JP2963046B2 - Image processing method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION [0001] 1. Field of the Invention [0002] The present invention relates to an image processing method for generating correction data used when performing a correction process for generating image data to be output to an output device. . 2. Description of the Related Art For example, Japanese Patent Application Laid-Open No.
JP-A-89577 discloses a method in which a test pattern is formed in an image forming apparatus, the density of the formed test pattern is detected, and a pulse per pixel of a modulation signal corresponding to a halftone image is detected in accordance with the detected density. A technique for controlling the width is disclosed. However, in the above-mentioned prior art, since the gradation characteristic is controlled by controlling the pulse width in the image forming apparatus, the reproduction of an image with rich gradation can be performed with high accuracy. If the control is attempted, there is a problem that the control of the pulse width becomes complicated and the configuration of the control circuit on the image forming apparatus side becomes complicated. On the other hand, on the side that supplies image data to the image forming apparatus, various types of image processing are usually performed, and if the image forming characteristics of the image forming apparatus are known, the image data is converted according to the characteristics. It is relatively easy to perform the conversion on this, and it is desirable to effectively use the capability of the supply side. SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and has an output characteristic and a floor characteristic of an output device so that a good output image can be obtained by the output device.
It is an object of the present invention to provide an image processing method capable of creating correction data based on tone characteristics . [0006] In order to achieve the above object, for example, the present invention has the following arrangement. That is, an image processing device that creates a correction table used when performing a correction process of generating image data to be output to an output device,
The correspondence relationship between a plurality of input data and output data having different levels indicating the original reference characteristic of the output device.
A reference characteristic table and a gradation correction table indicating a target gradation characteristic are stored, and test image measurement data obtained by measuring an image output based on test image data by the output device is input. Means for creating a correction table to be used when performing correction for output to the output device based on the reference characteristic table, the test image measurement data, and the gradation correction table. For example, when the correction table data is created, an interpolation process is used. Embodiments of the present invention will be described below in detail with reference to the accompanying drawings. <First Embodiment> FIG. 1 is a block diagram showing the arrangement of an image forming apparatus according to a first embodiment of the present invention, and shows a case where the present invention is applied to an electrophotographic digital copying machine. I have. Here, A is a reader unit, and B is a printer unit. A normal copying operation according to the configuration of FIG. 1 is as follows. First, the original 1 is read by the reading element (CCD) 2 of the reader unit A, and the original image is converted into an analog electric signal. The electric signal is transmitted to an amplifier (AMP) 3
, And 8 bits by the A / D converter 4 (= 256 gradations)
, And input to the RAM 10 for use gradation correction via the latch 5. Further, the RAM 10
D / A converter 1 of the printer unit B
3 and converted into an analog image signal, and the signal is input to the pulse width modulation circuit 14. FIG. 2 is a diagram for explaining an example of the pulse width modulation operation. In the figure, "image signal" is an analog image signal output from the D / A converter 13, and "pattern signal"
Is a triangular wave signal generated in the pulse width modulation circuit 14. The "image signal" and the "pattern signal" are synchronized as shown in the figure, and a pulse signal obtained as a result of level comparison of these by a level comparator in the circuit 14 is a PWM signal output from the pulse width modulation circuit 14. is there. The above is an example of the pulse width modulation operation, and is suitable for relatively high-speed image signal processing. On the other hand, when the image signal is relatively slow, D / A
Even without using the converter 13, for example, a pulse width modulation signal can be directly generated from the digital image signal by generating a digital pattern signal sufficiently faster than the digital image signal and comparing these digitally. Further, the PWM of the output of the pulse width modulation circuit 14
The signal is amplified by the laser driver 15 and the laser generation circuit 1
6 to control ON / OFF of the laser beam. The laser light emitted from the laser generation circuit 16 is irradiated on the photoreceptor 19 via an optical system including a polygon mirror 17 and an f-θ lens 18 and the like. The photoreceptor 19 is uniformly charged by the corona charger 20, and then receives the above-mentioned laser exposure to form an electrostatic latent image on the surface. This electrostatic latent image is visualized by a developing device 21 and then transferred onto a transfer material 27 by a transfer charger 24. The transfer material 27 is separated from the photoreceptor 19 by a separation charger 25 and then fixed. Vessel 26
Is fixed by On the other hand, the toner remaining on the photoreceptor 19 without being transferred is collected by the cleaner 22, the electrical history of the photoreceptor 19 is erased by the pre-exposure lamp 23, and the next print cycle is started again. Next, a method for testing the characteristics of the printer unit will be described. To test the characteristics of the printer section, first, a test image signal is generated by the pattern generator 12 and is input to the D / A converter 13 instead of a normal image signal. In the present embodiment, the pattern generator 12 is set to 0
Test image signals of 256 levels from 0H to FFH (H is hexadecimal) can be generated. However, during an actual test, a test image signal for every 16 levels is output. That is, the density level is changed from the white level of 00H to FF.
When the black level is H, the test image signal is 0
A total of 17 levels of 16 levels of 0H, 10H, 20H, 30H,..., E0H, F0H and the last FFH level are generated. Next, the density pattern based on the test image signal is transferred onto the transfer paper 27 in exactly the same manner as in a normal printing operation. FIG. 3 is a diagram showing an example of a test pattern formed on transfer paper. In the figure, a density pattern from a white level to a black level in the paper passing direction is represented in 17 levels. The positions where these density patterns are generated are positions that can be sequentially read by a densitometer 28 described later. Next, the density patterns thus transferred are sequentially read by a reflection type densitometer 28. And C
A select signal from the PU 6 selects and energizes an address line of the buffer 8 and a data write line (downward direction) of the bidirectional buffer 9, and sequentially reads density patterns via the I / O circuit 11. The detection data of
The write signal from the PU 6 is written into the table 2 of the RAM 10. FIG. 4 is a diagram showing output characteristics (Table 2) obtained by testing the printer section. In the figure, the horizontal axis is the D / A input level given by the pattern generator 12, and the vertical axis is the output density detected by the densitometer 28. still,
The output density is standardized so that the white level is 0.0 and the black level is 1.0. .., Etc., which are not included in the test image signal detected by the densitometer 28, are interpolated by appropriate calculations. Furthermore, in this embodiment, since the output density is processed by an 8-bit signal, the output density level can be assigned to 0.000 to 1.275 in 0.005 steps. That is, it is possible to cope with the case where the black density is higher or lower than the standard. Next, based on the test results, the RAM
A method for correcting the used gradation correction table (table 1) in 10 will be described. For this purpose, it is necessary to first describe the contents of various tables accommodated in the RAM 10 and their functions. That is, in the RAM 10, a used gradation correction table (Table 1 in FIG. 9) for performing gradation correction of a digital image signal at the time of actual printing, and a test printer output table (Table in FIG. 2) and a test gradation correction table (Table 3 in FIG. 8) used to correct the used gradation correction table based on the tested printer output characteristics.
The RAM 10 has a separate backup power supply so that the contents of the RAM 10 are retained even when the main body power supply is turned off. The ROM 7 of FIG. 1 stores a standard gradation correction table (Table 4) as shown in FIG. 5 and a standard printer output table (Table 5) as shown in FIG. In the standard gradation correction table, the input image signal level in the RAM 10 is normally corrected to a D / A input level having a characteristic as shown in FIG.
A standard property that should be done. The standard printer output table is a standard characteristic that, when a linear D / A input level (00H to FFH) is given to the horizontal axis, the printer output density on the vertical axis should be as shown in the figure. This is a reference for comparing the output characteristics of the test printer in FIG. Next, the CPU 6 creates a test gradation correction table shown in FIG. 8 from the standard printer output characteristics shown in FIG. 6 and the test printer output characteristics shown in FIG. That is, FIG. 7 shows the standard printer output characteristics and the test printer output characteristics simultaneously.
In FIG. 8, the correspondence between the D / A input level ns of the standard printer output characteristic and the D / A input level nm of the test printer output characteristic corresponding to the same output density is obtained as shown by an arrow. A test gradation correction table as shown in the figure is created by taking the horizontal axis and the corresponding nm on the vertical axis. In FIG. 7, when the output density exceeds dr, the value of ns is inherently present even if it exceeds nsr, but it can be seen that the value of nm is no longer saturated with FFH. This relationship is clearly shown in FIG. Next, the CPU 6 reads the D / A input level ns corresponding to each image signal level from 00H to FFH from the standard gradation correction table of the ROM 7 as shown by the arrow in FIG. This operation is a standard γ (gamma) correction conversion operation. However, since the above-described test gradation correction table is created in a non-linear form, a standard printer output as shown in FIG. 6 cannot be obtained if a standard gamma correction conversion operation is performed in the ROM 10 even in actual printing. I understand. Therefore, the CPU 6 creates or modifies the used gradation correction table of FIG. 9 by the following method. That is, FIG.
The value ns of the vertical axis (D / A input level) read by the standard gradation correction table is applied to the horizontal axis of the test gradation correction table of FIG. Each ns is converted into a corresponding nm, and the value nm is made to correspond to the D / A input level on the vertical axis in FIG.
That is, according to the used gradation correction table of FIG. 9, the horizontal axis indicates the actually used image signal level, and the vertical axis indicates the D / A which has been subjected to characteristic conversion so as to always obtain an ideal output characteristic for the printer. The input level will be obtained. In general, a sufficiently high image quality can be obtained by the use gradation correction table described above. However, if the output characteristics of the printer greatly deviate from the standard characteristics, for example, the density at the black level (FFH) becomes the standard. When the density becomes significantly higher than the black density, or when the density becomes significantly lighter, the gradation of the dark portion of the original image may be impaired.
In order to prevent this to some extent, it is more effective to make the following additional correction. That is, for example, in the case of the test printer output characteristic as shown in FIG. 4, the black level is slightly thinner than the standard.
For example, in the test gradation correction table shown in FIG.
In the range of sr, nm is always at the FFH level, and the gradation of the input image signal from nsr to the FFH level is lost. To prevent this,
For example, as shown in FIG. 10, when the value of mn or ns exceeds a certain threshold value nmt or nst, the relationship between ns and nm may be replaced by a specific functional relationship from there until FFH to perform interpolation. . FIG. 10 shows an example in which this section is linearly complemented. In addition, if interpolation is performed using a quadratic curve or the like, the gradation characteristics can be further smoothed. The above method is also effective when the test output characteristic of the printer is darker than the standard at the black level. In this case, the dynamic range of the output can be widened. In the above example, the black (FFH) level of the test printer characteristic fluctuates.
0H) Even when the level fluctuates, for example, mn ≦ nmt ′
A threshold value nmt 'may be provided to interpolate between the 00H level and the nmt' level. However, even if the fluctuation near the white level is subtle, it tends to cause fogging and flying,
Since deception is not effective in terms of visual characteristics, a method other than linear approximation, such as using a quadratic curve, is required as an interpolation method. The test tone correction table as shown in FIG. 8 can be used as a guide for checking the current state of the printer as it is. For example, at the time of maintenance, this correction table is output to an appropriate display means, and ns = n
The current state of the printer can be known by looking at the deviation from the ideal linear state of m. In particular, if the correlation between the deterioration of the developer of the developing unit 21 and the deterioration of the photoconductor 19 and the change in the correction table is known in advance, it is possible to detect an image trouble in advance and to determine the cause thereof. Can be easily determined. Alternatively, it is more effective to use the contents of the correction table in combination with the automatic self-diagnosis function of the printer. As described above, according to the present invention, since the state of the machine can be known based on the correction table, the serviceability is significantly improved. <Second Embodiment> In the first embodiment, the densitometer 28 as shown in FIG. 1 was required. This configuration will be extremely effective if the present invention is configured as a printer only. However, there are drawbacks that the densitometer itself becomes a cost-up factor and that the density test cannot be performed over a wide range. Therefore, a second embodiment in which this point is further improved will be described next. The apparatus according to the second embodiment uses the reader A in FIG. 1 in place of the densitometer 28. FIG. 11 is a block diagram of an image forming apparatus according to a second embodiment of the present invention. The same components as those in FIG. First, a method for testing printer characteristics will be described. Also when detecting the printer characteristics, a test image signal is formed by the pattern generator 12 and input to the D / A converter 13 instead of a normal image signal.
In the pattern generator 12, as in the case of the first embodiment, the levels from the 00H level (white) to 10H, 20H, 30
A total of 17 levels of H,..., E0H, FOH and FFH level (black) are generated. Next, the test image based on the test image signal is transferred onto the transfer paper 27 by a normal printing operation in the same manner as in the first embodiment. FIGS. 12A to 12D are views showing some examples of density patterns on the transfer paper 27. FIG. For example,
In FIG. 12 (a), margins a and b are provided at the leading end and the trailing end of the paper, and the density is changed stepwise from 00H (white) to FFH (black) every 16 levels in the other portions. The difference from FIG. 3 of the first embodiment is that the density pattern area is widened by outputting each density widely over the entire area in the main operation direction. Next, returning to FIG. 11, the original 31 of the test density pattern created as shown in FIG. 12 is read again by the CCD 2 from the reader unit A as shown in FIG. At this time, there is no need to operate the printer unit B. At this time, the document 31 needs to be placed on the document table in parallel with the paper passing direction at the time of print output as shown in FIG. However,
Which of the FFH level and the 00H level should be the top may be determined in advance. Also, the array of the test density patterns does not necessarily need to be such that the density changes in one direction. For example, dark and light ones may be arranged alternately, or may be random. In short, if the order of generation of the test patterns by the pattern generator 12 is known, it is possible to always cope with the sub-scanning position at the time of reading. Thus, by always setting the original 31 at a fixed position on the original platen in a fixed direction, the reader unit A knows what level of density is always read during scanning by associating it with the sub-scanning position. be able to. FIGS. 12B to 12D show another example in which a test density pattern is formed. Among them, for example, FIG. 12 (c) shows channel information (density information) corresponding to each test density pattern written with a bar code beside the pattern. It is possible to always know what level of density is being read without detecting the position. Next, after the test image signal read from the CCD 2 is converted into a digital signal by the A / D converter 4, in this case, the density data of the same group is cumulatively added through the adder 33 and the register 34. Calculates the average value for each density level. As described above, when density data is collected over a certain wide area and its average value is obtained, the influence of partial density unevenness and noise can be significantly reduced. Then, the data of the average value of the image signal level obtained by reading each density level is plotted on the vertical axis, and
Test printer output table like Table 4 (Table 6)
Create Again, untested horizontal axis 01
The levels on the vertical axis corresponding to H, 02H,... Are interpolated by appropriate calculations. As can be seen from FIG. 14, the test image data read from the CCD 2 includes all the characteristics of the CCD 2, the amplifier 3, the A / D converter 4, etc., and the vertical axis thereof is shown in FIG. For example, it is different from the one standardized on the basis of a physical quantity in consideration of visual characteristics such as reflection density. Of course, the test printer output table of FIG. 14 obtained as described above may be used as it is, but here, in order to clarify the physical meaning, this information is temporarily stored in the test printer of FIG. Output characteristics (Table 2)
Will be converted to a form like For this purpose, FIG.
A table (Table 7) indicating the input characteristics of the reader unit A is prepared in the ROM 32 in advance as shown in FIG. Table 7 shows how the input density is normalized with respect to the read image signal level value.
In order to match the input density on the vertical axis with that of FIG.
After standardizing the standard black level at the time of copying to 1.00, 00H is set to 0.0 and FFH is set to 1.70. Thus, the table 6 of FIG.
5, the test printer output characteristics (table 8) as shown in FIG. 16 are obtained. Accordingly, this actually shows exactly the same relationship as the printer output characteristics (Table 2) of FIG. Therefore, the subsequent processing may be the same as that described in the first embodiment. <Third Embodiment> In the first embodiment and the second embodiment described above, FIG.
Using a test density pattern as shown in (a) to (d), a method of knowing a test image signal value generated by the pattern generator 12 from a sub-scanning position or a code display of the pattern is used. Therefore, it is necessary to arrange the test density patterns with some level of margin (step) so as not to cause misreading between adjacent patterns.
On the other hand, although the overall configuration is not shown in the present embodiment, for example, the test image signal from 00H to FFH is continuously changed to some extent finely by the pattern generator 12 in FIG. 11 of the second embodiment. So that
A gradation-like test density pattern as shown in FIG. 17 is printed out, read out again from the reader unit A, and a cumulative frequency histogram at this time (FIG. 18) is created. To create. Also in FIG. 18, the ordinates of “frequency distribution” and “cumulative frequency distribution” are normalized from 0.0 to 1.0. As an example of a method for creating the test printer output characteristics, first, when creating the gradation pattern shown in FIG. 17, the time for outputting each of the test image signals from 00H to FFH is made equal. In addition, the test image signal is changed linearly by one gradation. Then, only by normalizing the ordinate of the cumulative frequency in FIG. 18 from 0 to 255, the same characteristics as in the case of the table 6 in FIG. 14 in the second embodiment can be obtained. As described above, the third embodiment is the same as the second embodiment up to the point where the test density pattern is output from the printer unit B, but the output test density pattern is read again. When reading from the unit A, it is only necessary to count the number of data for each input level that has been read. Therefore, it is only necessary to pay attention to the start and end of reading of the test density pattern. It is not necessary to take correspondence with the read density data.
Furthermore, even when shading due to mechanical or electrical noise occurs during the creation of the test density pattern, the density is always an increasing function due to the nature of the cumulative frequency histogram, so a decrease such as inversion of the correction curve occurs. Can be prevented, and furthermore, the unevenness of the curve due to noise is smoothed, and there is an advantage that the noise becomes stronger. When the gradation density pattern as shown in FIG. 17 is read from the reader section A, the test image signal may be read from the 00H side or from the FFH side. It is not necessary to pay attention to the left and right directions of the density original when setting the original in the copy section A, so that a mistake during service can be prevented. This is also due to the property of the cumulative frequency histogram. For the same reason, when a gradation pattern is created, even if the gradation pattern is not always linearly changed from 00H to FFH as shown in FIG. good. However, if an extremely different level is arranged adjacently, an erroneous density may occur due to an edge effect of development. As described above, according to the present invention, the output device is designed so that a good output image can be obtained by the output device .
Since the correction table data is created based on the output characteristics and the gradation characteristics of the device, it is possible to obtain a good output image. [0042]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an image forming apparatus according to a first embodiment of the present invention; FIG. FIG. 4 is a diagram illustrating an example of a density pattern formed on a transfer sheet according to the first embodiment; FIG. 4 is a diagram illustrating a tested printer output characteristic according to the first embodiment; FIG. 5 is a first embodiment; FIG. 6 is a diagram showing standard tone correction characteristics of the printer of FIG. 6, FIG. 6 is a diagram showing standard output characteristics of the printer of the first embodiment, FIG. 7 is a diagram showing standard printer output characteristics of the printer of the first embodiment; FIG. 8 is a diagram showing test printer output characteristics in comparison, FIG. 8 is a diagram showing test gradation correction characteristics of the printer of the first embodiment, and FIG. 9 is a use gradation correction of the printer of the first embodiment. FIG. 10 is a diagram showing characteristics, and FIG. 10 is another test chart of the printer according to the first embodiment. FIG. 11 is a diagram showing a tone correction characteristic, FIG. 11 is a block diagram of an image forming apparatus according to a second embodiment of the present invention, and FIG. 12 is a diagram showing a density pattern on a transfer sheet 27 according to the second embodiment. FIG. 13 is a diagram showing some examples, FIG. 13 is a diagram showing a relationship between a sub-scanning direction of a reader unit and a test density document, and FIG. 14 is a diagram showing output characteristics of a tested printer of the embodiment. FIG. 15 is a diagram showing density conversion characteristics of the second embodiment; FIG. 16 is a diagram showing tested printer output characteristics of the second embodiment; FIG. 17 is a diagram showing the third embodiment; FIG. 18 is a diagram illustrating an example of a density pattern on transfer paper. FIG. 18 is a diagram illustrating a histogram characteristic of a tested printer output according to the third embodiment. [Description of Signs] 1 Original 2 CCD 3 Amplifier 4 A / D converter 5 Latch 6 Microprocessor 7 ROM 8 Buffer 9 Bidirectional buffer 10 RAM 11 I / O port 12 Pattern generator 13 D / A converter 14 Pulse Width modulation circuit 15 Laser driver 16 Semiconductor laser 17 Polygon mirror 18 f-θ lens 19 Photoconductor 20 Corona charger 21 Developer 22 Cleaner 23 Pre-exposure lamp 24 Transfer charger 25 Separate charger 26 Fixing roller 27 Transfer paper 28 Densitometer 29 A / D converter 30 Bus 31 Reference density output pattern 32 ROM 33 Adder 34 Register

Claims (1)

(57) [Claims] An image processing method for creating a correction table used when performing a correction process for generating image data to be output to an output device, comprising: a plurality of input data having different levels indicating an original reference characteristic of the output device; Data correspondence
A reference characteristic table and a gradation correction table indicating a target gradation characteristic are stored, and test image measurement data obtained by measuring an image output based on test image data by the output device is input. An image processing method for creating a correction table to be used when performing correction for output to the output device based on the reference characteristic table, the test image measurement data, and the gradation correction table. 2. 2. The image processing method according to claim 1, wherein an interpolation process is used when creating the correction table data.