JPS63181568A - Gradation conversion method - Google Patents

Abstract

PURPOSE:To speed up the selection by sequentially accumulating the product of a brightness and a frequency in a density histogram to obtain a accumulated value and selecting one of plural gradation conversion curves according to the number of comparisons executed until then when it exceeds a specified value. CONSTITUTION:A CPU 2 prepares plural gradation transforming curves, forms the density histogram in a RAM 3 on a picture to be gradation-converted sequentially accumulates the product of the brightness and the frequency in this density histogram to obtain the accumulated value. Then, the accumulated value is compared with the prescribed specified value for every time of the accumulation, and when the accumulated value exceeds the prescribed specified value, one is selected among the plural gradation conversion curves according to the number of the comparison executed until then. Thereby, in case one of the plural gradation conversion curves stored in a ROM is selected based on the formed density histogram, the selection can be executed at high speed.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention uses a histogram method to select an optimal tone conversion curve for an image whose tone is to be converted in a tone conversion method. This invention relates to a gradation conversion method.

[Prior Art] In a conventional tone conversion method, a plurality of tone conversion curves are written in advance in a ROM, and when tone correction is performed based on the density histogram of an input image, first, the accumulation of cumulative histograms from the bright direction is calculated. Find the point where the value or the cumulative value of the cumulative histogram from the dark direction is greater than the specified value, then perform operations such as subtraction, division, and comparison at the found point, and based on the result of this operation. , one of the plurality of gradation conversion curves is selected. As mentioned above, if several tone conversion curves are stored in the ROM in advance, there is no need to create tone conversion curves each time.
The time will be shortened accordingly.

However, in the above conventional method, it is necessary to find a point where the cumulative value reaches a predetermined value based on the created density histogram, and then perform complicated calculations such as subtraction, division, and comparison on the point. Therefore, there is a problem that the processing speed becomes slow.

[Object of the Invention] The present invention has been made by focusing on the problems in the conventional method described above, and based on the created density histogram, R
It is an object of the present invention to provide a gradation conversion method that can perform the selection at high speed when selecting one from a plurality of gradation conversion curves stored in an OM.

[Embodiments of the Invention] FIG. 1 (1) is a diagram showing a density histogram. In this figure, the horizontal axis indicates the brightness of each dot, and the vertical axis indicates the number (frequency) of dots having that brightness. Therefore, in the above diagram, the closer the value is to 0, the darker the value. In addition, the brightness (brightness) of the human image
is input as an 8-bit digital value, this input image is sampled, and a density histogram is created based on the sampled values.

FIG. 1 (2) is a diagram showing the principle of the present invention, and is a diagram showing a cumulative density histogram from the dark direction in the density histogram, and FIG. 1 (3) is a diagram showing the principle of the present invention. FIG. 3 is a diagram showing a cumulative density histogram from the bright direction in the density histogram.

The number of dark tone switching points and the number of bright tone switching points are determined according to the number of tone conversion curves stored in a ROM (not shown). In other words, the greater the number of tone conversion curves stored in the ROM, the greater the number of tone switching points. Here, the number of dark gradation switching points and the number of bright gradation switching points are
Each is set to eight.

Then, each gradation switching point is associated with the characteristics of the gradation conversion curve. In other words, the dark area gradation switching point ■ corresponds to a gradation conversion curve of a certain characteristic, and the dark area gradation switching point ■ corresponds to a gradation conversion curve of another characteristic,... The correspondence is as follows.

When selecting a gradation conversion curve using the cumulative density histogram from the dark direction, first find the cumulative value from 0 to the dark gradation switching point If the cumulative value is smaller than the specified dark area value, the cumulative value from O to the next gradation switching point (■ in the above case) is calculated. Then, compare the cumulative value with the dark area specified value, and if the cumulative value is smaller than the dark area specified value, move from O to the next gradation switching point (■ in the above case).
Calculate the cumulative value up to. In this way, if the cumulative value is smaller than the specified dark area value, the operation of calculating the cumulative value from 0 to the next gradation switching point is repeated in sequence.

On the other hand, when the 'JA11i value exceeds the specified value, a tone conversion curve is selected based on the number of times it has been compared with the dark area specified value. In other words, the gradation conversion curve corresponding to the last compared dark area gradation switching point is the curve to be found. In this way, one of the plurality of tone conversion curves is selected.

1] When selecting a tone conversion curve using the cumulative density histogram from the direction, the tone conversion curve is selected in the same manner as described in h. In other words, first find the cumulative value from 255 to the bright area gradation switching point ■, and then set this cumulative value 1 to the bright area specified value (
This value is set in advance), and if the above cumulative value is smaller than the specified bright area value, calculate the cumulative value from 255 to the next gradation switching point ((2) in the above case). . Then, compare the cumulative value with the bright area specified value, and if the cumulative value is smaller than the bright area specified value, calculate the cumulative value from 255 to the next gradation switching point (■ in the above case). . In this way, if the cumulative value is smaller than the bright area specified value? 5
The operation of calculating the cumulative value from 5 to the next gradation switching point is repeated.

On the other hand, if the cumulative value exceeds the specified value, a gradation conversion curve is selected based on the number of times it has been compared with the bright area specified value. In other words, the gradation conversion curve corresponding to the last compared bright area gradation switching point is the curve to be found. In this way, one of the plurality of tone conversion curves is selected.

FIG. 2 is a block diagram showing the circuit configuration of a printer that is an embodiment of the present invention.

This embodiment has one image input section l, CPU2, and RAM3.
The 0 gradation conversion table 4, which includes a gradation conversion table 4, an image processing circuit 5, and an output section 6, expands and compresses gradations within a predetermined gradation range, and The processing circuit 5 performs image processing such as optimal color conversion.

The CPU 2 prepares a plurality of gradation conversion curves, creates a density histogram in the RAM 3 for the image whose gradation is to be converted, and sequentially accumulates the product of brightness and frequency in this density histogram to obtain a cumulative value. Each time this accumulation is performed, the cumulative value is compared with a predetermined specified value, and when the cumulative value exceeds the specified value, the plurality of floors are 1 from the key conversion curves
This is an example of means for selecting one.

Next, the operation of selecting a gradation conversion curve in the gradation conversion table in the embodiment described in "-" will be described.

FIGS. 3(1) and 3(2) are flowcharts showing operations in the above embodiment.

First, a histogram for density levels O to 255 is created. First, a tone selection routine for dark areas is performed. Set the cumulative histogram value (SUMH3T), reference histogram table (HST), and comparison count (CMPC) to O (Sl), and set the cumulative histogram value (SUMH3T) to O (Sl).
The contents of the reference hist table (HST) are added to ST) (S2).

The optimal conversion point (D
r (CMPC)) (this point exists for (number of switching stages - 1)) and the reference hist table (HST) (33). If they do not match, the reference hist table (HST) ) to the next point (S4) and perform the hill marking motion.

If they match (S3), compare the dark area specified value (DPI) and cumulative histogram value (SUMH3T) (S5), and if this is less than the dark area specified value (DPI), compare the number of comparisons (CM
After adding 1 to (PC) (S6), the number of comparisons (CMPC
) and the upper limit of the number of comparisons (DPMAXCMP),! If f comparison L (S7) is less than the upper limit of the number of comparisons, the reference hist table (STH) is moved to the next point (S4), and the above operation is repeated.

Also, if the cumulative histogram value (SUMHS T) is equal to or higher than the dark area specified value (DPI) or
) becomes greater than or equal to DPMAXCMP, the value of the number of comparisons (CMPC) at that time becomes the selected value on the dark side of the optimum gradation conversion curve for that image (s i o). The gradation conversion curve corresponding to this selected value becomes the optimum one.

In order to match the number of comparisons (CMPC) with the selected value, the initial value of the number of comparisons (CMPC) is matched to the characteristics of the gradation conversion curve.

By doing the above, subtraction in the conventional example,
Since division is not necessary, the search procedure is simplified and the search processing time is shortened.

The bright side is also processed in the same way as above. That is,
Set the cumulative histogram value (SUMH3T) to O, set the reference histogram table (HST) to 255, and set the number of comparisons (CM
PC) to 7 (Sll). Then, the reference histogram (HS T
) are added (S l 2).

The optimal conversion point (H
P (CMPC)) (this point exists for (number of switching stages - 1)) and the reference hist table (HST). HST) to the next point514)
, perform the above operations.

If they match (Sl3), compare the bright area specified value (HPI) and the cumulative histogram value (SUMHS T).S1
5) If this is less than the specified bright area value (HPI), after adding 1 to the number of comparisons (CMPC) (516), calculate the number of comparisons (CMPC) and the northern limit of the number of comparisons (HPMAXC).
If the comparison count is less than the upper limit value of the number of comparisons, the reference hist table (STH) is moved to the next point 514), and the above operation is returned in green.

In addition, if the cumulative histogram value (SUMHS T) is greater than or equal to the bright area specified value (HPI) or
) becomes HPMAXCMP or more, the value of the number of comparisons (CMPC) at that time becomes the selected value on the bright side of the optimal gradation conversion curve for that image (S20), and the gradation conversion curve corresponding to this selected value becomes optimal.

In the above embodiment, when selecting a gradation conversion curve,
A cumulative density histogram from the dark direction starting from luminance 0 and a cumulative density histogram from the bright direction starting from luminance 255 are used, and two starting points are used, but the number of starting points is two. but not limited to.

[Effects of the Invention] According to the present invention, when one of the plurality of gradation conversion curves stored in the ROM is selected based on the created density histogram, the selection can be executed at high speed. It has the effect of being able to.

[Brief explanation of the drawing]

FIG. 1(1) is a diagram showing a density histogram. FIG. 1(2) is a diagram showing the principle of the present invention, and is a diagram showing a cumulative density histogram from the dark direction in the density histogram. FIG. 1(3) is a diagram showing the principle of the present invention, and is a diagram showing a cumulative density histogram from the bright direction in the density histogram. FIG. 2 is a circuit configuration block diagram of a printer that is an embodiment of the present invention. FIGS. 3(1) and 3(2) are flowcharts showing one embodiment of the present invention. 2... CPU, 4... Gradation conversion table, 5... Image processing circuit. Patent Applicant: Canon Co., Ltd. Same Agent: Arata Yokubo - o -4/l 255 Saki #in III Kiritan~, BJJtP Floor 1 Dekiri Kyoushin Figure 2 Figure 3 (1)

Claims (2)

[Claims] (1) Prepare multiple gradation conversion curves, create a density histogram for the image whose gradation is to be converted, sequentially accumulate the product of brightness and frequency in this density histogram to find the cumulative value, and calculate the cumulative value. At each time, the cumulative value is compared with a predetermined specified value, and when the cumulative value exceeds the specified value, depending on the number of times the comparison has been performed so far,
A gradation conversion method comprising selecting one of the plurality of gradation conversion curves. (2) In claim 1, the gradation conversion method is characterized in that the number of times the product of luminance and frequency in the density histogram is accumulated is determined according to the number of the prepared gradation conversion curves. .