JPH07182516A - Judgement method of color component of computer graphic picture - Google Patents

Abstract

PURPOSE: To provide a color decision method which automatically sorts an image into a specific category based on a pixel color component that exists in a computer graphics image. CONSTITUTION: An image is automatically sorted by deciding the color component of a computer graphics image. The primary image types which are automatically sorted include (1) a binary image, (2) a gray scale image and (3) a color image, and these automatically sorted images are stored in an assigned memory 120 as the graphics images according to the image conditions. In this case, the gray scale and binary images can be stored by means of 8-bit/pixel and 1-bit/pixel respectively after automatic sorting with higher precision.

Description

Detailed Description of the Invention

[0001]

FIELD OF THE INVENTION This invention relates to computer systems, and more particularly to methods for determining the color components of scanned computer images and displaying and storing such images.

[0002]

BACKGROUND OF THE INVENTION Computer graphic images usually consist of multiple points or pixels. Each pixel has a code representing the gradation range of that pixel. In monochrome systems, this tonal range is from pure black on the gray scale to pure white. This sign or grayscale level is typically 8 bits and provides a tonal range of 0 to 255, where 0 represents either pure black or white and 255 is the opposite color (ie pure white or not). Represents pure black).

[0003]

In a color graphics system, colors are created by "mixing" the three primary colors red, green and blue (RGB). The RGB model can be described as an RGB color cube in three-dimensional space, where black (ie the absence of color) is the origin and white is at the opposite end of this cube. Any point on this cube or on its side represents the color given by: C = rR + gG + bB where C is the color representing the pixel, R, G, B represent the primary colors, and r, g, b correspond to the relative amounts of R, G, B. Thus, each pixel is some combination of red, green and blue corresponding to the above equation.

Color images are usually stored by allocating 24 bits per pixel. in this way,
8 bits are allocated for red, 8 bits for green, and 8 bits for blue. The range of each of these 8-bit codes ranges from 0 to 255. The 24-bit method becomes difficult depending on the number of pixels per color image. The number of pixels in the scanned image depends on the size of the image and the resolution of the scanner.
For example, an 8.5 x 11 inch letter size sheet scanned at 300 dots per inch (dpi) requires 25 megabytes of memory. Therefore, processing of image data requires a large amount of memory and time. Another problem with using the 24 bit / pixel scheme is that all 24 bits are used regardless of the actual color components of the target image. Therefore, 24 bits per pixel are used to represent and store the image, regardless of whether the scanned image actually has color.

Another color model useful in computer graphics is the double hexagonal pyramid model,
That is, there is an HLS (Hue, Lightness, Saturation) model. The term hue is used here in the conventional sense. That is, it represents the color of the image. The term saturation describes the purity of a color. "Pure" color means a color with 100% saturation. The degree of saturation decreases as white, black, or gray is added to a color until it becomes a single color with a saturation of 0%. Brightness is the intensity or brightness of a pixel.

A hexagon is formed when an RGB color cube is projected onto a plane along the diagonal from white to black. Using the relationship of this shape, the RGB color cube is divided into David.
F. Rogers is Procedural Elem
nts for Computer Graphics
It can be converted to an HLS double hexagonal pyramid using a simple non-trigonometric algorithm such as the one described in. Basically, the value of lightness (I) changes along the axis from the black vertex to the white vertex. The saturation (S) is determined by the radial distance from the central axis. When the saturation (S) = 1, color saturation occurs. The hue (H) fluctuates at 360 ° around the central axis and is indefinite when the saturation (S) = 0.

[0007]

The present invention provides a new method for classifying the type of information present in a computer graphics image. In the following description, "type", "auto sort", "image type" and "information type" refer to the content of the scanned image. There are three main image type classifications. That is, (1) binary image,
(2) Gray scale image and (3) Color image. Color types are further classified as continuous tone or spot colors. Binary type refers to an image consisting of black and white information. Examples of binary type images include text pages and line drawings (ie, black ink drawn on white paper).

An image of the gray scale type consists of black, white and gray components. Grayscale images are monochromatic. That is, such an image has no meaningful color components. An example of a grayscale image is a picture taken in black and white film.

Color type images are usually three primary colors, red,
It consists of a mixture of green and blue (RGB). Color images also contain some monochromatic components, but enough meaningful color components to distinguish them from grayscale images. Color type images can be further classified as either continuous tone or spot color. A continuous tone color image is a mixture of many colors of different hues. An example of a continuous tone color image is a landscape photograph taken using a color film. A spot color image is usually an artificial figure containing a few distinct conceptual colors. A color bar diagram is an example of spot colors.

The present invention automatically classifies an image by determining the color components of the computer graphics image. The method of the present invention is to first convert the RGB values of the scanned or stored image to HLS values using a non-triangulation algorithm. Next, the HLS value is used to determine the color component of the image. According to the method of the present invention, it is possible to classify images more accurately than the conventional method.

In the method of the present invention, the saturation value is first used to determine whether the image is chromatic or monochromatic. A low saturation indicates that the image is a monochromatic image, and a high saturation indicates that the image is a chromatic image. If the saturation value is an intermediate value, the hue is used as an indicator of saturation. A yellowish hue represents a monochrome image. The lightness value is used to further characterize the monochrome image as either a grayscale image or a binary image.

An advantage of the present invention is that images can be automatically classified by their color components. This automatic image classification enables the memory allocation of graphic images according to the conditions of the images, which facilitates more efficient representation and storage of images.
A full color image can be stored using the 4 bit / pixel method, a grayscale image can be stored using 8 bit / pixel, and a binary image can be stored using 1 bit / pixel. .

[0013]

1 is a block diagram of a general purpose digital computer system 100 using the present invention. Processor 11
0 is connected to the system bus 105. The system bus 105 includes a processing device 110, a memory 120, a data storage disk 130, and an input / output interface device 14.
Facilitates communication between 0s. This memory stores the software of the present invention and all the data collected and generated by the present invention. An area 125 is set in the memory 120 for storing the present invention, which will be described in more detail below. The input / output interface device 140 is provided between the system bus 105 and the display device 142 and the keyboard 1
Controls data communication between 44 and scanner 146.

The scanner 146 can be either a grayscale scanner or a color scanner. The grayscale scanner returns a grayscale value for each pixel scanned. This value is typically 0 to 25
8 bits in the range of 5, where 0 represents pure black and 2
55 represents pure white. The present invention is not limited to these values, and the display of these values can be interchanged. Color scanners typically return three values for each pixel (ie red, green, blue). Each value is 8 bits in the range of 0 to 255, and a total of 24 bits per pixel is red-green-blue (RGB) image data.
Various scanner schemes can be used without departing from the scope of the invention. For example, the scanner may return more than 8 bits for each RGB value, which increases the total bits of RGB image data per pixel. In one embodiment, the scanner 146
Can be referred to as "ScanJet IIc" or "ScanJet IIp", both of which are sold by Hewlett-Packard Company, a California corporation of Palo Alto, California.

FIG. 2 shows the outline of the software of the present invention. Scanner software 200 is provided to control the flow of image data between computer system 100 and scanner 146 (see FIG. 1). The scanner software 200 executes several functions such as a data structure allocation / initialization function 210, a measurement preview function 220 and an automatic classification function 230. Scanner software 200 performs other scanner functions, which are not included in the invention.

The measurement preview function 220 includes a 24-bit processing routine 211 and an 8-bit processing routine 215.
Call one routine. 24-bit processing routine 21
1 calls the HLS function 212 and the RGB pixel measurement function 213. The automatic classification function 230 is the color automatic classification function 23.
1 and the grayscale automatic classification function 235 are called. The color auto-classification function 231 has the ability to call the grayscale auto-classification function 235.

The 24-bit processing routine 211 works in conjunction with the RGB pixel measurement function 213 to "scanjet II.
Collect and "measure" data retrieved by a color scanner such as "c". "Measuring" an image simply means recording a statistic of pixels (eg, determining the number of pixels having a red value greater than 200). Since a typical image has a large amount of data, 24-bit processing routine 211 can perform its calculations on strips of image data. Therefore, the 24-bit processing routine 211 is called many times until the entire image is acquired and measured. On the other hand, the automatic classification function 230 uses the measurement preview function 220 (the 24-bit processing routine 211 and the RG
Called only once to make a classification decision after all measurements have been made (via the B pixel measurement function 213). The 24-bit processing routine 211 can also collect and measure image data in a single call, depending on the components of the image and the amount of memory available (120 in FIG. 1).

FIG. 3 is a flow chart showing the overall flow of the method of the present invention. The method begins at block 310
In this block, red-green-blue (RGB) values are generated from the scanned raster image. Such RGB values can be assigned by the scanner (146 in Figure 1) or generated by scanner software (200 in Figure 2). In block 320, the RGB values are
For example, it is converted to Hue-Lightness-Saturation (HLS) values according to a non-triangulation algorithm such as that proposed by Rogers. At block 330, the color components of this image are determined based on these HLS values, as described in detail below with reference to FIGS. 6-9. At block 340, the final result (color components of the image) is output. This output is one of three major types: (1) color, (2) grayscale, or (3) binary (ie black and white). Color types are further classified as either continuous tone or spot color.

At block 350, block 340.
The image is displayed on the display device (142 in FIG. 1), the image is stored in the memory (120 in FIG. 1), or both, based on the output from the device. According to this method, it is easy to represent and store the image. This is because the number of bits used to represent and store the image is minimal. For example, if the output from block 340 is automatic color classification, then at block 350 the image is displayed and stored using a 24-bit / pixel scheme. Similarly, if the output from block 340 is grayscale auto-classification, the display (142 in FIG. 1) is monochromatic and at block 350 the image is stored using the 8-bit / pixel scheme. Binary images are represented and stored using only one bit per pixel.

FIG. 4 shows the relationship between the RGB color space and the HLS color space. The origin 410 of the RGB color cube is transformed into the first vertex 410 of the HLS hexagonal pyramid. This 410 represents black. White is represented by the opposite corners of the 420 RGB color cube. This can be transformed into a second vertex 420 of the HLS hexagonal pyramid, the grayscale image is less saturated (ie, closer to the axis of the HLS hexagonal pyramid), the black vertex 410 to the white vertex 420. Range up to.

FIG. 5 shows the RGB-HLS conversion function (3 in FIG. 3).
20 and a more detailed flow diagram of 212 of FIG. 2). For this function, floating point calculations were converted to fixed point calculations to speed up the RGB-HLS conversion process.

This function results in one pixel of the image scanned in block 510. Block 5
At 20, the brightness value of the pixel is determined based on half of the sum of the maximum RGB value and the minimum RGB value. For example, when the RGB value of a pixel is red = 10 green = 20 blue = 30, the brightness value at this time is (30 + 10) / 2 = 20, the maximum value is blue = 30, and the minimum value is Red = 10.

At block 520, control is passed to block 530 which determines the saturation value of the pixel. Maximum R
A small value where there is a difference between the GB value and the minimum RGB value (empirically this small value is 15), the pixel is grayscale and therefore the saturation is set to zero. Alternatively, the degree of saturation is set (1) based on some number equal to or less than the ratio of this difference and its sum, or (2) equal to a number indicating complete saturation. It should be noted that when the function sets the saturation value to a large value, this large number is equivalent to "1" because the calculation has been converted to fixed point.

Once the saturation is determined, control is passed to block 540 which determines the hue value of the pixel. Hue is RG
Based on the determination of the maximum B-value, this maximum is used to obtain the difference between this maximum and the other RGB values. For example, using the above values, the maximum value is blue =
Thirty. The difference is then calculated as: (Blue-green)-(blue-red) That is, (30-20)-(30-10) =-10

Next, this difference is used as the numerator and the numerator of the ratio of the difference between the maximum RGB value and the minimum RGB value.

At blocks 520, 530 and 540, the decision is modified by adding and / or multiplying the result by an integer to use fixed point arithmetic. By this fixed point calculation method, the processing device (see FIG.
110) of RG is faster than when using floating point calculation
The B value can be converted to an HLS value.

Once the RGB values have been converted to HLS values, at block 550 the pixel is assigned to the appropriate HLS array. The array is a counter that counts the pixels corresponding to the specified characteristics. A preferred routine for recording the HLS characteristics of each pixel is shown in Table 1 below.

[0028]

[Table 1]

Other color space conversions can be performed without departing from the scope of the invention. For example, conversion from RGB to HIV (hue, lightness and value) translational motion or RG
HLS conversion of RGB can be performed without departing from the scope of the present invention such as conversion from B to Munsell color system.

Once the pixel has been recorded and counted, control is passed to decision block 560 which determines if the pixel can be classified as yellowish gray. The lightness value is between 178 and 228 and the hue value is 6
If less than 0, the pixel is considered to be a pale yellowish gray and block 565 increments the counter by one. Returning from block 570 to block 510, the next pixel is obtained until each pixel of the scanned image has been transformed. Once each pixel has been transformed, block 570 passes control to a color auto-classification function, which is described in detail below with reference to FIGS.

FIG. 6 shows the automatic color classification function (FIG. 3) of the present invention.
330 and 231) of FIG. 2 are shown. Once the RGB to HLS conversion function has converted all pixels to the HLS color space, block 620 counts the total number of pixels associated with the image, and if this number is a small threshold number, then as the type of image. Returns "color". In one embodiment, 10 is selected as this threshold number.
The threshold number is used because of uncertainty when determining the image type of an image that has only a few pixels. The default value "color" is selected as a user preference. Different thresholds and default image types can be selected without departing from the scope of the invention.

Next, block 640 calls the grayscale auto-classification function (235 in FIG. 2) when the image is gray or greyish-yellow. The grayscale automatic classification function determines whether the image is a grayscale image or a binary image based on the RGB values of the pixels included in the image.

At block 660, it is determined whether the image is a text page based on the number of light yellowish gray pixels (see 560 and 565 of FIG. 5 and associated description) and high saturation. Is performed. If the image is a text page, block 660 returns a "binary" image type.

If the conditions of blocks 620, 640 and 660 are not met, block 680 returns "color" as the image type.

FIG. 7 shows a more detailed flow chart of this automatic color classification function. In particular, FIG. 7 shows 620 and 64 in FIG.
The beginning of 0 is shown enlarged. At block 710, which is the starting point of the enlarged view of 620 in FIG. 6, variables used in the automatic color classification function are initialized. Such variables include the total number of pixels used, the number of pixels having a hue value, and characteristic percentages such as gray percentage, yellow percentage, and the like. When such a variable is initialized to zero, the number of pixels used in block 712 is counted. The number of pixels used is not necessarily the total number of pixels in the image. The image data can be analyzed using every nth pixel (n is an integer greater than 1). In one embodiment, n is the ratio of the total number of pixels in the scan area of interest divided by the optimal number of pixels needed to determine the type of image, the optimal number being 20000.

When the number of pixels used is counted,
At decision block 720, it is determined whether there are enough pixels to perform automatic classification. If the number of pixels used does not exceed the threshold amount, block 720
Passes control to block 722 which sets the automatic classification to the default value "COLOR" and then to block 724 to return to the main program. If you have enough pixels,
Block 720 passes control to block 730 for further analysis. In one embodiment, this pixel threshold amount is 10 and any less than that is considered inadequate and the method stops the process, simply block 72.
The "color" automatic classification is returned via 2 and 724. "COLOR" returned by automatic classification is selected as a default value according to user preference. Most users use computer systems that have color capabilities and expect color images. "Grayscale" or "binary" has been used as the default automatic classification.

If the number of pixels used is greater than 10,
The method enters 640 of FIG. 6 to determine if the image is a grayscale or binary image. Initial calculations are performed for saturation and hue. At block 730, the number of pixels having a hue is calculated. Hue [12]
The colorless phase pixels in the array are basically grayscale pixels. Next, at block 732, the percentage of grayscale pixels is calculated. The gray percentage is equal to the number of low saturation pixels divided by the number of pixels used. Low saturation pixels are saturated [0 ..
2] in the array.

FIG. 8 is a continuation of FIG. 7 and is a further enlargement of 640 of FIG. At decision block 810, it is determined whether there are pixels that have a hue value, and if there are no pixels that have a hue, then control is passed to decision block 820.
And in decision block 820 it is determined whether the percentage of gray pixels is less than or equal to 0.9. If there are pixels with hue, block 810
Passes control to block 812 which calculates the percentage of yellowish pixels. A "yellowish" pixel is a pixel in the array of hues [1..2]. The percentage of yellowish pixels is determined by dividing the number of yellowish pixels by the number of pixels having a hue value. After the percentage of yellowish pixels has been calculated in block 812, block 812 determines control block 8
Return to 20.

When this point in the flow chart is reached, it is determined whether this image is a grayscale image or a binary image. This method makes this determination by determining whether the image is gray or gray with a yellowish hue. Decision block 820
The threshold for a "gray" image, which first determines if the image is gray, is 90% or 0.9. Therefore, if the percentage of grayscale pixels calculated in block 732 of FIG. 7 is greater than or equal to 0.9, control is passed to decision block 830 to make further grayscale determinations. If the percentage of grayscale pixels is less than 0.9, control is passed to block 910 of FIG.

Decision block 830 determines if the percentage of grayscale pixels is more grayed than the one passed in block 820 as being very close to pure gray. If the percentage of grayscale pixels is 0.99 or greater, then the image is likely to be a grayscale or binary image. Control is passed to block 832, where block 832
Prior to the invocation of the grayscale auto-classification function by block 840 (235 in FIG. 2), two counters, a COLOR counter and a SOME-COLOR counter, are initialized to zero. These two variables are zeroed so that the auto-classification function bypasses the color decision after the presence of color components is denied by blocks 820 and 830. Grayscale automatic classification function (23 in FIG. 2)
5) calls "Grayscale" or "Binary" autoclassification.

If the percentage of grayscale pixels is less than or equal to 0.99, control is passed to decision block 834 which determines if the percentage of yellowish pixels is greater than 0.9. The hue is yellowish and the percentage of grayscale pixels is greater than 0.9 (block 82).
(Determined by 0), this means a binary image. Therefore, if the percentage of yellowish pixels is greater than 0.9, control is passed to block 832, which calls “COLO” before the grayscale auto-classification function call by block 840.
2 of "R" counter and "SOME-COLOR" counter
Two variables are initialized to zero. If the percentage of yellowish pixels is less than or equal to 0.9, control transfers to block 910 of FIG.

FIG. 9 is a continuation of FIG. 8 and is an enlarged view of 660 of FIG. When you reach this point in the flow diagram, whether the image is a color image or a text page (ie 2
Value) must be determined. At block 910, the percentage of pixels that can be classified as light yellowish gray is calculated. As described above in connection with FIG. 5, the brightness values are 178 and 22.
If it is between 8 and the hue value is less than 60, then this pixel is considered as a pale yellowish gray. Block 565 of FIG. 5 maintains a counter of the number of pixels that are considered light yellowish gray. Therefore, block 910 divides this number by the number of used pixels determined in block 712 of FIG. 7 to determine the percentage of light yellowish gray pixels. At block 912, the percentage of highly saturated pixels is calculated. Saturation [7 ..
9] is considered to have high saturation. These arrays are added and the sum is divided by the number of pixels used to obtain the percentage of highly saturated pixels.

Next, at decision blocks 920 and 922 it is determined whether the image is a text page. The percentage of light yellowish gray pixels is greater than 0.75 (block 920) and the percentage of highly saturated pixels is less than 0.01 (block 9).
22), the image is considered a text page, control is passed to block 924, where the automatic classification of the image is set to binary (binary). If the result of block 920 or block 922 is no, then in block 930 automatic classification is set to color. In an alternative embodiment, block 930 may further determine if the color image is continuous tone or spot color. The continuous tone type is a type in which many colors having different color tones exist. Spot color types are generally artificial graphic images with a few distinct colors that are distinguishable from each other. Block 94
At 0, control is returned to the main program, in the case of an optical data link the scanner program (200 in FIG. 2).
A preferred routine for implementing this automatic color classification function is shown in Table 2 below.

[0044]

[Table 2]

Although illustrated and described with reference to embodiments of the invention, the invention is not limited to the particular structure illustrated. Those skilled in the art will appreciate that various modifications can be made within the scope of the claims without departing from the spirit and scope of the invention. For example, the automatic classification function (230 in FIG. 2) can be modified to identify the actual color of a particular image. This is for example block 93.
This can be done by adding a new routine that determines the color of the image by examining the hue value of each pixel after the 0 and making an overall color determination that represents the scanned image.

As described above, the method of determining the color component of the image according to the present invention can be configured as shown below. [1] A method for determining a color component of graphic information in a scanned image having one or more pixels,
(A) scanning the image into a computer system, (b) counting the one or more pixels scanned in the computer system to produce a result, and outputting the result to the scanned image. First in total
(C) assigning a value to the one or more pixels contained in the scanned image, each of the assigned pixels having one or more memory locations. And storing each value in the first color space, (d) converting each value in the first color space to a second space,
(E) determining the color components of the scanned image based on the values converted in step (d).

[2] The method according to [1], wherein the first color space is composed of red values, green values, and blue values.

[3] The method according to [2], wherein the second color space comprises a hue value, a lightness value, and a saturation value.

[4] The method described in [3] above, wherein
(F) determining whether the total number of scanned pixels is equal to a threshold amount, and (g) returning the automatic color classification if the total number of scanned pixels is less than the threshold amount. Step (h) determining if the scanned image is gray or greyish yellow if the total number of scanned pixels is greater than or equal to the threshold amount; (i) the image Is gray or greyish-yellow, determining the grayscale type of the scanned image, (j) determining whether the scanned image is a text page, and scanning the scanned image. From the step of returning a binary auto-classification if the image is a text page, and (k) returning the color auto-classification if the scanned image is not a text page It is that way.

[5] The method described in [4] above, wherein
Calculating a first number representing the number of pixels having a hue value by subtracting (i) representing the number of pixels having no hue value from (i) the total number of pixels scanned, (m) ) Calculating a second number representing the percentage of less saturated pixels by dividing the number of less saturated pixels by the total number of scanned pixels, (n) said first number being one or more. Calculating a third number representing the percentage of pixels having a yellowish hue by dividing the number of pixels with a low hue value by the first number, (o) said second number being 0. 9 or more, and if the second number is smaller than 0.9, proceed to step (k), (p) if the second number is 0.9 or more, Whether the number of 2 is more than 0.99 Constant, and if the second number is 0.99 or more, the step proceeds to step (i), (q)
When the second number is smaller than 0.99 but is 0.9 or more, it is determined whether the third number is 0.9 or more, and when the third number is smaller than 0.9 , Step (j), and (r) if the third number is 0.9 or more, the method proceeds to step (i).

[6] The method described in [5] above, wherein
Step (j) (s) expresses the percentage of pixels having a light yellowish gray hue value by dividing the number of pixels having a light yellowish gray hue value by the total number of pixels scanned. The step of calculating the number of
(T) calculating a fifth number representing the percentage of pixels with high saturation values by dividing the number of pixels with high saturation values by the total number of pixels scanned, (u) wherein the fourth number is Determining if it is greater than 0.75 and returning an automatic classification of colors if the fourth number is less than 0.75, (v) if the fourth number is greater than 0.75, Determining if the number 5 is less than 0.01, and returning the automatic color classification if the fifth number is greater than 0.01; (w) the fourth number is greater than 0.75. , The fifth number is less than 0.01, a binary automatic classification is returned.

[0052]

The method of determining the color components of the computer graphic image according to the present invention, which has the above-described configuration,
Scanning the image, counting the pixels scanned, storing the result in a first memory location as the total number of scanned images, assigning a value to the pixels contained in the scanned image, Each of the stored pixels in one or more memory locations, each value being the first
Of the first color space and the second value of the first color space.
By converting to the space of, and determining the color component of the scanned image based on the converted value, it becomes possible to perform the memory allocation of the graphic image according to the condition of the image, A more efficient representation and storage of the image can be easily performed, for example, a 24-bit / pixel system can be used to store a full-color image, and 8-bit / pixel can be used to store a grayscale image. Binary images can be stored using bits / pixels, which is a great advantage in that computer graphics images can be classified more accurately and automatically than in the conventional method.

[Brief description of drawings]

FIG. 1 is a block diagram of a general purpose digital computer system using the present invention.

FIG. 2 shows an outline of software of the present invention.

FIG. 3 is a flow diagram of the overall flow of the method of the present invention.

FIG. 4 is a relationship diagram between an RGB color space and an HLS color space.

FIG. 5 is a more detailed flow chart of the RGB to HLS conversion function.

FIG. 6 is a flowchart of the automatic image classification function of the present invention.

FIG. 7 is a more detailed flow chart of the automatic image classification function.

FIG. 8 is a more detailed flowchart of the automatic image classification function.

FIG. 9 is a more detailed flowchart of the automatic image classification function.

[Explanation of symbols]

100 general-purpose digital computer system 110 processor 105 system bus 120 memory 130 data storage disk 140 input / output interface device 125 area 142 display device 144 keyboard 146 scanner 200 scanner software 210 data structure allocation / initialization function 220 measurement preview function 230 automatic classification Function 211 24-bit processing routine 215 8-bit processing routine 212 HLS function 213 RGB pixel measurement function 231 Color automatic classification function 235 Grayscale automatic classification function 310, 320, 330, 340, 350 block 410 RGB color cube origin 420 HLS hexagonal pyramid Second vertices of 510, 520, 530, 540, 550, 560, 5
65, 570, 620, 640, 660, 680, 71
2, 720, 722, 724, 730, 732, 81
0, 812, 830, 832, 834, 910, 91
2,920,922,924,924,930,940 blocks

Claims (1)

[Claims] 1. A method of determining color components of graphical information in a scanned image having one or more pixels, the method comprising: (a) scanning the image into a computer system; The one scanned into the computer system
Count one or more pixels and give the result,
Storing the result in a first memory location as the total number of scanned images, (c) assigning a value to the one or more pixels contained in the scanned image Storing each of the pixels in one or more memory locations, each value representing a first color space, (d) converting each value of the first color space to a second space A method for determining a color component of a computer graphic image, which comprises the steps of: (e) determining a color component of the scanned image based on the values converted in step (d).