JP2011209523A - Conversion device, image display device, and data conversion method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technology on a conversion device converting image data.SOLUTION: The conversion device which outputs, as output data of a predetermined number bits, the result obtained by applying a predetermined arithmetic processing to input data includes: an LUT (Look Up Table) constituted of data of larger number of bits than the output bits, and having a significant data value outside the range expressed by the number of output bits; and an output means which outputs a result of applying a predetermined arithmetic processing to input data using the data value read from the LUT, by reducing the number of bits to the number of output bits.

Description

  The present invention relates to image data conversion.

  In image display devices that handle image data such as projectors, liquid crystal televisions, and plasma televisions, in order to convert input image data into a data format that can be processed by the image display device, a lookup table for data conversion (hereinafter, referred to as a data conversion lookup table). (Also referred to as LUT). When performing data conversion using such an LUT, the LUT does not perform data conversion corresponding to all input data values, but the LUT stores only data values for some input data (hereinafter, the LUT stores data). In general, the remaining input data is calculated by an interpolation operation based on the stored value of the LUT (Patent Document 1 below).

JP-A-7-99586

  By the way, the range of data values stored in the LUT is limited by how many bits are used to represent the data value. For example, an 8-bit unsigned integer can represent a value having a minimum value of 0 and a maximum value of 255 (hereinafter also referred to as an allowable data value). However, an ideal data value that is desired to be stored in the LUT may not be expressed by the number of bits used to encode the data value. As long as the interpolation calculation is performed within the allowable data value range, the expected result is obtained, but if the value outside the allowable data value range must be handled during the interpolation calculation, the data within the allowable data value range However, it has been pointed out that an accurate interpolation operation cannot be performed because it can only be handled. It is known that the problem caused by such a problem is manifested as output distortion in a conversion device that converts image data represented in a YUV color space into image data represented in an RGB color space. It has been. In the above-mentioned patent document, in order to avoid such a problem, the value is offset so that the data value falls within the range of values that can be stored in the LUT. Therefore, it is necessary to perform correction processing for removing the offset of the data value used for the interpolation calculation from the calculation result after the interpolation calculation.

  In order to solve at least a part of the problems described above, the present invention can take the following forms or application examples.

[Application Example 1]
A conversion device that outputs a result of performing predetermined arithmetic processing on input data as output data of a predetermined number of output bits, and is composed of data having a number of bits larger than the number of output bits, and The LUT having a significant data value outside the range that can be expressed by the number of output bits, and the result obtained by performing a predetermined arithmetic processing using the data value read from the LUT on the input data, the output bit number A conversion device comprising output means for cutting the output into two.

  According to this conversion apparatus, since the LUT has a significant data value outside the range that can be expressed by the number of output bits, the arithmetic processing can be performed using the significant data value. Compared to a case where the LUT can have a data value only within a range that can be expressed by the number of output bits, the significant data value can be used for the calculation, and an accurate calculation process can be performed.

[Application Example 2]
The conversion apparatus according to Application Example 1, wherein the predetermined calculation process includes an interpolation calculation, determines an interpolation section based on the input data, and is read from the LUT and corresponds to the data corresponding to the interpolation section A conversion apparatus including an interpolation calculation unit that performs the interpolation calculation on the input image using a value.

  According to this conversion apparatus, since the interpolation operation is performed using the LUT having a significant data value outside the range that can be expressed by the number of output bits, the data value can be held only within the range that can be expressed by the number of output bits. Compared with the case where the interpolation calculation is performed using the LUT, an accurate interpolation calculation can be performed.

[Application Example 3]
The conversion apparatus according to Application Example 2, wherein the LUT includes two LUT1 and LUT2, and the LUT1 is a first output data value corresponding to a lower limit value of an interpolation section when performing the interpolation calculation. The LUT 2 determines a second output data value corresponding to the upper limit value of the interpolation interval, and the interpolation calculation unit performs the interpolation calculation using the first and second output data values. Conversion device to perform.

  According to this conversion apparatus, since the upper limit value and the lower limit value of the interpolation section are determined by separate LUTs, the processing speed can be increased as compared with the case where the upper limit value and the lower limit value are determined by one LUT.

[Application Example 4]
The conversion device according to application example 3, wherein the LUT1 and the LUT2 are conversions that determine the first and second output data values based on a value represented by a predetermined upper number of bits of the input data. apparatus.

  According to this conversion apparatus, the upper limit value and the lower limit value are determined with reference to predetermined higher-order bits of the input bit data, so that complicated calculation processing is not required, and the interpolation interval is determined by relatively simple processing. Can be determined.

[Application Example 5]
The conversion device according to any one of Application Examples 2 to 4, wherein the interpolation calculation executed in the interpolation calculation unit is linear interpolation in the predetermined interpolation section, and the interpolation calculation unit is A conversion device that outputs the maximum value as a result of the interpolation calculation when a result of the interpolation calculation by linear interpolation exceeds a maximum value that can be expressed by the number of output bits.

  According to this conversion device, when the result of the interpolation operation exceeds the maximum value that can be expressed by the number of output bits, the maximum value is output, so the result of the interpolation operation that is larger than the range that can be expressed by the number of output bits and the output The error from the data value can be minimized.

[Application Example 6]
The conversion device according to any one of Application Examples 2 to 5, wherein the interpolation calculation executed in the interpolation calculation unit is linear interpolation in the predetermined interpolation section, and the interpolation calculation unit is A conversion device that outputs the minimum value as a result of the interpolation calculation when a result of the interpolation calculation by linear interpolation is less than a minimum value that can be expressed by the number of output bits.

According to this conversion device, when the result of the interpolation calculation is less than the minimum value that can be expressed by the number of output bits, the minimum value is output, so the result of the interpolation calculation that is smaller than the range that can be expressed by the number of output bits, The error with the output data value can be minimized.
[Application Example 7]
An image display device comprising: the conversion device according to any one of application examples 1 to 6; and a display unit that displays an image based on the output data from the conversion device.

  According to this image display apparatus, since the LUT provided in the conversion apparatus has a significant data value outside the range that can be expressed by the number of output bits, the arithmetic processing can be performed using the significant data value. Compared with an image display device including a conversion device that can only have a data value within a range that can be expressed by the number of output bits of the LUT, an accurate calculation process can be performed.

[Application Example 8]
A data conversion method for outputting a result of performing predetermined arithmetic processing on input data as output data having a predetermined number of output bits, the data conversion method including data having a number of bits larger than the number of output bits, and A significant data value is stored outside the range that can be expressed by the number of output bits, and a result obtained by performing a predetermined arithmetic processing on the input data using the data value is output by cutting it down to the number of output bits. Data conversion method.

  According to this data conversion method, significant data values outside the range that can be expressed by the number of output bits are stored, and a predetermined operation is performed using the stored data value. Since processing is performed, it is possible to perform data conversion by performing accurate arithmetic processing as compared with a case where arithmetic is performed by storing only data values within a range that can be expressed by the number of output bits.

  Note that the present invention can be realized in various modes. For example, the present invention can be realized in the form of a data conversion method and apparatus, a data processing system, an integrated circuit for realizing the functions of the method or apparatus, a computer program, a recording medium on which the computer program is recorded, and the like.

1 is an explanatory diagram showing a configuration of an image display system 10. FIG. It is explanatory drawing which shows roughly the image data conversion process which the image data conversion circuit 310 performs. It is a block diagram explaining the outline | summary of the image data conversion process which the image data conversion circuit 310 performs. It is explanatory drawing explaining LUT1,2. It is explanatory drawing explaining the comparative example for demonstrating an effect.

Next, embodiments of the present invention will be described based on examples.
A. First embodiment:
(A1) System configuration:
FIG. 1 is an explanatory diagram showing the configuration of an image display system 10 as a first embodiment of the present invention. The image display system 10 includes an image data supply device 20, a projector 30, and a screen SC. The projector 30 projects and displays an image represented by the image signal supplied from the image data supply device 20 on the screen SC. Further, image data (hereinafter also referred to as YUV data) expressed in a color space of Y (lightness) U (color difference) V (color difference) is input from the image data supply device 20 to the projector 30.

  The projector 30 includes an image data conversion circuit 310, a gamma correction circuit 320, a color unevenness correction circuit 330, a liquid crystal panel drive unit 340, a liquid crystal panel 350, a control unit 370, and a memory 380. The projector 30 also includes an illumination optical system 360 for illuminating the liquid crystal panel 350, and a projection optical system 365 that projects modulated light (hereinafter also referred to as image light) emitted from the liquid crystal panel 350 onto the screen SC. I have.

  The image data conversion circuit 310 represents image data (YUV data) input from the image data supply device 20 as gradation data (hereinafter also referred to as RGB data) expressed in RGB format of 8 bits for each color that can be processed by the projector 30. ). The image data conversion process performed by the image data conversion circuit 310 will be described in detail later. The gamma correction circuit 320 is a circuit that performs gamma correction on the RGB data input to the liquid crystal panel 350. Specifically, by correcting the RGB data so that the projected display image on the screen SC has a natural color expression, the light input / output characteristics (hereinafter referred to as VT characteristics) for each device of the liquid crystal panel 350 are corrected. Say).

  The color unevenness correction circuit 330 corrects the image data after the gamma correction by the gamma correction circuit 320 so as to control the color unevenness generated in the image displayed on the screen SC. “Color unevenness” refers to a variation in gradation in the liquid crystal panel in-plane direction due to uneven thickness of the liquid crystal layer of the liquid crystal panel 350 or variations in TFT operation characteristics within the liquid crystal panel surface. . Therefore, the color unevenness correction circuit 330 performs color unevenness correction for each predetermined region in the liquid crystal panel surface.

  The liquid crystal panel drive unit 340 forms an image on the liquid crystal panel 350 based on the image data after the color unevenness correction. The liquid crystal panel 350 is a light valve (light modulator) that modulates the illumination light emitted from the illumination optical system 360 under the control of the liquid crystal panel drive unit 340. Although not shown, the projector 30 has three liquid crystal panels 350 for three colors of RGB and is processed by the image data conversion circuit 310 to the color unevenness correction circuit 330 to drive the liquid crystal panel. Based on the RGB data input to the unit 340, the liquid crystal panel drive unit 340 drives the three liquid crystal panels 350 of RGB. The control unit 370 includes a CPU, and has a function of reading the driver program for the projector 30 stored in the memory 380 and controlling the operation of the projector 30.

(A2) Image data conversion process:
Next, image data conversion processing performed by the image data conversion circuit 310 will be described. FIG. 2 is an explanatory diagram schematically showing an image data conversion process performed by the image data conversion circuit 310. The image data conversion process is a process of converting 8-bit YUV data input from the image data supply device 20 into 8-bit RGB data that can be processed by the projector 30. The image data conversion circuit 310 includes a lookup table (hereinafter also referred to as LUT) that is referred to when converting (Y, U, V) to (R, G, B). The LUT does not store RGB data values corresponding to all input YUV data values (Y, U, V), but stores RGB data values corresponding to some YUV data values. The RGB data values corresponding to the remaining YUV data values are calculated by performing an interpolation operation with reference to the stored value of the LUT. Depending on the data value of the input YUV data, the data value of the RGB data after the image data conversion process may exceed the range of values that can be handled inside the projector 30. The details of the image data conversion processing by the image data conversion circuit 310 in this embodiment will be described below, including a method for dealing with such a case.

  FIG. 3A is a block diagram illustrating an outline of image data conversion processing performed by the image data conversion circuit 310. As shown in the figure, data input to the image data conversion circuit 310 is assumed to be input data Ain. Ain is data corresponding to YUV data that is image data input to the image data conversion circuit 310. The input data Ain is 8-bit data. The input data Ain input to the image data conversion circuit 310 is subjected to image data conversion processing by calculation using the LUTs 1 and 2 and the interpolation calculation circuit, and is output from the image data conversion circuit 310 as 8-bit output data Aout. The The output data Aout is data corresponding to RGB data that is converted by the image data 310 and output.

  First, the LUTs 1 and 2 included in the image data conversion circuit 310 will be described. FIG. 4 is an explanatory diagram for explaining the interpolation calculation. The horizontal axis in FIG. 4 corresponds to the input data Ain, and the vertical axis corresponds to the output data Aout.

  FIG. 4 shows an ideal interpolation function Fi (solid line) corresponding to the input data Ain and an interpolation straight line G (broken line) applied in the interpolation section of Ain = 192 to 255. As shown in the figure, the interpolation operation based on the interpolation function Fi is performed in the section where Ain = 0 to 192, whereas the interpolation function is performed until Aout reaches 255 in the section where Ain = 192 to 255. Interpolation is performed based on Fi, and in the section where the value of the interpolation function Fi exceeds 255, Aout is limited to 255 by the interpolation straight line G. FIGS. 3B and 3C are diagrams showing the structures of LUT1 and LUT2 used to determine a complementary section in which interpolation of output data Aout is performed based on input data Ain. LUT1 is configured to hold output data Q1 corresponding to the lower limit of the interpolation interval, and LUT2 is configured to hold output data Q2 corresponding to the upper limit of the interpolation interval. In this embodiment, it is assumed that data Q1 and Q2 are read by referring to LUT1 and LUT2 using the upper 2 bits including the MSB (Most Significant Bit) of input data Ain as an index.

  Here, an interpolation operation when Ain = 100 is input as input data Ain to the image data conversion circuit 310 will be described as an example. When the decimal number 100 is expressed in binary, it becomes “01100100”. That is, the index for referring to LUT1 and LUT2 is “01”. Therefore, when the value of the input data Ain is 100, data Q1 corresponding to Ain = 64 is read from LUT1, and data Q2 corresponding to Ain = 128 is read from LUT2.

  The LUT1 (FIG. 3) included in the image data conversion circuit 310 in this embodiment stores a 10-bit value as data Q1. Similarly, the LUT 2 included in the image data conversion circuit 310 stores a 10-bit value as data Q2. Of the 10-bit data stored in LUT1 and LUT2, the most significant 1 bit is used as a sign bit. Therefore, each lookup table can store values of “−512 to +511”. As an example, LUT1 in this embodiment can store Q1 = 240 corresponding to Ain = 192, and LUT2 can store Q2 = 270 corresponding to Ain = 255. With such a configuration, it is possible to use Fi, which is an ideal interpolation function, when performing interpolation calculation on the output data Aout corresponding to the input data Ain included in the interpolation section Ain = 192 to 255. Further, when the result of the interpolation calculation is calculated to exceed a data value that can be expressed by an unsigned 8-bit integer, that is, when a value that satisfies Aout> 255 is calculated, a “clipping process” to be described later. Thus, the value of Aout is set to 255.

  Next, an interpolation calculation method performed by the image data conversion circuit 310 using the LUTs 1 and 2 will be described with reference to FIG.

  When Ain = 100 is input to the image data conversion circuit 310, LUT1 and LUT2 are referred to using the upper 2 bits when the value of Ain is expressed in binary numbers as an index. That is, when Ain = 100, as described above, data corresponding to “01”, which is the upper 2 bits of Ain, is read from LUT1 and LUT2, respectively.

  On the other hand, among Ain = 100 (binary representation: 01100100) input to the image data conversion circuit 310 as input data, the lower 6 bits “100100 (decimal representation: 36)” are directly input to the interpolation arithmetic circuit as a coefficient α. Is entered as This coefficient α is used as a weighting coefficient for weighted average calculation.

  When data Q1, data Q2, and coefficient α are input to the interpolation calculation circuit included in the image data conversion circuit 310, a weighted average calculation is performed based on these data values and the following equation (1). Calculated. Note that “L” in the following equation (1) is the interval between the data values stored in the LUT, that is, the input data Ain associated with the stored values. That is, in this embodiment, the stored values are stored in the respective lookup tables at intervals of 64, Ain = 0, 64, 128. Therefore, in this embodiment, L = 64. In other words, “L” can also be said to be the width of the interpolation section in the image data conversion circuit 310.

ここで、αはＡｉｎの下位６ビットの値である。

Here, α is a value of lower 6 bits of Ain.

  When the interpolation calculation circuit performs the above interpolation calculation and calculates the weighted average value Y, clipping processing is performed on the weighted average value Y. The clipping process is a process of setting the output data Aout of the interpolation operation circuit to 255 when the weighted average value Y is larger than 255 that can be expressed by an unsigned 8-bit integer. In this embodiment, the clipping process for the maximum value has been described. For example, when the value of the weighted average value Y is smaller than 0, which is the minimum value that can be expressed by an unsigned 8-bit integer, that is, the weighted average value Y When A takes a negative value, Aout = 0. That is, when the weighted average value Y exceeds the allowable data value range (0 to 255) by clipping processing, the weighted average value Y is truncated and can be expressed as an unsigned 8-bit integer. And output as output data Aout. After that, the output data Aout from the image data conversion circuit 310 is subjected to other data processing by the gamma correction circuit 320 and the color unevenness correction circuit 330, and is applied to the screen SC via the liquid crystal panel driving unit 340, the liquid crystal panel 350, and the like. Displayed as a projected display image.

  As described above, the image data conversion circuit 310 converts the 8-bit input data Ain input from the image data supply device 20 into 8-bit output data Aout and outputs it, and interpolation performed in the process In the calculation processing, interpolation calculation is performed using LUTs 1 and 2 in which data values are expressed by 10 bits. Therefore, the interpolation calculation by the ideal interpolation function Fi can be performed in all the interpolation sections.

  On the other hand, for example, as shown in FIG. 5, let us consider a case where the lookup table included in the image data conversion circuit 310 is an LUT 3 that can store only 8-bit data. As described above, there are cases where it is desired to perform an interpolation operation using a value in a range that cannot be expressed by a predetermined number of bits. For example, the LUT can store only up to an 8-bit value, but the ideal interpolation function Fi has a value exceeding 255. As an example, the case where the value of the ideal interpolation function Fi is 270 with respect to the input data Ain = 255 is illustrated as in the example described with reference to FIG. In the LUT 3 that stores unsigned 8-bit integer data Q3, a value exceeding 255 cannot be stored. In such a case, an interpolation function Fc having a smaller gradient than the ideal interpolation function Fi is normally applied to the interpolation section Ain = 192 to 255 in FIG. That is, when Ain = 255, the interpolation calculation is originally performed with Q3 = 270, but the interpolation calculation is performed with Q3 = 255, resulting in a problem that the value of the output data Aout is distorted. On the other hand, in the projector 30 according to the present embodiment, the interpolation value can be accurately calculated for data whose output value (RGB data) after the original data conversion does not exceed the range that can be expressed by the projector 30 (see FIG. 4). ). Note that the value of the region to be clipped has little influence on the image finally projected from the projector 30, so that even if the output data Aout is saturated to a predetermined value (255 in this embodiment), it is practical. No problem.

  Further, the data value exceeding the range expressed by the projector 30 is stored in the LUT by offsetting the value of Aout, and read out as a value offset with respect to the input Ain, and then the correction calculation for the offset. Compared with the case of performing the above (for example, Patent Document 1 above), in this embodiment, it is not necessary to perform the correction calculation for the offset. As for the correspondence relationship with the claims, the image data conversion circuit 310 corresponds to the conversion device described in the claims, and the clipping processing corresponds to the output means described in the claims.

B. Variations:
The present invention is not limited to the above-described examples and embodiments, and can be implemented in various modes without departing from the gist thereof. For example, the following modifications are possible.

(B1) Modification 1:
In the above embodiment, the input data Ain input to the projector 30 is YUV data, and the output data Aout after data conversion in the image data conversion circuit 310 is RGB data. However, the present invention is not limited to this. The data Aout can be a data format expressed in a color space such as HLS, YCbCr, YPbPr, or a data format expressed in a color space standardized by xvYCC (Extended-gamut YCC color space for video applications). It is. Even if Ain and Aout are in such a data format, the same effect as in the above embodiment can be obtained.

(B2) Modification 2:
In the above embodiment, Ain and Aout are unsigned 8-bit integer data, and the LUTs 1 and 2 included in the image data conversion circuit 310 are 10-bit data in the range of data values used for the interpolation calculation. In the range that can be processed by the projector 30, Ain and Aout can be N-bit data (N is an integer of 2 or more), and the output data of LUT1 and LUT2 can be M-bit data (M is an integer of N or more). . Even if it does in this way, the effect similar to the said Example can be acquired.

(B3) Modification 3:
In the above embodiment, the image data conversion circuit 310 is provided with two LUTs LUT1 and LUT2 in order to determine the upper limit value and the lower limit value of the interpolation section. However, the present invention is not limited to this, and only one LUT is provided. You may not. In this case, two addresses, an address corresponding to the upper limit value of the interpolation section and an address corresponding to the lower limit value, are designated for a predetermined number of bits of Ain (the upper 2 bits in this embodiment). The upper limit value and the lower limit value of the interpolation interval may be determined in order from one LUT. Even if it does in this way, the effect similar to the said Example can be acquired.

(B4) Modification 4:
In the above embodiment, the LUT in the image data conversion circuit and the image data conversion process using the LUT are applied to the projector. However, the present invention is not limited to this, and image data is input and an image is displayed based on the input image data. It can be applied to image display devices such as liquid crystal televisions and plasma televisions. Also, the present invention can be applied to an image processing apparatus that is externally connected to the image display apparatus, inputs image data from the image data supply apparatus, converts the image data into data that can be processed by the image display apparatus, and outputs the data to the image display apparatus. it can. Even if it does in this way, the effect similar to the said Example can be acquired.

(B5) Modification 5:
In the above embodiment, a part of the functions realized by software may be realized by hardware, or a part of the functions realized by hardware may be realized by software. Even if it does in this way, the effect similar to the said Example can be acquired.

(B6) Modification 6:
In the above embodiment, the case where the data value (R, G, B) after data conversion exceeds the upper limit value of the allowable data value in the data conversion from YUV data to RGB data in the image display device has been described. Without limiting to this, as described above, the LUT in the first embodiment can store up to the range of “−512 to +511” as Aout. Even when G, B) exceeds the lower limit value of the allowable data value to the minus side, the output value (RGB data) after the original data conversion does not exceed the range that can be expressed by the projector 30 correctly. Interpolation can be performed.

DESCRIPTION OF SYMBOLS 10 ... Image display system 20 ... Image data supply apparatus 30 ... Projector 310 ... Image data conversion circuit 320 ... Gamma correction circuit 330 ... Color unevenness correction circuit 340 ... Liquid crystal panel drive part 350 ... Liquid crystal panel 360 ... Illumination optical system 365 ... Projection optics System 370 ... Control unit 380 ... Memory Ain ... Input data Aout ... Output data SC ... Screen

Claims (8)

A conversion device that outputs a result of performing predetermined arithmetic processing on input data as output data of a predetermined number of output bits,
An LUT composed of data having a larger number of bits than the number of output bits and having a significant data value outside the range that can be expressed by the number of output bits;
A conversion device comprising: an output unit that outputs a result obtained by performing predetermined arithmetic processing on the input data by using the data value read from the LUT, and truncating the result to the number of output bits. The conversion device according to claim 1,
The predetermined calculation process includes an interpolation calculation,
A conversion apparatus comprising: an interpolation calculation unit that determines an interpolation interval based on the input data and performs the interpolation calculation on the input image using the data value corresponding to the interpolation interval read from the LUT. The conversion device according to claim 2,
The LUT consists of two, LUT1 and LUT2.
The LUT 1 determines a first output data value corresponding to a lower limit value of an interpolation section when performing the interpolation calculation,
The LUT2 determines a second output data value corresponding to the upper limit value of the interpolation interval,
The interpolation calculation unit performs the interpolation calculation using the first and second output data values. The conversion device according to claim 3,
The LUT1 and the LUT2 determine the first and second output data values based on a value represented by a predetermined upper number of bits of the input data. A conversion device according to any one of claims 2 to 4,
The interpolation calculation executed in the interpolation calculation unit is linear interpolation in the predetermined interpolation section,
The interpolation device outputs the maximum value as a result of the interpolation calculation when a result of the interpolation calculation by the linear interpolation exceeds a maximum value that can be expressed by the number of output bits. A conversion device according to any one of claims 2 to 5,
The interpolation calculation executed in the interpolation calculation unit is linear interpolation in the predetermined interpolation section,
The said interpolation calculating part outputs the said minimum value as a result of the said interpolation calculation, when the result of the said interpolation calculation by the said linear interpolation is less than the minimum value which can be expressed with the said output bit number. The conversion device according to any one of claims 1 to 6,
An image display device comprising: a display unit that displays an image based on the output data from the conversion device. A data conversion method for outputting a result of performing predetermined arithmetic processing on input data as output data having a predetermined number of output bits,
It is composed of data having a larger number of bits than the number of output bits, and stores a significant data value outside the range that can be expressed by the number of output bits,
A data conversion method for outputting a result of subjecting the input data to a predetermined calculation process using the data value, truncated to the number of output bits.

Images