JPS6052164A - Picture processing device - Google Patents

Abstract

PURPOSE:To process properly picture information by setting a slice level used for binary-coding or multi-value-forming a picture element density signal by means of the operation processing so as to widen a region obtaining an average value of the picture element density through the circuit constitution of a minimum scale. CONSTITUTION:An input picture element density signal is added sequentially by a full adder 1, the result of addition is stored once in a D FF2 and the result is fed sequentially to a comparator 3, a D FF4, a comparator 8 and a D FF9. The output of the FFs 4, 9 is fed respectively to the comparators 3, 8 the result is compared with an average value of an input picture element density, a value larger than the average value and smaller than the average value are outputted respectively to D FFs 10, 5 from the comparators 8, 3. The maximum value and the minimum value of the average value of the input picture element density from the FFs 5 and 10 are latched respectively to the FFs 4, 9 and fed to an operation processing unit 7 via D FFs 6 and 11. Then the slice level to be binary -coded is set at the device 7 by the operation, the result is fed to a comparator 12 thereby simplifying the configuration of the device.

Description

[Detailed Description of the Invention] [Technical Field] The present invention relates to an image processing device that detects the pixel density of a document image using an image sensor, converts it into at least two values, and reproduces the image. This allows image information to be automatically reproduced with high contrast.

[Prior art]

In this type of image processing apparatus, it is conceivable to detect only the ground color of an image original, or the density of the ground color and image information to determine the slice level used for λ value conversion.

Therefore, when sequentially comparing pixel densities during main scanning with threshold values in order to obtain image information of a document image, when setting a required threshold value based on the maximum and minimum values of image density during main scanning, It is possible to detect the uniform background color of image originals, but in diazo originals etc.
It is impossible to set an accurate threshold value to detect pure white pixels and perform so-called background skipping.

Furthermore, it is also impossible to deal with bit defects in the image sensor. Therefore, it is recognized that it is effective to detect the average value of pixel density within a certain appropriate range.

However, since pixel density data is formed by main-scanning a document image with an image sensor at high speed,
It is difficult to temporarily accumulate sequential pixel density data using a multi-line buffer memory and then detect the average value of pixel density within a predetermined image area spanning multiple main scanning lines by sub-scanning.) The wider the image area in which the average value of pixel densities is calculated, the better, but problems such as cost disadvantages arise.

〔the purpose〕

It is an object of the present invention to solve the above-mentioned problems and eliminate their drawbacks, and to widen the image area in which the average pixel density is calculated using a minimum-scale circuit configuration, so that the average pixel density in the original image can be calculated using the minimum circuit configuration. It is an object of the present invention to provide an image processing device capable of detecting almost in real time and performing binarization processing of image data corresponding to an original image.

In order to achieve such an object, the image processing apparatus of the present invention calculates the maximum and minimum values for each of multiple main scans with respect to the average density value obtained by averaging the image pickup output pixel density signal of the original image for multiple pixels in the main scanning direction. Then, based on the average of the maximum and minimum values for each predetermined area, the slice level used to convert the pixel density signal into λ values or multi-values is set by arithmetic processing. The slice level is set in near real time without using a special memory device.

〔Example〕

In the following, the invention will be described in detail by way of example embodiments with reference to the drawings.

First, an example of the configuration of the image processing apparatus of the present invention is shown in FIG. In the illustrated configuration, 1 is a full adder with 1O bit configuration, and 2 is a D flip-flop, which temporarily holds the data of the addition result of the full adder and also supplies the pixel density to the outside. The signals are sequentially supplied to the full adder l in synchronization with the signal transfer clock, and added to the sequential input pixel density signals. Note that since the input pixel density signal has a t-bit configuration, 16 input pixel density signals can be sequentially added in the main scanning direction. Therefore, if the input pixel density signals in the effective area are added sequentially from the beginning of the main scan, and when the addition of 76 input pixel density signals is completed, the upper 6 bits of the data of the addition result are extracted. An average value of 76 pixel densities is obtained.

The data of the pixel density average value obtained as described above is supplied to the comparator 3 and the D flip-flop l, and the output of the D flip-flop l is supplied to the comparator 3.
The input pixel density average value is compared with the input pixel density average value.
Q", so no matter what value the first input pixel density average value is, the comparator 3 determines that the input average value is larger, and controls based on the determination result. The output of the D flip-flop 5 is set to H'' level. Therefore, the D flip-flop l controlled by the "H" output of the D flip-flop 5 latches the average value data of pixel density /B as an initial value, and at the same time, as described above, sequential addition of input pixel density signals is performed. D slip 70 with lθ bit configuration for holding
The knob is cleared to "O".

When the average value of the pixel density of the continuously input / is obtained as described above, the comparator 3 calculates the difference between the previous pixel density average value latched in the D flip-flop 41 and the input pixel density average value. The comparison is made, and only when the current input pixel density average value is larger, the D flip-flops 1 and 2 are operated sequentially according to the operation procedure described above, and as a result, the D7 flip-flop 1 stores the input up to that point. The maximum value of the pixel density average values will be retained.

The above-described operation of detecting and holding the maximum average value of the image tip density is sequentially repeated within the effective image area of the pixel density signals that are sequentially transferred along the main scanning direction. therefore,
At the end of the main scan, the average pixel density of 110 pixels obtained by dividing an image area with a width of '//6'' and a length of approximately/lrOmm into unit areas of l≦number×/ This means that the maximum value, that is, the pixel density value closest to black, is obtained.

Furthermore, this pixel density maximum average value detection operation is continuously repeated in the range of each main scan in the sub-scanning direction, and at the end of the main scan, the output of the D-flip flitter is used. The maximum pixel density value D within the A tnn X /lrOtntg region is detected.

ax The maximum pixel density value D obtained every l main scan by sub-scanning as described above is supplied to the D flip 70 which is operated by applying a clock to the latch aX every l main scan, and then the next pixel density is It is held during the scanning period and is also supplied to an arithmetic processing unit (CPU) 7 which applies a latch clock to the interrupt terminal every l main scan. Therefore, in the arithmetic processing unit 7, maximum pixel density value data can be taken in at a speed of about 6 ms.

Moreover, the comparators t and D in the configuration shown in FIG.
flip flop? , /θ, //c, the above-mentioned comparator 3 and D flip-flop l.

It operates in exactly the same way as 4 and has the same function, and detects the minimum pixel density value D in the '/4' x /10 am area. Note that, unlike the above-mentioned D flip-flop l, the D7 flip-flop 91n can preset its output in the initial state and latch the pixel density average value for the top/≦pixel of the main scan as the initial value. Do it like this.

As described in detail above, according to the configuration shown in FIG. 1, the image area over the main scanning section can be scanned by the combination of the comparator and the D flip-flop without using an expensive memory device. It is possible to detect the maximum and minimum values of pixel density in the image data, and the data acquisition by the arithmetic processing unit (CPU) can be performed at a relatively low speed of every 6 mS, so a dedicated arithmetic processing unit is required. (apu) can be used for other purposes, such as subsequent image data processing.

Furthermore, as described above, in the image processing apparatus of the present invention, since the average value of the pixel density is detected for each pixel along the main scanning direction, bit defects may occur in the output of the image sensor. It is possible to prevent the influence on pixel density data in the event that this occurs. Furthermore, since the detected maximum pixel density value Dmax is relatively small for a thin line image, there is no risk of missing pixel density data that is required to be processed based on the maximum pixel density value. On the other hand, on the contrary, the minimum pixel density value D ・ is, for example, 1
n To accurately and faithfully detect the density of the ground color even when the density of the ground color is not uniform, such as in a diazo original image, and the ground color is created by a mixture of white and black pixels. This makes it possible to perform pixel density detection for image processing extremely effectively and appropriately.

Next, based on the maximum value D and minimum value D of the pixel density obtained as described above, a slice λ of the pixel density signal is performed in the arithmetic processing device (CPU).
The algorithm for determining the slice level for value conversion will be explained in detail.

First, it is generally said that it is effective to select the slice level for converting pixel density signals to -i (Dmax+Dmin), but the above-mentioned Dmax and Dmin
The value of n is a value obtained based on the image area of ///≦X/mm, and it is not necessarily appropriate to select the slice level based only on this value. In other words, when determining the slice level by blisscanning as in conventional devices, or when determining the ground color visually, the average value of pixel density is calculated over a wide range of kanashi. Therefore, although the device of the present invention is basically a sequential type, the arithmetic processing device (C
By intervening with PU), it is possible to add past data to the new pixel density average value data for each main scan that is sequentially captured, and perform averaging processing over a wider image area. becomes.

Generally, the average value 7N of N pieces of data What are the determining factors? Therefore, any value can be obtained as long as it is an integral multiple of . On the other hand, based on each data of DTnaxn and Dmin n in the nth pixel block and the slice level Ln+ for the nth pixel block, the slice level Ln+ for the n+/@th pixel block is determined by the algorithm.

In this equation, for example, if N=32, the average value will be found over the past 32 blocks, that is, over the range of fms. Therefore, for example, if the beam exists along the main scanning direction by about 1 m, the influence will be extremely small since it will not affect the slice level, and disturbances in the reproduced image due to sudden changes in the slice level can be suppressed. On the contrary, if a pitch-black image occurs in steps over several G, the slice level will reach the highest threshold level later than the t-simplified image, and the entire screen will become white with λ value. This will cause an inconvenience. Therefore, the processing unit (
Maximum and minimum pixel density value data D supplied to CPU)
It is necessary to impose restrictions on max and Dmin. −
For example, proceed as follows.

When Dmaxn<A' Dmax n-4n-4D
When n-+Dminn>B・Dmin n″Dmin
n-1, that is, if the maximum value Dmax is extremely close to pure white, or if the minimum value Dmin is extremely close to pure black, such maximum or minimum value is not used as slice level calculation data, and the immediately preceding n- The /th maximum value or minimum value held in a register is substituted for the nth data.

D used for a 1ftran area at the leading edge of an image original
As the value of maxlDmin, it is sufficient to use the values of A and B, respectively.

Next, in this type of image processing apparatus, the relationship with the variable resistor for adjusting pixel density, which is usually provided in the console section, will be described. In the image processing apparatus of the present invention, an optimal reproduced image can be obtained without requiring manual operation. However, it is also possible to control the image density by manual operation. That is, for example, when the adjustment lever is set at the center of the density adjustment scale range, ll! according to the above formula! i
7. When automatic adjustment of image density is performed and the adjustment lever is set to a scale other than the center, it is also possible to configure the slice level to be manually adjusted in accordance with the set scale. That is, Ln+,=[Ln+i((Dmaxn-Dminn)×
vR+Dminn-Ln)] In the formula, ■ corresponds to the relative scale range of the image density adjustment lever, and as shown in Figure 1, o-i, o
shall take the value of . Note that by using the conditions for A and B described above, even if the density adjustment lever is set to the lightest level or the lightest level, the slice level will not reach a level below A or above 8, It is possible to faithfully reproduce images even for solid white or solid black images.

/7 in the configuration shown in the first figure represents such a variable resistor for adjusting image density, which supplies the density setting value as an 1-bit signal to the arithmetic processing unit (CPU) 7.

Furthermore, 12 in the configuration shown in the first diagram is a comparator that compares the 6-bit image data of the input pixel density signal with the 6-bit slice level signal from the arithmetic processing unit (CPU) 7 to convert it into λ-valued image data. Convert.

Next, in the image processing apparatus of the present invention, an algorithm for slice level calculation when performing multi-value conversion of pixel density signals will be explained. Generally, a multilevel latent image visualized by a laser beam printer is visualized by controlling the laser beam irradiation time per pixel and changing the area of the beam spot. For example, 3
When visualizing using a value-level printer, the beam spot that makes up 7 pixels is divided into two halves, the first half and the second half.
Image reproduction is performed for each pixel using three density levels obtained when both are turned on and irradiated, when both are turned off and not irradiated, and when only one is turned on and only one is irradiated. Let's do it. therefore,/
ii! The input pixel density signal is converted into a ternary density signal using different slice levels in the first half and the second half of the ii element.

Two types of slice levels L'' t L'''l'l used when performing image reproduction at such ternary density levels are set by the following equations as shown in FIG.

Therefore, when converting a pixel density signal into three values, as described above, the slice levels L11+ and L12+ of one stage are linearly set at the same rate as the setting scale of the density adjustment lever changes. How to change the slice level Lf of one stage as shown in Figure G or Figure S.
il and L(21 are both changed linearly,
There is a method of making the rate of change different from each other, or making it further different between the first half and the second half of the adjustment scale range. The slice level L111 shown in the first diagram and the fifth diagram,
The settings of L(21 are performed using the following calculation formulas.

In the case shown in FIG. 4: Next, FIG. 6 shows an example of the circuit configuration of the portion that should be modified from the configuration shown in FIG. 1 when the pixel density signal is ternarized as described above. In the illustrated configuration, the arithmetic processing unit (CPU) 7,! Slice levels L[1+ and Lll of 1l12 steps are set and supplied to comparators 2/ and n, respectively, and the input pixel density signals are converted into λ values, respectively. Each λ-valued image data from the comparators l and 22 is supplied to a multiplexer n, and is configured into an l-bit configuration by alternately switching to a clock signal that is twice as fast as the image data transfer clock applied from the outside. Convert to ternary image data.

〔effect〕

As is clear from the above description, according to the present invention, when a pixel density signal obtained by reading an original image by an image sensor is converted into λ values or multivalued and reproduced, the ground color of the original image and the density of image information are For example, first, /7 along the scanning direction
Determine the average pixel density value for each pixel, then detect the maximum and minimum values of the average pixel density for each main scan, and further,
The maximum and minimum values are calculated by 1crIL through calculation processing.
Since the slice level for λ-value conversion or multi-value conversion is determined based on the average value for each degree range, a simple configuration consisting of comparators, D flip-flops, etc. can be used without the need for expensive memory devices. Appropriately captures only the image information of the original image in real-time based on the circuit device.
The special effect of being able to convert into values or multi-values can be obtained.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a configuration example of the image processing apparatus of the present invention, and FIGS. 2 to S show examples of slice level settings for binarization and ternarization of the image density, respectively. Graph FIG. 6 is a block diagram showing another configuration example in which a part of the configuration example shown in FIG. 1 is changed. l... Full adder, λ, 4', j, O* 9 + 10 * //...D
Flip-flop, j, I, /2... Comparator, 7... Arithmetic processing unit (CPU), /7... Variable resistor for concentration adjustment, 2/, 22... Comparator, B... multiplexer. Patent applicant Canon Co., Ltd. Figure 1 Figure 6

Claims (1)

[Claims] 1) In an image processing device that converts a pixel density signal formed by reading a document image using an image sensor into a λ value, the pixel density of a plurality of consecutive pixels along the main scanning direction of the document image is calculated. average value detection means for detecting an average value; maximum/minimum value detection means for detecting a maximum value and a minimum value of the average value of the pixel density for a plurality of unit blocks over a plurality of main scans; and for each of the plurality of unit blocks. A slice level setting means for calculating a slice level by averaging the maximum and minimum values of the average values of the pixel densities, and a λ value conversion means for binarizing the pixel density signal based on the slice level. An image processing device characterized by: 2. The image processing apparatus according to claim 1, further comprising comparison means for comparing the maximum and minimum values of the average values of pixel densities with predetermined values, respectively, wherein the maximum value and the minimum value are respectively compared with the predetermined values. An image processing apparatus characterized in that when a predetermined value range is exceeded, the slice level is calculated by replacing the slice level with the maximum value and the minimum value l unit blocks before. 3) The image processing apparatus according to claim 1, further comprising a density specifying means for specifying an image density, and the image density specified by the density specifying means and the slice level value calculated by the slice level setting means. An image processing device characterized in that a slice level is set according to 4) In the image processing apparatus according to claim 1, the average value detection means detects the average value of the pixel density by excluding a plurality of consecutive pixels at both ends of the main scan. Characteristic image processing device. 5) An image processing apparatus according to claim 1, wherein at least one of said means is constituted by a microprocessor. 6) In an image processing device that converts a pixel density signal formed by reading a document image by an image sensor into an M value, where M is a positive integer, the pixel density is calculated for a plurality of consecutive pixels along the main scanning direction of the document image. average value detection means for detecting the average value of the pixel density, maximum/minimum value detection means for detecting the maximum and minimum values of the average values of the pixel densities for the unit blocks spanning multiple main scans, and for each of the plurality of unit blocks. slice level setting means for calculating M-1 slice levels by averaging the maximum and minimum values of the image density average values, and calculating the pixel density signal based on the M-/ slice levels, respectively. An image processing device comprising M-/ binarization means each converting into a λ value. 7) An image processing apparatus according to claim 6, wherein the M-/ slice levels calculated by the slice level setting means are sequentially increased.