JPH08251408A - Image processing unit - Google Patents

Abstract

PURPOSE: To eliminate a background properly by detecting a parameter representing noted color image information to be a background portion based on lightness information and saturation information and deciding a threshold level and a coefficient based on the result of detection of the parameter. CONSTITUTION: A lightness detection section 50 detects a lightness and a saturation detection section 70 detects a saturation from color image data read by an optical scanner 10. A range of data detected by a lightness range gate section 60 and a saturation range gate section 80 is given to a parameter control section 90. The parameter control section 90 applies control in response to the detected lightness, saturation and range of data to a background detection section 20 and a background elimination section 30 in real time. A frequency detection memory section 100 connects to the parameter control section 90 to store a histogram, of the lightness information and the saturation information equivalent to one line of a color original and the control section 90 controls background detection and background elimination in response to the content of the memory.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for reading color image data from the outside and removing background components from the read color image data and outputting the color image data to the outside, such as a color printer or a color copying machine.

[0002]

In Japanese Patent Laid-Open No. 5-207280, image data read by a scanner is converted into lightness L component data and chromaticity a and b component data, and only the L component is the background. A background removal processing method for performing detection / removal processing is disclosed. Since the a and b components do not change due to the background removal processing, the background removal output (image data) can be obtained with the hue / saturation information stored.

[0003]

In the case of a color manuscript, the background portion may have a color, and in the background removal processing that focuses on only density information, an appropriate background detection / removal processing is performed. Difficult to do. Generally, the brightness information of color image data obtained by reading the background portion of a color original is high and the saturation information is low in many cases. However, in the conventional method, the background detection / removal processing is performed only on the L component of the Lab color system corresponding to the lightness information of the color image data, but the control considering other components (a and b components) is not possible. Not done.

The present invention aims to improve the above points. That is, it is an object of the present invention to provide an image processing apparatus capable of performing more appropriate background detection / removal by paying attention to information other than brightness.

[0005]

According to a first aspect of the present invention, there is provided an image processing device which detects background information from color image information read by an image reading means, and the color corresponding to the background information according to the background information. Background removal means for performing background removal processing on image information; brightness detection means for extracting brightness information from the color image information; saturation detection means for extracting saturation information from the color image information; From the detection result of the brightness detection means and the detection result of the saturation detection means,
Detection control means for controlling the background detection processing in the background detection means based on the detection result of the parameter detection means, the parameter detection means detecting a parameter indicating the degree to which the color image information of interest is the background portion. When,
Is provided.

The image processing apparatus according to the second aspect of the present invention is a background detecting means for detecting background information from the color image information read by the image reading means, and a background for the color image information according to the background information. Background removal means for performing removal processing, brightness detection means for extracting brightness information from the color image information, saturation detection means for extracting saturation information from the color image information, and detection result of the brightness detection means From the detection result of the saturation detection means and a parameter detection means for detecting a parameter indicating the degree to which the color image information of interest is a background portion, the background removal is performed based on the detection result of the parameter detection means. Removal control means for controlling the background removal processing in the means.

[0007]

According to the image processing apparatus of the first aspect of the present invention, the background information of the color image data is extracted by the color background detecting section, and the background information is output to the color background removing section. The color background removal unit performs background removal processing according to the background information on the color image data and outputs it. The parameter control unit detects a parameter indicating the degree of a background portion from the detected brightness and saturation information, and controls background detection and background removal processing in accordance with the detected parameter. For example,
Control is performed so that background detection and background removal are strengthened for image data with high brightness and low saturation. More appropriate background detection processing and / or background removal processing can be performed.

[0008]

【Example】

-First Example- The configuration of the first example of the present invention is shown. First, the color image data obtained by reading a color original with the optical scanner unit 10 is
The color background detection unit 20 detects the background information, and the color background removal unit 30 performs a predetermined color background removal process according to the result. After that, the printer unit 40 outputs it to a predetermined recording sheet.

Further, the color image data input from the optical scanner unit 10 includes the lightness detecting unit 50 and the saturation detecting unit 7.
Input to 0, these detect lightness and saturation. The detected lightness information and saturation information have their data ranges detected by the lightness range gate unit 60 and the saturation range gate unit 80, respectively, and the range detection result is input to the parameter control unit 90. The parameter control unit 90 controls the background detection unit 20 and the background removal unit 30 in real time according to the detected brightness and saturation information and the data range. Further, the parameter control unit 90 is connected to a frequency detection unit 100 having a memory capable of storing the occurrence frequency (histogram) of the lightness information and the saturation information for one line of the color original, and the parameter control unit 90 is The background detection control and background removal control are performed according to the contents of the memory. The specific configuration of each unit will be described below.

Optical Scanner Section 10: FIG. 2 shows the configuration of the optical scanner section 10. The optical scanner unit 10 reads image data from a document and outputs color component image data R (red component), G (green component), and B (blu component) to the outside. The light emitted from the fluorescent lamp 11 strikes the document, and the reflected light passes through the first mirror 12, the second mirror 13, and the third mirror 14, and is imaged by the lens 15 on the light receiving surface of the CCD 16. The light and shade of the original is converted into an electric signal by the CCD 16, and the signal is amplified by the amplifier 17, and then the shading correction, dark current correction,
A / D converter 19 that has been subjected to signal correction such as density setting
Thus, each pixel is converted into image data of 8 bits of R, G and B. Here, in the CCD 16, the photoelectric conversion element for each pixel is arranged in the main scanning direction (direction perpendicular to the paper surface of FIG. 2).
Are linear CCDs arranged in a row in the
The CD 16 moves along the document in the sub-scanning direction (horizontal direction in FIG. 2) to read the document. The R, G, B image data has a dynamic range of 255 → 0 from light to dark, and the larger the value represented by the image data, the brighter.

Color Background Detection Unit: 20 FIG. 3 shows the configuration of the color background detection unit 20. The color background detection unit 20 detects background information on the document from the input R, G, B image data. Optical scanner unit 10
The R, G, and B image data taken in are given to the comparator 21, R and G, to the comparator 22, G and B, and to the comparator 23, B and R. Comparator 2
1 to 23 compare R and G, G and B, and B and R, and give the comparison result to the logic circuit 25. The logic circuit 25 is
The maximum value of the R, G, B image data is detected. The detection result is transmitted to the multiplexer 24 and the maximum value of the input R, G, B image data (hereinafter,
max (RGB)) is output from the multiplexer 24 to the comparator 26.

The above max (RGB) is the comparator 2
6, the comparator 26 is compared with the threshold value held in the register 27, and when max (RGB) is larger than the threshold value, the comparator 26 causes the multiplier 28 to output max (R).
GB) × weighting coefficient output, and when the threshold value is larger, threshold value × weighting coefficient output is instructed. In response to this, the multiplier 28 makes
When x (RGB) is large, the data obtained by multiplying max (RGB) by a predetermined weighting coefficient is output to the color background removing unit 30 as background information GND, and when the threshold value is larger. , The threshold value multiplied by the weighting factor
Is output to the color background removal unit 30 as background information GND. The threshold value given to the comparator 26 by the register 27 is the value written in the register 27 by the parameter control unit 90 (background control information 1). When the threshold value is large, the background detection is weak and small. And become stronger. If max (RGB) is smaller than the threshold value, the background information G
ND is proportional to the threshold value, and the color background removing unit 30 converts the image data showing a higher (brighter) value than the background information GND into a brighter higher value as shown by a chain line in FIG. Therefore, the background removal is weak when the threshold is large, and strong when the threshold is small.

Color background removal section 30: FIG. 4 shows the structure of the color background removal section 30. In the color background removal unit 30,
The background removal processing is performed by using the output GND of the color background detection unit 20. The R, G, B image data read by the optical scanner unit 10 is independently subjected to the background removal processing, and each of the R, G, B has a configuration as shown in FIG. In each of the skin removing sections 30R, 30G, and 30B, each image data is first compared with the output GND of the color background detecting section 20 by the comparators 31R, 31G, and 31B, and the image data is GND (the horizontal axis of FIG. "Image part" and "background part"
And the boundary value) and the multiplexer 34R,
Instructions for selecting the outputs of the multipliers 33R, 33G, 33B are given to the 34G, 34B. Multipliers 33R, 33G, 3
3B, R, G, B image data, the coefficient value corresponding to the image data (corresponding to the image data at that time in the coefficient value group in the registers 32R, 32G, 32B) The product obtained by multiplying the image data is output to the multiplexer 24. Thereby, the multiplexers 34R, 34G, 34
As for the output of B, when the input image data is the image portion (less than GND), the data indicated by the solid line in FIG. 5 (the same as the input image data) is output, but the input image data is the background portion. In the case of (greater than or equal to GND), the data (data with the background removed) shown by the alternate long and short dash line in FIG. 5 is output. Incidentally, the registers 32R, 32
The parameter control unit 90 writes a coefficient value group (background control information 2) in G and 32B. If the coefficient value is large, the image data in the background area is converted into a brighter high value, and the background removal is strong. If the coefficient value is small, the background removal is weak.

Brightness detecting section 50: FIG. 6 shows the brightness detecting section 50.
Shows the configuration of. In the lightness detection unit 50, the optical scanner unit 1
The brightness information is extracted from the R, G, B image data starting from 0. The R, G, B image data from the optical scanner unit 10 is input to the matrix calculation unit 51. The matrix calculation unit 51 performs multiplication calculation for converting R, G, B image data into lightness information. The matrix memory unit 52 holds the multiplication coefficient matrix data for the above calculation,
The contents can be rewritten from the outside. A detailed description of the matrix calculation is omitted.

Saturation detection unit 70: FIG. 7 shows the saturation detection unit 70.
Shows the configuration of. In the saturation detection unit 70, the optical scanner unit 10
Saturation information is extracted from the R, G, B image data from.
With the same configuration as the brightness detection unit 50, R, G, B
Image data is converted into lightness information.

Brightness range gate unit 60: FIG. 8 shows the structure of the brightness range gate unit 60. In the lightness range gate unit 60, the lightness information provided by the lightness detection unit 50 is stored in the register 62.
The comparator 61 detects whether it is higher than the threshold value of, and outputs the result (low: H / high: L) to the parameter control unit 90.

Saturation Range Gate Unit 80: FIG. 9 shows the structure of the saturation range gate unit 80. In the saturation range gate unit 80, the comparator 81 detects whether the saturation information given by the saturation detection unit 70 is larger than the threshold value of the register 82,
The result (large: H / small: L) is output to the parameter control unit 90.

Frequency Detection Unit 100: FIG. 10 shows the configuration of the frequency detection unit 100. The frequency detection unit 100 calculates the occurrence frequency (the number of occurrences of the same value: histogram) of the brightness information and the saturation information of the R, G, B image data of one line before read by the optical scanner unit 10 from the color original. It is held and the frequency information is sent to the parameter control unit 90. Frequency detector 1
00 has a lightness frequency memory 102 that holds the lightness frequency and a saturation frequency memory 104 that holds the saturation frequency. The memory contents are initialized at the beginning of each line, and after initialization, the contents of the address corresponding to the input value is incremented for each frequency memory each time the brightness information and saturation information is input. It As a result, every time reading of one line of the color original is completed, the frequency distribution of the brightness information and the saturation information of the line is stored in the memories 102 and 104. At this time, a parameter control unit 9 described later
0 reads the frequency distribution (histogram) of the brightness information and the saturation information of the memories 102 and 104, and reads the brightness data of the highest frequency.
The data (memory address) and the most frequent saturation data (memory address) are taken into the latch 1 (D91, E91) and the latch 2 (D95, E95).

Parameter control unit 90: FIG. 11 shows the configuration of the parameter control unit 90. The parameter control unit 90 receives the outputs of the brightness detection unit 50, the saturation detection unit 70, the brightness range gate unit 60, the saturation range gate unit 80, and the frequency detection unit 100, and detects the color background from the combination of these outputs. Detection parameter of the unit 20 (threshold of the register 27: background control information 1) and removal parameter of the color background removal unit 30 (coefficient group of registers 32R to 32B:
The background control information 2) is calculated and parameters are set for each. That is, write to each register. The brightness information input to the parameter control unit 90 is stored in the register 1 (D9
3) and the multiplier 1 (D94), the saturation information is similarly output from the register 2 (D97) and the multiplier 2 (D94).
98), weighting (coefficient multiplication) is performed, and the weighted data are added by the adder D99. Here, registers 1 (D93), 2 (D
97) holds weighting coefficients of lightness and saturation information respectively, and the CPU 1 built in the parameter control unit 90.
Can be rewritten sequentially. The above addition result is
It is input to the decoder D100 as a decode address,
The CPU 1 stores data corresponding to the address value in the ROM (D
101). The read data is held by the latch D102.

The output of the saturation range gate unit 80 and the output of the brightness range gate unit 60, which are input to the parameter control unit 90, are ANDed by the AND circuit D103, and the brightness range is high (H) and the saturation range is high. Is low (H), it is output to the latch D102 as a data latch signal. Only when both outputs of the two gate sections 60 and 80 are H (high brightness and low saturation), the contents of the latch D102 are stored in the ROM (D101) addressed by the decoder output. The read data is rewritten, and if not, the latch D102 holds the previous contents.

Further, the CPU 1 of the parameter control unit 90 is
Lightness / saturation frequency memory 102, 10 of the frequency detection unit 100
The contents of No. 4 are read every time one-line processing is completed, and the lightness and saturation information having the highest frequency is detected. The detected highest frequency lightness and highest frequency saturation are
Latch 1 (D91) and Latch 2 (D95) respectively
The difference between the brightness information and the saturation information, which are stored in the parameter control unit 90 and are input to the parameter control unit 90, is calculated by the difference unit 1 (D92) and the difference unit 2 (D96). The CPU 1 rewrites the values of the register 1 (D93) and the register 2 (D97) according to the difference value. For example, when the difference value between the maximum frequency saturation and the input saturation information is small, the CPU 1 sets the content of the register 2 (D97) holding the weighting coefficient for the input saturation information to the value currently set. Set to a large value (same for brightness information).

The parameter control unit 90 has the above-described configuration.
It has a pair (CPU1, CPU2) independently, and each has
Threshold for color background detection unit 20 (background control information 1)
And the coefficient group (background control information 2) to the detection parameter calculation unit 30 is stored in the ROM (D101, E1) as described above.
01), stores them in the latches D102 and E102, and supplies them to the color background detecting section 20 and the color background removing section 30, respectively. Incidentally, ROMD101 and E10
1, the larger the address (higher lightness, lower saturation), the stronger the background detection and background removal (smaller threshold for background detection) background control information 1 and (background removal coefficient). The background control information 2 is held.

Overall Operation: The overall operation of the first embodiment will be described below. R captured by the optical scanner unit 10,
The background information of the G and B image data is extracted by the color background detecting unit 20, and the background information is extracted from the color background removing unit 30.
Is output to The color background removal unit 30 performs background removal processing according to the background information on the R, G, B image data read by the optical scanner 10 and outputs it to the printer unit 40.

The parameter control unit 90 calculates and sets parameters (background control information 1, background control information 2) for adjusting the processing of the color background detection unit 20 and the color background removal unit 30 from the detected lightness and saturation information. To do. For example, when an input with low saturation and high brightness is made, the parameter control unit 90 makes the background detection and background removal processing parameters (background control information 1 and background control information 2) stronger for background detection and background removal. Values to be processed (background control information 1 representing a small threshold value, background control information 2 representing a large coefficient) are set. If the lightness information detected by the lightness detection unit 50 is lower than a predetermined threshold value (a portion that is considered to be other than the background portion), the latch D102,
Since E102 does not latch and holds the previously latched value, the parameter calculated last time is used (the same applies to the saturation information). Further, by reading the frequency distribution of the brightness and saturation information of the previous line held in the frequency detection unit 100, even when the input brightness and saturation information is a value close to the maximum frequency value of the previous line. , The parameters (background control information 1 representing a small threshold value, background control information 2 representing a large coefficient) on which the background detection and background removal processing are strongly performed are selected.

Other Embodiments: In the first embodiment, the color image information is read by the optical scanner 10, but the color image information may be read by other means. Further, in the first embodiment, the printer unit 40 is provided at the final stage, but the output unit may not be included. Further, in the first embodiment, inputs R, G,
Although the color background detection processing (FIG. 3) is performed using the maximum value max (RGB) of B image data, the present invention can be applied to background detection means using another method. In addition, in the first embodiment, the color background removal processing using the data GND obtained by the color background detection processing (FIG. 3) as a threshold value is performed (FIG. 4), but the background removal means using another method is used. Can also be applied to the present invention. Further, in the first embodiment, the lightness range gate unit 60 detects the lightness information larger than the threshold value, but it may be configured to detect the lightness information in the predetermined range. Further, in the first embodiment, the saturation range gate unit 8
Although 0 detects the saturation information larger than the threshold value, it may be configured to detect the saturation information in a predetermined range.

[0026]

The image processing apparatus of the present invention has a lightness detecting means for detecting lightness information from color image information and a saturation detecting means for detecting saturation information, and further from the lightness information and the saturation information. , A parameter (output of adders D99, E99 & D103, E of the adder D99, E99) indicating the degree to which the region of interest (processing target pixel updated momentarily) is the background portion of the document
Since the output (103 output) is detected, an appropriate background detection process and / or background removal process can be performed by using the above parameters.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an exemplary embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of an optical scanner unit 10 shown in FIG.

3 is a block diagram showing a configuration of a color background detection unit 20 shown in FIG.

4 is a block diagram showing a configuration of a color background removing unit 30 shown in FIG.

5 shows color image data (input R, G, B data) output by the optical scanner unit 10 shown in FIG. 1 and color image data output by the color background removing unit 30 shown in FIG. 3 is a graph showing the relationship with the data (output R, G, B data).

FIG. 6 is a block diagram showing a configuration of a brightness detection section 50 shown in FIG.

7 is a block diagram showing a configuration of a saturation detection unit 70 shown in FIG.

8 is a block diagram showing a configuration of a lightness range gate unit 60 shown in FIG.

9 is a block diagram showing a configuration of a saturation range gate unit 70 shown in FIG.

10 is a block diagram showing a configuration of a frequency detection unit 100 shown in FIG.

11 is a block diagram showing a configuration of a parameter control unit 90 shown in FIG.

Claims (7)

[Claims] 1. A background detecting means for detecting background information from color image information read by an image reading means, and a background removing means for performing background removing processing on the color image information according to the background information. From the brightness detection means for extracting the brightness information from the color image information, the saturation detection means for extracting the saturation information from the color image information, and the detection result of the brightness detection means and the detection result of the saturation detection means. A detection control for controlling a background detection process in the background detection means based on a detection result of the parameter detection means, the parameter detection means detecting a parameter indicating the degree to which the color image information of interest is a background portion. An image processing device comprising: 2. A background detecting means for detecting background information from color image information read by the image reading means, and a background removing means for performing background removal processing on the color image information according to the background information. From the brightness detection means for extracting the brightness information from the color image information, the saturation detection means for extracting the saturation information from the color image information, and the detection result of the brightness detection means and the detection result of the saturation detection means. A removal control means for controlling the background removal processing in the background removal means based on the detection result of the parameter detection means, which has a parameter detection means for detecting a parameter indicating the degree to which the image information of interest is a background portion. An image processing apparatus comprising: 3. The image processing apparatus according to claim 1, wherein the brightness detection means detects whether the brightness information is larger than a predetermined threshold value and outputs the brightness information. 4. The image processing apparatus according to claim 1, wherein the brightness detecting means detects whether the brightness information is within a predetermined range and outputs the brightness information. 5. The saturation detecting means detects and outputs whether or not the saturation information is smaller than a predetermined threshold value and outputs it.
Alternatively, the image processing apparatus according to claim 2. 6. The image processing apparatus according to claim 1, wherein the saturation detecting means detects whether or not the saturation information is within a predetermined range and outputs it. 7. A storage unit for storing a detection frequency distribution of the lightness information and the saturation information, wherein the control processing in the detection control unit is changed according to the distribution information stored in the storage unit. The image processing apparatus according to claim 1, claim 2, claim 3, claim 4, claim 5, or claim 6.