scaler image data; and selectively scaling one of the original image data and the pre-scaler image data according to the post-scaling ratio to generate post-scaler image data.

8. The method of claim 7, further comprising: scaling the original image data according to a first post-scaling ratio to generate the post-scaler image data if the scaling ratio is greater than a reference value; and scaling the pre-scaler image data according to a second post-scaling ratio to generate the post-scaler image data if the scaling ratio is less than the reference value.

9. The method of claim 8, further comprising scaling the original image data to generate the pre-scaler image data if the scaling ratio is less than the reference value.

10. The method of claim 7, further comprising: generating the post-scaler image data according to a full-color scaling if the original image data comprises full-color data; and generating the post-scaler image data according to a bayer mosaic scaling if the original image data comprises bayer color data.

11. The method of claim 7, wherein the post-scaling ratio is proportional to the scaling ratio and proportional to the pre-scaling ratio.

12. The method of claim 7, further comprising synchronizing the post-scaler image data to generate synchronized scaler image data.

13. A scaler comprising: a scaling ratio generator configured to output a pre-scaling ratio, a post-scaling ratio, and a pre-scaler valid signal; and a post-scaler configured to: receive the post-scaling ratio and the pre-scaler valid signal; scale original image data according to the post-scaling ratio if the pre-scaler valid signal is off to generate post-scaler image data; and scale pre-scaler image data according to the post-scaling ratio if the pre-scaler valid signal is on to generate the post-scaler image data.

14. The scaler of claim 13, further comprising a pre-scaler configured to: receive the pre-scaling ratio and the pre-scaler valid signal; and scale the original image data according to the pre-scaling ratio to form the pre-scaler image data if the pre-scaler valid signal is on.

15. The scaler of claim 13, wherein the scaling ratio is further configured to: generate the pre-scaling ratio and the post-scaling ratio according to a scaling ratio; and wherein the post-scaling ratio is proportional to the scaling ratio and the pre-scaling ratio.

16. The scaler of claim 13, wherein the post-scaler is further configured to: generate the post-scaler image data according to a full-color scaling if the original image data comprises full-color data; and generate the post-scaler image data according to a bayer mosaic scaling if the original image data comprises bayer color data.

17. The scaler of claim 13, wherein the post-scaler is further configured to: receive addresses from a timing controller if the pre-scaler valid signal is off; and receive addresses from the pre-scaler if the pre-scaler valid signal is on.

18. The scaler of claim 17, wherein: the addresses comprise a first set of values if the original image data comprises full-color data; and the addresses comprise a second set of values if the original image data comprises bayer color data.

19. A system for scaling image data, the system comprising: a means for generating a pre-scaling ratio according to a scaling ratio and for generating a post-scaling ratio according to the scaling ratio; a means for generating pre-scaler image data by scaling original image data according to the pre-scaling ratio; and a means for generating post-scaler image data by selectively scaling one of the original image data and the pre-scaler image data according to the post-scaling ratio.

20. The system of claim 19, further comprising: a means for scaling the original image data according to a first post-scaling ratio to generate the post-scaler image data if the scaling ratio is greater than a reference value; and a means for scaling the pre-scaler image data according to a second post-scaling ratio to generate the post-scaler image data if the scaling ratio is less than the reference value.

21. The system of claim 20, further comprising a means for scaling the original image data to generate the pre-scaler image data if the scaling ratio is less than the reference value.

22. The system of claim 19, further comprising a means for generating the post-scaler image data according to a full-color scaling if the original image data comprises full-color data; and a means for generating the post-scaler image data according to a bayer mosaic scaling if the original image data comprises bayer color data.

23. The system of claim 19, wherein the post-scaling ratio is proportional to the scaling ratio and proportional to the pre-scaling ratio.

24. The system of claim 19, further comprising a means for synchronizing the post-scaler image data to generate synchronized scaler image data.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The above and other objects and features of the present invention will become apparent from the following description of the preferred embodiments given in conjunction with the accompanying drawings, in which:

(2) FIG. 1 is a block diagram of an image sensor;

(3) FIG. 2 is a block diagram of a CMOS image sensor in accordance with an embodiment of the present invention;

(4) FIG. 3 is a block diagram of a scaler of FIG. 2;

(5) FIG. 4 is a flowchart diagram illustrating an image scaling algorithm of the scaler in FIG. 2;

(6) FIG. 5 is a block diagram of a scaling ratio generator of in FIG. 3;

(7) FIG. 6 is a block diagram of a pre-scaler of FIG. 3;

(8) FIG. 7 is a block diagram of a post-scaler of FIG. 3;

(9) FIG. 8A illustrates an output of the pre-scaler in a full color data level when the pre-scaling ratio is 1/4;

(10) FIG. 8B illustrates an output of the pre-scaler in a bayer data level when the pre-scaling ratio is 1/4;

(11) FIG. 9A illustrates an output of the post-scaler in a full color data level when the post-scaling ratio is 5/7 with respect to a post 1414 pixel array;

(12) FIG. 9B illustrates an output of the post-scaler in a bayer data level when the post-scaling ratio is 5/7 with respect to a post 1414 pixel array; and

(13) FIG. 10 is a flowchart diagram illustrating an algorithm of generating row/column addresses for the image scaling.

DETAILED DESCRIPTION OF THE INVENTION

(14) An image sensor with a scaler and an image scaling method thereof in accordance with exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

(15) FIG. 2 is a block diagram of a CMOS image sensor in accordance with an embodiment of the present invention.

(16) Referring to FIG. 2, the CMOS image sensor includes a pixel array 20 with MN unit pixels (M and N are positive integers), a timing controller 21, an analog signal processor 22, an image signal processor 24, a line memory 23, and a scaler 25.

(17) The analog signal processor 22 is provided with a CDS and analog amplifier 221, and a CDS 222.

(18) The timing controller 21 is called a control and external system interface. The timing controller 11 controls an overall operation of the image sensor using an FSM and serves as an interface for an external system. Also, since the timing controller 21 has a batch register, it can program contents related to several internal operations and controls an overall chip operation according to the program information.

(19) The pixel array 20 includes MN unit pixels configured to maximize light-sensitive characteristics. The pixel array 20 is a core of the image sensor and detects information about an image inputted from the outside.

(20) In the CDS and analog amplifier 221, the CDS removes a fixed pattern noise of a pixel using a CDS method, and the analog amplifier converts a pixel signal into an electric signal.

(21) The ADC 222 converts an analog voltage detected by each pixel of the pixel array 20 into a digital voltage that can be processed at a digital system.

(22) The line memory 23 stores the digital voltage of the pixel, which is converted by the ADC 222. The line memory 23 includes a plurality of lines for executing various functions of the image signal processor 24.

(23) The image signal processor 24 executes several functions for improving the performance of the image sensor, based on the pixel output value stored in the line memory 23. Examples of the functions are a color interpolation, a color correction, a gamma correction, an auto white balance, an auto exposure, and so on.

(24) The scaler 25 receives a scaling factor for adjusting the image size and calculates a scaling ratio. Then, the scaler 25 creates row/column address and data corresponding to the scaling ratio and outputs an image with a desired size.

(25) FIG. 3 is a block diagram of the scaler 25 illustrated in FIG. 2.

(26) Referring to FIG. 3, the scaler 25 includes a scaling ratio generator 250 for determining the scaling ratio according to the scaling factors SCALEM and SCALEN, a pre-scaler 251 for selectively executing the pre-scaling under control of the scaling ratio generator 250 and the timing controller 21, a post-scaler 252 for selectively executing the post-scaling with respect to the pre-scaled image and the non-pre-scaled image, and an output data synchronizer 253 for synchronizing the scaled image from the post-scaler 252 with the output clock of the image sensor.

(27) An operation of the scaler 25 will be described in detail below.

(28) The scaling ratio generator 250 calculates the scaling ratio using the scaling factors SCALEM and SCALEN. The scaling ratio is expressed as
Scaling ratio=SCALEM/SCALEN(1)

(29) Also, the scaling ratio generator 250 determines a pre-scaler valid signal and a pre-scaling ratio according to the scaling ratio. The pre-scaler valid signal determines on/off operation of the pre-scaler 251.

(30) Only when the pre-scaler valid signal is generated, the pre-scaler 251 generates pre-scaler row/column addresses according to the pre-scaling ratio calculated by the scaling ratio generator 250. Also, the pre-scaler 251 generates an image of an appropriate size by sampling only data located at positions matched with the generated addresses.

(31) The post-scaler 252 adjusts a final image size of the image scaler 25. When not passing through the pre-scaler 251, the scaling ratio generated from the scaling ratio generator 250 becomes the post-scaling ratio and thus the size of the final image is adjusted according to the scaling ratio. On the contrary, when passing through the pre-scaler 251, the post-scaling ratio of the post-scaler 252 is given by
Post scaling ratio=[(SCALEM/SCALEN)/Scaling ratio](2)

(32) When the pre-scaler 251 is off, the post-scaler 252 receives the row/column addresses from the timing controller 21 to generate new row/column addresses according to the post-scaling ratio. On the contrary, when the pre-scaler 251 is on, the post-scaler 252 receives the row/column addresses from the pre-scaler 251 to generate new row/column addresses according to the post-scaling ratio.

(33) Since the output timing of the image sensor due to the post-scaler 252 is not constant, the output data synchronizer 253 receives image data of the post scaler 252 and then synchronizes the scaled image data with the output clock of the image sensor.

(34) FIG. 4 is a flowchart diagram illustrating an image scaling algorithm of the scaler of FIG. 2.

(35) Referring to FIG. 4, in the step S901, if scaling factors SCALEM and SCALEN for outputting an image with a desired size are transferred through the timing controller 21, a scaling ratio is generated. Then, in step S902, it is determined whether to execute a pre-scaler according to the scaling ratio.

(36) In step S903, if it is determined that the pre-scaler is necessary to execute, a pre-scaling using the pre-scaler is performed to adjust the image size. Then, in step S904, it is determined whether the post-scaler is necessary to execute. In step S905, if the post-scaler is necessary to execute, a post-scaling is performed through the post-scaler to adjust the image size. If the post-scaler is unnecessary to execute, the image size obtained through the pre-scaling is finally outputted.

(37) On the contrary, if it is determined in step S902 that the pre-scaling is not unnecessary to execute, the pre-scaling is bypassed and the image size obtained through the post-scaling in step S905 is finally outputted.

(38) In step S906, the image data scaled through the pre-scaling and the post-scaling is synchronized with the output clock of the image sensor. That is, data timing synchronization process is performed.

(39) FIG. 5 is a block diagram of the scaling ratio generator 250 illustrated in FIG. 3.

(40) Referring to FIG. 5, the scaling ratio generator 250 includes a divider 250-1, a plurality of comparators 250-2A to 250-2C, a selector 250-3, a post-scaling ratio determiner 250-4. The divider 250-1 outputs the scaling ratio given by dividing SCALEM by SCALEN using the scaling factor. The plurality of comparators 250-2A to 250-2C have different reference values ranging from 1/2 to 1/8 for comparing the scaling ratio with a preset scaling ratio. The selector 250-3 outputs the corresponding pre-scaling ratio among the outputs of the comparators 250-2A to 250-2C. The post-scaling ratio determiner 250-4 determines the post-scaling ratio using the pre-scaling valid signal (the output of the comparator 250-2C having the scaling ratio reference value of 1/2) and the selected pre-scaling ratio.

(41) Although the post-scaler 250 having the scaling ratio of 1 to 1/8 has been described in the above embodiment, the present invention can also be applied to a greater or smaller scaling ratio.

(42) An operation of the post-scaler 250 will be described below in detail.

(43) The divider 250-1 calculates the scaling ratio by dividing the scaler factor SCALEM by the scaler factor SCALEN. The scaling ratio outputted from the divider 250-1 is inputted to negative () terminals of the comparators 250-2A to 250-2C having the reference values 1/2, 1/4, and 1/8, respectively.

(44) When the scaling ratio is greater than 1/2, the outputs of all the comparators 250-2A to 250-2C are a logic low and the scaler valid signal is also a logic low. Consequently, the pre-scaler 251 is turned off. At this point, the scaling ratio becomes SCALEM/SCALEN.

(45) When the scaling ratio is less than or equal to 1/2 and greater than 1/4, the output of only the 1/2 comparator 250-2C is a logic high. The pre-scaling ratio becomes 1/2. At this point, the post-scaling ratio becomes (SCALEM/SCALEN)/(1/2).

(46) When the scaling ratio is less than or equal to 1/4 and greater than 1/8, the outputs of only the 1/2 comparator 250-2C and the 1/4 comparator 250-2B are a logic high. The pre-scaling ratio becomes 1/4. At this point, the post-scaling ratio becomes (SCALEM/SCALEN)/(1/4).

(47) When the scaling ratio is greater than 1/2, the scaler valid signal is a logic low. When the scaling ratio is less than or equal to 1/2, the post-scaling ratio is a logic high.

(48) FIG. 6 is a block diagram of the pre-scaler 251 illustrated in FIG. 3.

(49) Referring to FIG. 6, the pre-scaler 251 includes a plurality of AND gates 251-1A to 251-1C, an address shift coefficient selector 251-2, a pre-scaler address generator 251-3, and a pre-scaler data sampler 251-4. The AND gate 251-1A has one terminal receiving the reference value of 1/8 and another terminal receiving the pre-scaling ratio, the AND gate 251-1B has one terminal receiving the reference value of 1/4 and another terminal receiving the pre-scaling ratio, and the AND gate 251-1C has one terminal receiving the reference value of 1/2 and another terminal receiving the pre-scaling ratio. The address shift coefficient selector 251-2 selects the shift coefficient of the corresponding address using the outputs of the AND gates 251-1A to 251-1C. The pre-scaler address generator 251-3 generates the pre-scaler row/column addresses by shifting the addresses from the timing controller 21 as much as the corresponding coefficient selected by the address shift coefficient selector 251-2. The pre-scaler data sampler 251-4 samples the original image data and the data stored in the line memory 23 through the pre-scaler row/column addresses and outputs the pre-scaler image data.

(50) The pre-scaler 251 operates only when the scaler valid signal is a logic high, and shifts the row/column addresses provided from the timing controller 21 in each bit according to the pre-scaling ratio determined by the scaler ratio generator 250. That is, a 1-bit shift is performed when the scaling ratio is 1/2, a 2-bit shift is performed when the scaling ratio is 1/4, and a 3-bit shift is performed when the scaling ratio is 1/8. Through this shift process, the pre-scaler row/column addresses are generated.

(51) Meanwhile, even in the same pre-scaling ratio, the pre-scaler row/column addresses are differently generated when the image scaler is performed in the full color data level and in the bayer data level.

(52) By outputting only data located at positions matched with the pre-scaler address position, which is newly constructed among the original image data through the pre-scaler data sampler 251-4, the output data of the image sensor can be implemented to match with the pre-scaling ratio.

(53) In various manners, the output data of the pre-scaler 251 can be determined using the data of the previous line and the currently inputted data through the line memory 23 at the position where the address is matched.

(54) Accordingly, the image quality of the pre-scaler output image is determined by the method of determining the output data.

(55) FIG. 7 is a block diagram of the post-scaler 252 illustrated in FIG. 3.

(56) Referring to FIG. 7, the post-scaler 252 includes a post-scaler address generator 252-1, transmission gates 252-3A and 252-3B, and a post-scaler data sampler 252-2. The post-scaler address generator 252-1 receives the row/column addresses from the timing controller 21, the pre-scaler row/column addresses, the pre-scaler valid signal, and the post-scaling ratio and generates the post-scaler row/column addresses. The transmission gate 252-3A receives the pre-scaler valid signal and its inverted signal through gates of NMOS and PMOS transistors and outputs the pre-scaler image data. The transmission gate 252-3B receives the pre-scaler valid signal and its inverted signal through gates of NMOS and PMOS transistors and outputs the original image data. The post-scaler data sampler 252-2 samples the original image data or the pre-scaler image data through the post-scaler row/column addresses and outputs the post-scaler image data.

(57) The post-scaler address generator 252-1 selects one of the pre-scaler row/column post-scaling ratio and the row/column address of the entire size of the image sensor, which is generated by the timing controller 21 by using the post-scaling ratio and the pre-scaling valid signal in Eqs. (1) and (2), and then generates the post-scaler row/column addresses. At this point, like in the pre-scaler 251, the post-scaler row/column addresses are differently generated when the image scaler is performed in the full color data level and in the bayer data level.

(58) Also, like the pre-scaler data sampler 251-3, the post-scaler address generator 252-2 selects one of the original image data and the pre-scaler image data according to the pre-scaler valid signal and outputs data at the position matched with the new post-scaler address. The image quality of the post-scaler output image is determined by the method of outputting the output data.

(59) FIG. 8A illustrates the output of the pre-scaler in the full color data level when the pre-scaling ratio is 1/4, and FIG. 8B illustrates the output of the pre-scaler in the bayer data level when the pre-scaling ratio is 1/4.

(60) Specifically, FIGS. 8A and 8B illustrate a case where the pre-scaling ratio is 1/4 with respect to a 1616 pixel array. Even though the pre-scaling ratios are equal to 1/4, the pre-scaler row/column address is (1,5,9,13) in the full color data level and (1,2,9,10) in the bayer level, whereby the pre-scaler addresses are differently implemented.

(61) Consequently, the data of the positions matched with the row/column addresses are outputted as the output data of the pre-scaler 251.

(62) FIG. 9A illustrates an output of the post-scaler in a full color data level when the post-scaling ratio is 5/7 with respect to a post 1414 pixel array, and FIG. 9B illustrates an output of the post-scaler in a bayer data level when the post-scaling ratio is 5/7 with respect to a post 1414 pixel array.

(63) Even though the post-scaling ratios are equal to 5/7, the row/column addresses is (1,2,3,5,6,8,9,10,12,13) in the full color data level and (1,2,3,4,5,6,9,10,11,12) in the bayer data level, whereby the post-scaler addresses are differently implemented.

(64) When the post-scaling ratio is greater than 1/2, it does not pass through the pre-scaler. When the scaling ratio is less than 1/2, it passes through the pre-scaler and is newly constructed like Eqs. (1) and (2). Therefore, the post-scaling ratio is always in a range from 1 to 1/2.

(65) For example, when the scaling ratio is 3/7, the pre-scaling ratio becomes 1/2 and the post-scaling ratio becomes 6/7 by Eq. (2). When the scaling ratio is 6/7, the post-scaling ratio becomes 6/7 by Eq. (1).

(66) Therefore, when the scaling ratio is 3/7 and 6/7, the post-scaling ratio becomes 6/7 in both cases.

(67) Like this, since the scaler method of the present invention can share the same post-scaling ratio, the post-scaler can be implemented in the same configuration, thereby reducing the complexity of hardware.

(68) The output data synchronizer 253 acts as a first input first out (FIFO) that synchronizes the image scaler output data with a random period, which is outputted from through the post-scaler, with the output clock of the image sensor and then outputs the data at a constant period. The output data synchronizer 253 is configured with a plurality of line memories. Also, the output data synchronizer 253 constantly matches the intervals of the respective lines, as well as the constant period of the output data of the image sensor.

(69) FIG. 10 is a flowchart diagram illustrating an algorithm of generating row/column addresses for the image scaling.

(70) Referring to FIG. 10, in step S101, it is determined whether an image scaler is executed or not. If the image scaling is unnecessary, the row/column addresses are unnecessary to generate. Therefore, the row/column addresses from the timing controller are outputted without change.

(71) In this case, the size of the output image from the pixel array is identical to that of the output image from the image sensor.

(72) On the contrary, in step S102, if the image scaling is necessary, it is determined whether the scaling is executed in a full color RGB data level or a bayer mosaic pattern data level. The scaling in the full color RGB data level is referred to as a full color scaling (FCS), and the scaling in the bayer mosaic pattern data level is referred to as a bayer mosaic scaling (BMS).

(73) Unlike the full color data, the bayer data has one color information per pixel in B/Gb or R/Cr line. Therefore, an address of the B/Gb or R/Gr pixel has to be generated at the same time. Thereafter, when the pre-scaling is executed according to the scaling ratio, the row/column addresses for the pre-scaling are generated and used in the pre-scaler. Also, when the post-scaling is executed, the row/column addresses for the post-scaling are generated and used in the post-scaler.

(74) When the FCS or BMS is selected in step S102, the row/column addresses for the pre-scaling and the post-scaling are generated through steps S103 to S109.

(75) In accordance with the present invention, arbitrary scaling factors (SCALEM, SCALEN) are received and the scaling ratio (=SCALEM/SCALEN) is determined through the scaling ratio generator. Then, the row/column addresses of the entire size of the image generated by the timing controller are reconfigured using the row/column addresses corresponding to the scaling ratio. Consequently, only data located at positions matched with the row/column addresses newly generated are outputted.

(76) The image size can be arbitrarily adjusted through the pre-scaler and the post-scaler. When the scaling ratio is greater than 1/2, the pre-scaler is off and the image size is adjusted only by the post-scaler.

(77) Since the size of the output image of the image sensor can be arbitrarily changed according to the scaling ratio, the image size can be adjusted according to the requirement of an image displaying system such as the digital camera or mobile phone.

(78) Typically, in case that a scaling ratio is small, a whole image size should be converted to a very small size. Thus, damage in an image is generated. However, in case that the scaling ratio is small, a pre-scaler serves a role in controlling an image size to a medium size of the scaling ratio to reduce the damage in the image.

(79) Therefore, it is possible to satisfy the technical requirements of the high image quality/high resolution of the capture image and the small-sized image display. Thus, it is easy to adjust the image size regardless of the size of the original image in the image sensor.

(80) Also, the hardware of the scaler can be simplified by separately installing the pre-scaler and the post-scaler depending on the scaling ratio, thereby increasing the degree of integration.

(81) Although the CMOS image sensor has been taken as an example, the present invention can also be applied to various kinds of image sensors.

(82) The present application contains subject matter related to the Korean patent application No. KR 2005-0015506, filed in the Korean Patent Office on Feb. 24, 2005, the entire contents of which being incorporated herein by reference.

(83) While the present invention has been described with respect to the particular embodiments, it will be apparent to those skilled in the art that various changes and modifications may be made without departing from the scope of the invention as defined in the following claims.