JPS6222182A - Picture information processor - Google Patents

Abstract

PURPOSE:To eliminate the difference of the brightness of a different type source picture and to obtain a reproducing picture easy to see by converting a picture data so as to be distributed by setting a prescribed brightness as a reference according to the brightness of the output of a discriminating means. CONSTITUTION:A picture input means 1 selects and inputs one of the different type source picture data VCHa-VCHd by an indicating input from a keyboard 7, and stores it on a picture memory a2. A discriminating means 9 reads a picture data L at a memory a2 and outputs a brightness discriminating signal RS. A luminance conversion ROM3 performs a luminance conversion on the picture data L according to a conversion function that is regulated so that the brightness of the picture data is distributed and converted in full scale setting a prescribed brightness as the reference, and outputs a picture data M. A picture memory b4 synthesizes and displays plural pictures on which the luminance conversion is applied.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image information processing device, and particularly to a method for controlling the brightness of a reproduced image caused by differences in the characteristics of an image data generating device and various conditions during image data generation. The present invention relates to an image information processing device that improves the problem of differences.

[Prior Art] In recent years, various image editing devices have been developed, and it has become possible to freely combine and enjoy various video images as raw materials. Input source images for this type of device include television images, laser disc images, still videos, graphics images from microcomputers, etc.; for example, as shown in Fig. 5(b).
When these disparate source images are synthesized and reproduced as shown in Figure 2, generally the bright source images are reproduced as bright, and the dark source images are reproduced because the characteristics of the image data generating device and the imaging conditions at the time of image data generation are generally different. Therefore, for example, when a composite multi-image is output using a device such as a video hard copy or an inkjet video printer, the brightness of one screen varies depending on the location, making it very unsightly. was there. Moreover, when the image data generation device generates a 6-bit gradation image and an 8-bit gradation image with different levels, there is a problem as to where to set the brightness standard when reproducing one image.

[Problems to be Solved by the Invention] The present invention aims to eliminate the difference in brightness between different source images with a simple configuration and provide a reproduced image that is easy to see.

[Means for Solving the Problem] As a means for solving this problem, for example, the image information processing apparatus of the embodiment shown in FIG.
The image input means 1 selects and inputs one of CHa, NVCHd, and the image data Va input to the image input means 1 (or the image data L output from the image memory a) K have uniform brightness. The image data V on the changing coordinates
A determining means 9 that determines the brightness of an image based on the brightness of image data when the cumulative frequency of a or L is 2% and q% of the total, and the brightness of the output of the determining means 9. In order to obtain image data M whose brightness is converted so that the brightness of the image data L is distributed based on a predetermined brightness according to the discrimination output R5 of A brightness conversion read-only memory (RO) containing a conversion function that specifies that the brightness of the image data L is distributed and converted to full scale based on a predetermined brightness is stored at the address selected by the data L.
M) Equipped with 3.

[Operation] In the configuration shown in FIG.
One of Ha-VCHd is selected and input and stored in the image memory a2. Next, the determining means 9 scans and reads the image data L in the image memory a2 for one screen,
Take the cumulative frequency of image data L on coordinates where the brightness changes uniformly, and calculate the brightness of the image data when the cumulative frequency is 2% of the total and the brightness of the image data when it is q% of the total. The brightness distribution state of the image is determined based on the brightness, and a brightness determination signal R3 is output.

Next, the brightness conversion ROM 3 is addressed by the brightness discrimination signal R3 output from the discrimination means 9 and the image data L, so that the brightness of the image data L is distributed to be converted to full scale based on a predetermined brightness. The image data L is luminance-converted according to a specified conversion function, and the converted image data M
The 0-image memory b4 outputs a plurality of images whose brightness has been converted in this way, and further reads and outputs the multi-image data to the display means 6 via the D/A converter 5.

[Examples] Examples of the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a block diagram of an image information processing apparatus according to an embodiment of the present invention. In FIG. Image input means; 2 is an image memory a for temporarily storing image data Va taken in by the image input means 1; 3 is a brightness discrimination signal R3 output from the discrimination means 9 and image data read from the image memory a2; The brightness of the image data L is converted according to a predefined conversion function such that the brightness of the image data L is distributed to a full scale based on a predetermined brightness, and the converted image data is A brightness conversion read-only memory (ROM) 4 for reading and outputting M is used for the purpose of storing so-called multi-images, and an image memory b for compressing the images in the image memory a and sequentially combining and storing them in a part. Image data v read from image memory b4
6 is a display means for displaying multiple images as shown in FIG. 5(b), and 7 is designation information for specifying an image processing method and image processing A keyboard 8 for inputting necessary parameters is a central processing unit (8) comprising a ROM for storing the image processing program of the embodiment shown in FIG. 4 and a RAM for storing data and flags necessary for image processing.
CPU).

Inside the CPUa, various functional blocks realized by executing the program shown in FIG. 4 are shown.

9 takes the cumulative frequency of the image data L on the coordinates where the brightness uniformly darkens, and the cumulative frequency is p% and q% of the whole.
10 is an address conversion unit that thins out and reads out the contents of the image memory &2 and writes a substantially compressed image as a composite to the image memory b4. Means 11 is a control means for selecting a video channel (VCHa - VCHd) for inputting an image according to a specified input from the keyboard 7, and for controlling the activation of the determining means 9 and the address converting means 1O.

The image input means 1 has a plurality of video channels VCHa~
For example, VCHa is a TV image, VCHb is an image from a laser disk device, VC
Hc has different sources of image data, such as images from a graphics computer. Also, VCHd
The image input means 1 includes an A/D converter because the image signal formats may be different, such as VCHa being a digital image and VCHa being an analog luminance image signal.

The determining means 9 reads out the contents of the image memory a2 and detects the brightness of one image.For example, when 8-bit gradation image data can be reproduced, it performs any correction on the 6-bit gradation input image data. If you play back the 6-bit gradation without changing the image, the 6-bit gradation image will be relatively dark.Also, even if the image data is the same 8-bit gradation, the image data may have brightness and darkness due to differences in the optical conditions at the time of imaging. arise. In order to correct this inconvenience, the brightness of the input image data is first evaluated. As one method, in this embodiment device, the cumulative frequency of the brightness of the input image data is calculated, and the threshold value is calculated as a percentage. We decided to determine the brightness of the image. This relationship is shown in the histogram of FIG. That is, for example, take the frequency distribution HGb of the input image data VCHb, accumulate it from the brightest side, and when the cumulative number exceeds 1% of the total for the first time, the luminance (L) at that time is used to calculate the frequency distribution HGb of the image VCHb. Let it be a highlight point HPb. Furthermore, the cumulative number is 99% of the total.
When the brightness (L) at that time exceeds the image V
Let it be the dark point (shadow point) SPb of CHb. By determining these two points, the spread of the luminance distribution of the image VCHb can be determined, and it can be determined whether the image is a dark image or a bright image. In addition, input image data VCH
The same applies to the frequency distribution HGa of a.

FIGS. 3(a) and 3(b) are diagrams showing the conversion principle of the brightness conversion ROM 3. In general, image data representing luminance (L) can be applied to a display device such as a CRT, but when used in a printing device, it is usually logarithmically converted to density data representing density CD). Since a multi-image is displayed on the display means 6, there is no need to convert the input image data L to density (D), but an example of density conversion is shown taking into account the explanation when outputting to a color printer etc. If you want to obtain the luminance conversion output data M as in this embodiment, you can return to the luminance again or convert the luminance from the beginning. An example of the density conversion characteristic R41 is shown. When printing multiple images, in order to equalize the brightness of multiple images, for example, the width of the frequency distribution HGa (Habo SP
This is because it is sufficient to convert to full-scale density (D) using the slope characteristic R41 corresponding to HP1 to HP4). On the other hand, the third
Figure (b) shows a density conversion characteristic R1a suitable for the frequency distribution HGb of the image VCHb. In this case, the image VC
Since the brightness of Hb is distributed from the maximum to the minimum, the conversion width of the conversion characteristic R14 (from SP4 to HP) is also close to full scale. Therefore, for example, the image data Ln of the frequency distribution HGa is subjected to density conversion using the characteristic R41. Then, the density is converted to the density Dp, and if the image data Lm of the frequency distribution HGb is density-converted using the characteristic R14, the same density Dp can be obtained. This relationship is
Image VCHa is 6-bit gradation data, and image VCH
The same applies when b is 8-bit gradation data, so even if both images are printed on the same screen, it will not be difficult to see.

The brightness-brightness conversion characteristic of the embodiment is calculated in advance using the following equation and stored in the brightness conversion ROM 3. That is, ROM
Set the address input of 3 to N.

When the data read output is M, M = [255XlO] However, [X] is the largest integer that does not exceed )Z=DXL
o g (L/HPf) ÷ L o g (S
Pj/HPi)SPi<L<HPi, where D is preferably 1.5 to 2.0.

In the example, HP I = 250, M P 2 =
210.

HP 3 = 170, HP a = 130, and 5r1 = 22, SP, = 18.5P3 = 14.5
A total of 16 combinations (HPi, 5rj
), create a conversion characteristic R1j suitable for this, and store the data in the brightness conversion ROM 3 in advance. Therefore, the discriminating means 9 uses HPi which is closest to HPb obtained according to FIG.
Outputs the OM selection signal R3 (HPi, 5Pj) and selects the conversion characteristic (RiD. This conversion output M is always 255
Enter X l O-0~255. Note that in the above example, the gradation curve is uniquely determined from HP, but for example, HP and SP are calculated separately as values of 1% and 99% of the cumulative frequency function,
A total of 16 lI curves may be prepared, including 4 ways for HPi and 4 ways for SPj.

The address converting means 10 thins out and reads out the contents of the image memory a2 in order to form a multi-image in the image memory b4. For example, when considering a four-part multi-image as shown in FIG. 5(b), the processing is simple. That is, by reading out one logical column address X-a and one logical row address Yb assumed to be in the image memory a2, each side of one image becomes 172.
Therefore, the area ratio becomes 18. In the case of the image Va in the same figure (SL), this is the address (XO, YO) in the same figure (b).
), the compressed image shown in the figure will be transferred. The same applies to the other images vb to vd, just by changing the base of the write address.

FIG. 4 is a flowchart showing the image processing procedure of the embodiment. In step S1, it is determined whether the designation from the keyboard 7 is a video channel selection designation V CHa to VCHd. The channel selection designation also serves as an image reading designation. Therefore, if the determination is YES, step S2
Then, the selection signal vS is sent to the image input means 1 to activate image reading. In step S3, the input image Va is stored in the image memory a2. In step S4, 6 determinations to determine whether the automatic brightness correction flag AUTOF is 1 is YES.
Then, in step S5, an energizing signal STR is sent to the determining means 9, and the determining means 9, which energizes the processing, determines HPi and SPj and selects a ROM characteristic suitable for brightness correction. In step S6, an energizing signal STA is sent to the address conversion means 1O to energize the processing. The address conversion means 10 is an image memory a
At this time, the read data of the image memory a2 is subjected to luminance conversion via the luminance conversion ROM 3. Further, if AUTOF is not 1 in step S4, step S5 is bypassed. That is, the brightness conversion is performed using the ROM characteristics selected previously. This is convenient when you want to display the brightness according to the operator's specifications in advance, rather than normalizing the brightness of the input image.Furthermore, in step S1, if the key manual selection is not VCH selection, proceed to step S7 and select AUTO. Determine whether it is a specified input or not. If AtJTO is specified, AUTOF is set to 1 in step S8, and the process proceeds to step s5. As a result, the image already in the image memory a2 is reprocessed in the AUTO mode. If the AUTO mode is not specified in step S7, the process advances to step S9 and AUTOFt-0 is set. In step SIO, the specification information R5 for forced ROM selection from the keyboard 7 is sent to the determination means 9, and the information R5
is set as the selection signal R5 of the brightness conversion ROM3, and step S6
Therefore, in this case the key manually forced RO
Since the brightness is converted using M, the brightness of the multi-image can be changed according to the operator's preference.

FIG. 6 is a block diagram showing the application of the embodiment to color signals. In FIG.
The input signal is input converted and input to frame memories 20R, 20G, and 20B provided for each color in one image memory 2. The output signals LR, LG, and LB of the respective frame memories 20R, 20G, and 20B are input to the gamma correction ROMs 24R, 24G, and 24B in the luminance conversion ROM 3', respectively, to perform conversion and gamma correction of R, G-+M, and B4Y. , and brightness normalization processing is performed.

As mentioned above, the gamma correction circuit 24
In this embodiment, the gamma correction circuit 24 performs logarithmic conversion, gamma correction, and automatic normalization or forced conversion of brightness.
is composed of a ROM, and a total of 16 types of gamma correction reference tables are arranged in the ROM. In other words, it has four points SPl to SP4 as dark points for use, and four points HP to HP4 as highlight points for processing.
It has 16 types of tables as combinations thereof. Therefore, the gamma correction circuit is
8' can be freely switched by a turf point/chat point designation signal R5 output from a determining means (not shown). In this case, the discrimination means should normally sample each luminance signal LR, LG, and LB to judge the brightness, but considering that generally about 60% of the luminance signal is formed from the green signal. Therefore, it is sufficient to sample only the green signal LG as shown in FIG.

Furthermore, the data MC, MM, MY whose characteristics have been converted by the gamma correction ROMs 24R, 24G, and 24B are further input to the masking circuit 26, where unnecessary color removal processing is performed, and each color signal MC', MM', MY', BK is ' are obtained and converted into analog signals by digital-to-analog converters (hereinafter referred to as D/A converters) 28C, 28M, 28Y, and 288, respectively. Each analog signal C, M, Y, BK is
For example, the light is applied to the inkjet head via the inkjet head's driver circuit to obtain an image with desired brightness.

[Effects of the Invention] As described above, according to the present invention, all image p data can be converted by converting the brightness according to the input image, regardless of the brightness or number of gradations. The standard is between 255X1O and 255X1O, and if this video output is hard-copied, an image with uniform brightness can be obtained over the entire screen.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an image information processing apparatus according to an embodiment of the present invention. FIG. 3 is a diagram showing a frequency curve of brightness of the image data of FIG. 2; FIGS. 3(a) and (b) are diagrams showing the conversion principle of the brightness conversion ROM 3, FIG. 4 is a flowchart showing the image processing procedure of the embodiment, and FIGS. 5(L) and (b) are one mode of display. Diagram showing. FIG. 6 is a block diagram showing the application of the embodiment to a color signal, and FIG. 7 is a diagram showing a frequency curve of brightness of green data. In the figure, l: image input means, 2: image memory a,
3... Brightness conversion conversion ROM, 4... Image memory b, 5
...D/A converter, 6.Display means, 7.Keyboard, 8.Central processing unit (
CPU), 9...Discrimination means, 1O...Address translation means, 11...Control means. 2nd cause 1 demon (L) □ 3rd rIA (0) - ROM7F'LS (L) □ 3rd ward (b) ROM7F'LS (L) □

Claims (3)

[Claims] (1) A determining means for determining the brightness of an image based on the frequency distribution of the image data, and converting the image data so that it is distributed based on a predetermined brightness according to the brightness output from the determining means. An image information processing device characterized by comprising a conversion means for converting. (2) The determining means calculates the cumulative frequency of the image data on coordinates where the brightness uniformly changes, and determines whether the image has the brightness of the image data when the cumulative frequency is p% and q% of the total. 2. The image information processing device according to claim 1, wherein the image information processing device determines the brightness of the image. (3) The conversion means is configured to apply a conversion function to an address selected by the brightness of the output of the discriminator and selected by the image data, the conversion function being defined to transform the distribution of the image data to full scale based on a predetermined brightness. 3. An image information processing apparatus according to claim 1, characterized in that the image information processing apparatus comprises a read-only memory including: