JP2605806B2 - Image processing device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus such as a digital copying machine that handles a color image as an electric signal.

2. Description of the Related Art In recent years, colorization of copy originals has been progressing in the market, and color digital copying machines have been used for general office work. A color original has various background colors as compared with a black-and-white original, and often includes a photograph or the like. Therefore, the image processing apparatus needs to consider not only the processing corresponding to the color original and the binary processing but also the halftone processing.

Hereinafter, an example of the above-described conventional image processing apparatus will be described with reference to the drawings.

 FIG. 9 is an explanatory diagram of a conventional image processing apparatus.

The black-and-white digital copier detects the density distribution of the scanned black-and-white image original, and determines the quantization level SL from the background part that is the minimum value distribution (Wp) and the image information that is the maximum value distribution (Bp). And binary processing was performed. (For example, Japanese Patent Application Laid-Open No. 61-214867) Problems to be Solved by the Invention However, there is a problem that the background density of a rectangular document arbitrarily arranged cannot be accurately removed.
SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and provides an image processing apparatus that appropriately removes the background of a rectangular document arranged arbitrarily.

Means for Solving the Problems To solve the above problems, an image processing apparatus according to the present invention comprises: an image reading means for sequentially reading a document image by a line sensor; and an image signal from the image reading means at a predetermined determination level. Comparing means for outputting a predetermined judgment value by comparing with the scanning line; area detecting means for detecting the start position and the end position of the original image on the scanning line for each scanning line from the predetermined judgment value; Density detection means for sequentially detecting the background density of the document image from the start end position to the end position of the document image, and detecting the background density level of the document image using a plurality of the background densities detected in each scanning line; Processing means for removing the background density from the image signal of the image reading means based on the background density level detected by the density detection means.

According to the present invention, the start position and the end position of an image area on a scan line of a rectangular document arbitrarily arranged are sequentially detected by the above-described configuration, and the image area from the start position to the end position on the scan line is detected. The background density level is detected using a plurality of background densities detected from each scanning line, and the background density removal processing can be performed in accordance with the background density level. The background density can be removed.

Embodiment Hereinafter, an image processing apparatus according to an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing an image processing apparatus according to an embodiment of the present invention. In FIG. 1, reference numeral 1 denotes a color original image of three colors red (hereinafter simply referred to as R), green (hereinafter simply referred to as G), and blue (hereinafter simply referred to as B).
An image reading device that reads color separation into two, an edge emphasizing unit that corrects blurring of an image or the like with respect to output signals R1, G1, and B1 of the reading device; and 3, an output signal R2, G2 of the edge emphasizing unit ,
A gamma correction unit for performing characteristic correction such as brightness, contrast, and background removal for B2, and converting from linear reflectance to linear density, and 4 outputs yellow of print image data from output signals R3, G3, and B3 of the gamma correction unit. (Hereinafter simply referred to as Y), magenta (hereinafter simply referred to as M), cyan (hereinafter simply referred to as C), and a masking processing section for generating a block (hereinafter simply referred to as K), 8 is an output signal R3, G3, B3
A brightness generating unit for generating a brightness signal L from the masking processing unit; a selector for selecting print data GX from output signals Y, M, C, K and L of the brightness generating unit; A binarization processor for generating binary data from the value data; 16 a printer for printing output data GY of the binarization processor;
Reference numeral 7 denotes an area detection unit for detecting a document area from the output signals R1, G1, and B1 of the image reading apparatus. Reference numeral 9 denotes brightness data L from the brightness generation unit and the area detection unit, and density for detecting document density from the area signal CMPO. A detection unit, 12 is a CPU that performs data processing, 11 is a RAM that stores processing data, and 10 is an RO that stores fixed data.
M and 13 are display units for displaying brightness and the like, 14 is an operation unit for setting processing contents, and 15 is a bus connecting each unit.

The operation of the image processing apparatus configured as described above will be described with reference to FIG. 1, FIG. 2, FIG. 3, FIG. 4, FIG. 5, FIG. Will be described.

First, FIG. 8 (a) is an enlarged view of a part of the image reading apparatus 1, wherein 40 is a platen cover for pressing a color original, 41 is a color original, and 42 is an original using a transparent material such as glass. The color document 41 is placed downward on the table. 43 is a fluorescent lamp, 44 is a SELFOC lens array, 45 is a color sensor, and the light emitted from the light source of the fluorescent lamp 43 is reflected on the color original 41 or the platen cover 40 and passes through the SELFOC lens array 44 to the color sensor. Imaged on 45. The color sensor 45 is composed of, for example, a contact type CCD or a bipolar line sensor, obtains image information by reflected light from the color document 41, and separates the read pixels into R, G, B three-color analog signals. I do. The color-separated analog signal is A / D-converted in a signal processing unit 47 and is replaced with a digital signal. Shading correction is performed to correct the variation between the pixels or chips of the color sensor 41 and to correct the uneven light amount of the fluorescent lamp 43. Further, the three colors R,
The G and B signals are read pixels at the main scanning direction X and the sub-scanning direction Y
Is adjusted to match the output signal R.
Get 1, G1, B1. The color sensor 45 scans in the sub-scanning direction Y from a position before the position detection sensor 46 that detects the document leading edge position PO, and reads the color document 41.

FIG. 8B shows the range of the reading area of the color sensor 45, and the original area EFGH is present in the entire reading range ABCD. Since the platen cover 40 is formed of a specular reflection plate or a black absorbing plate with respect to the color document surface, the amount of light received outside the document area EFGH is small. As a result, the output signal of the image reading device 1
R1, G1, and B1 go low.

FIG. 2A is a block diagram of the area detecting section 7, and FIG. 2B is an explanatory diagram of the block diagram.

Reference numerals 7a to 7c denote comparators, which compare the output signals R1, G1, B1 (A side) of the image reading device 1 with the value of the comparison slice level value CSL (B side) set by the CPU 12 to obtain an input signal A. When the side is larger than the B side, it outputs low level signals "L7d to 7f. If any of the signal levels is" L "by the AND circuit 7i, the output signal 7g becomes" L ".
Since the platen cover 40 uses a mirror plate or a black absorbing plate that reduces reflected light (secondary scattered light) to the color sensor 45 with respect to the color document surface, the color sensor 45 is compared with the inside of the document area EFGH. Receives a small amount of light. As shown in FIG. 8B, outside the document area EFGH, the light received from the platen cover 40 is detected by the color sensor 45, so that the amount of received light is small. Therefore, as shown in FIG. 2B, if the comparative slice level value CSL (B side) set by the CPU 12 is set to a level higher than the level corresponding to the reflected light from the platen cover 40, the document area Only the output signal 7g becomes "L". As shown in FIG. 2B, the position where the signal R1, G1, or B1 larger than the comparison slice level value CSL is first obtained from the falling edge of the LENBL of each line is determined by a reflection object other than the platen cover 40. The first position to read and scan the area where
That is, it is the start position of the document. At the beginning of the document,
This is the first position where the output signal 7g changes from “H” to “L”. In addition, the position where all of the signal R1, the signal G1, and the signal B1 are smaller than the comparison slice level value CSL from the fall of the LENBL of each line at the end is scanned by reading an area other than the platen cover 40 on which a reflection object is placed. This is the last position, that is, the end position of the document. The end position of the document is the last position where the output signal 7g changes from "L" to "H". Therefore, the pixel address (Ast) of the first falling position where the output signal 7g changes from “H” to “L” and the pixel address of the last rising position where the output signal 7g changes from “L” to “H” (Aend) is detected by counting the pixel clock, the original area can be detected. Area generation circuit 7h
Generates an “L” enable signal CMPO from the first fall to the last rise of the signal 7g as an effective area.
While the signal LENBL indicating the pixel effective range is “L”,
The first falling pixel address (Ast) and the last rising pixel address (Aend) of the signal 7g are stored in the pixel clock CLK.
By counting the signal LENB on the next line
When L becomes "L", an enable signal CMPO that makes the pixel address (Ast) to the pixel address (Aend) "L" according to the pixel address is generated. Therefore, the signal CMPO
Is a signal that is always generated with one line delay. The signal 7g becomes "L" if any of the signal levels of the signals R1, G1, and B1 exceeds the comparison value CSL, so that the recognition accuracy for the color original is increased. Although the drawing of this embodiment shows a case where the background is red, when a normal lightness signal L is used, the lightness signal L = (a · R3 + b · G3 + c · B3) / (a + b + c) (1−1) When equation 1) is used, R3 signal component = a.R3 / (a + b + c) (1-2). As a result, the signal component level of R3 decreases. Also,
Since the output signals R2, G2, and B2 of the edge emphasizing unit 2 have their images enhanced by a Laplacian filter or the like, the output signals R2, G2, and B2 become signals having peaks in the case of a document with a coarse background, and When the detection sensitivity is increased by lowering the level, a malfunction may occur. Output signals R3, G3, B3 of the gamma correction unit 3
Is CP for brightness correction, contrast correction, etc. via bus 15.
Since it is set from U12, the level setting of the signal CSL becomes complicated, and the signal CSL is raised at the time of brightness adjustment, etc.
As a result, the detection sensitivity may decrease. Therefore, the input data R1, G1, and B1 of the area detection unit 9 are color component signals R, G, which are in a stage before the edge emphasis 2 and the gamma correction unit 3, that is, a stage in which characteristic correction such as edge emphasis, brightness correction, and contrast correction is not performed. G,
By extracting from B, highly accurate region detection can be performed.

FIG. 3 is a block diagram of the density detector 9. When performing density detection, the edge enhancement unit 2 does not perform enhancement processing.
Signal R1 = signal R2 and signal G1 = G by SW1 in FIG. 5 (b).
2. The signal B1 = B2.

From FIG. 1, the signals R1, G1,
B1 is converted into signals R2 (= R1) and G2 (=
G1) and B2 (= G1).

Further, the gamma correction unit 3 receives the signals R2, G2, and B2 and outputs the signals R3, G3, and B3. The gamma correction unit 3 performs logarithmic correction based on the conversion curve of Y = −LOG (X). Therefore, the gamma correction unit 3 performs conversion so that the larger the input signal, the smaller the output signal. The brightness generator 8 uses the signal R3, the signal G3, and the signal B3,
The brightness signal L is generated according to the equation (1-1).

In the area of the platen cover 40 where the amount of reflected light is small, the amount of light received by the color sensor 45 is small. Therefore, the brightness signal L indicates a large value, and in the area of the color document 41 where the amount of reflected light is large (the background area of the document), The lightness signal L shows a small value because the light receiving amount of the light-emitting element becomes large. By sampling the minimum value of the brightness signal L from the start position to the end position of the document on each line, the background density of the document can be detected.

The brightness data L output from the brightness generation unit 8 is latched by a latch 9a.
Is stored. The stored signal QA is the A of the comparator 9b.
Side and the input D of the latch 9c. The output Q of the latch 9c is preset to FF [HEX] while the signal LENBL is at the high level “H”. The condition that the signal QA is latched in the comparator 9c is when a rising pulse is output from the AND circuit 9e. The AND circuit 9c outputs the pixel clock CLK when the signal CMPO is “L” and the output signal QC of the comparator 9b is “H”.
Is valid. Comparator 9b outputs latch 9c to B side
QB is connected, compares the input signal QA and the comparison signal QB on a pixel-by-pixel basis, and outputs an output signal QC. Pixel effective range signal LENB
When L becomes “L”, that is, the first pixel Q in the main scanning direction X
When A is input, the signal QB is FF [HEX]. If the signal QA <the signal QB, the signal QC becomes “H”, the signal QA is latched by the latch 9c, and the signal QB is updated to the data value of the signal QA. It is not updated when signal QA> signal QB or signal QA = signal QB. This procedure is repeatedly performed in a section where the pixel effective range signal LENBL is “L”. As a result, when the final pixel data QA ends, that is, immediately before the signal LENBL rises, the minimum value of the one-line pixel data in the main scanning direction X is stored in the latch 9c. The minimum value QB is latched by the latch 9d at the rise of the signal LENBL, and the signal S
Synchronize with the YNC interrupt from the data bus 15 via the CPU 12
Stored in the RAM 11. Therefore, the minimum value of the brightness data L is sampled for each line.

Data FF [HEX] is preset in the latch 9d outside the document area, that is, when the signal LENBL is "H" or the signal CMPO is "H".

FIG. 5 (a) is an explanatory diagram of processing of the detected brightness data L, and FIG. 5 (b) is a block diagram of an edge emphasizing unit.

The brightness data L of the brightness invention unit 8 is configured by a LUT (look-up table), and in this embodiment, a ROM is used.
In the equation (1-1), constants such that a = 1, b = 2, and c = 1 are set. The signals R3, G3, and B3 are input to the ROM address, and data that stores the operation results is output. The ROM can be configured by hardware using an adder. Further, the above constants can be changed and can be set with this configuration.

When the brightness detection unit 9 samples the brightness data L, the switch SW1 of the edge enhancement unit 2 is switched to the B side. This is because the edge enhancement 2a enhances the image with a Laplacian filter or the like, so that the output signals R2, G2,
This is because B2 is differentiated to generate a peaked signal. The brightness data L for the output signal having this peak
Is sampled, a peak density distribution is generated below the actual original background level. When the density distribution is used, there occurs a problem that it is not possible to perform accurate background removal. Therefore, accurate sampling can be performed by using a signal that is not emphasized. During normal copying without sampling, SW1 is set to the A side to perform edge emphasis.

The minimum value of the brightness data L is sampled for each line from the density detector 9 and stored in the RAM 11 in the form of a table. The brightness table is prepared for the data range of the brightness data L value to be detected, and all data values are initialized to 00 [HEX] before performing density detection. In the present embodiment, since the brightness data L changes from 00 [HEX] to FF [HEX], 256
What is necessary is just to prepare the staple of a step. Every time the signal SYNC is interrupted, the CPU 12 increases the storage of the address corresponding to the brightness data L in the table by +1. The byte configuration of one step is determined by the resolution of the image reading device 1 and the printer 16. With a 2-byte configuration, the resolution is 16 lines / m
As m, 420 mm can be sampled in the sub-scanning direction Y, which corresponds to an A3 original. Data FF in table
Since [HEX] contains data other than the manuscript area,
Not used as processing data. The minimum value sampling of the brightness data L is performed up to the reading area in the sub-scanning direction Y. When the reading area is completed, the background density distribution of the read color document 41 is generated in the brightness table.

Brightness table is d when brightness data L = 00 [HEX]
(0), when d = 255 when L = FF [HEX], each of d (0) to d (25) at the time when reading in the sub-scanning direction Y is completed.
The distributions of the detection frequencies f (0) to f (255) up to 5) are generated, for example, as shown in FIG. 5 (a). The frequency of the detected density values of d1 to d4 is the background density distribution of the document. The average density D (Ave.) is calculated as follows: Average density D (Ave.) = ｛F (1) · d (1) + f (2) · d (2) +... + F (254) · d (254) {/ {F (1) + f (2) +... + F (254)} (1-4) Here, since f (255) includes data other than the document area, it is not used as processing data.

FIG. 6 (a) is a diagram showing the brightness characteristics, FIG. 6 (b) is a configuration diagram thereof, and FIG. 7 is a configuration diagram of the operation unit 14.

The brightness adjustment is performed by the CPU 12 to the bus 1 for the gamma correction unit 3.
Set via 5. In this embodiment, brightness adjustment, contrast adjustment, background removal, and gamma correction are configured by a LUT (look-up table), with 3 bits for brightness characteristics, 2 bits for contrast adjustment, 3 bits for background removal,
Eight bits are used for each R2, G2, B2 input. LUT is R
OM etc., independently configured for R2, G2, B2,
Three bits for brightness characteristics, two bits for contrast adjustment, and three bits for background removal are commonly input. There are seven levels of brightness adjustment, three levels of contrast adjustment, and eight levels of background removal, and operation buttons 14a to 14b are used for brightness adjustment, contrast adjustment, and background removal.

Contrast adjustment depends on the set brightness adjustment characteristics.
This is performed by changing the inclination. Contrast operation button
Each time 14d is pressed, the high-contrast display 13h, the normal display 13i, and the low-contrast display 13j are sequentially turned on, and setting is performed. By performing the brightness adjustment and the contrast adjustment independently, a desired image can be obtained.

The brightness is adjusted in seven stages by pressing the brightness adjustment buttons 14a and 14b. The operation button 14a moves the lighting of the display 13a to 13g in the dark direction and the light of the display 13a to 13g moves in the light direction, and the brightness characteristic of the number of the lighting is set. Number 1
If the lightness characteristics corresponding to F1 to F7 are F1 to F7, the characteristics shown in FIG. 6A are obtained. As preprocessing for brightness adjustment, the input signals R2, G2, and B2 are subjected to logarithmic correction such that Y = -log (X), and gradation correction is performed so that the maximum density value Dmax becomes FF [HEX]. . When the binary processing section 6a of the binarization processing section 6 performs the binary slice level at a certain fixed level SL, the output signals R3, G3, B3 are binarized at the levels of C1 to C7. Accordingly, in the present embodiment, the binary slice level SL is fixed without being varied, and the brightness adjustment is obtained by equivalently varying the binary slice level SL by adjusting the brightness. In the halftone processing section 6a, the brightness changes according to the brightness characteristics F1 to F7, and the brightness adjustment is performed. When the automatic operation is not performed, the background is not removed in this embodiment.

When the automatic operation button 14c is pressed and the automatic operation is performed, the CPU 12 sets the brightness characteristic to F4, the contrast to normal, and the background removal level to D8 without removal, and then scans the color original 41 with the image reading apparatus 1. Then, the density detector 9 detects the average density D (Ave.). The detected average density D (Ave.) is stored in a comparison value table stored in the ROM 10 by the CPU 12 shown in FIG.
Compared with T1 to T7. Further, according to the result of the comparison, one of the lightness characteristics F1 to F7 shown in FIG.
The CPU 12 determines and sets one of the lightness characteristics F1 to F7 by setting the background removal bit of the LUT shown in FIG. 6B. When the binarized slice level of the binary processing unit 6a shown in FIG. 4 is set to a fixed level SL as shown in FIG. 6A, the brightness characteristics F1 to F7 correspond to C1 to C7. The original densities D1 to D7 are compared and binarized.
As a result, when the lightness characteristic F7 is selected, the original density equal to or less than the original density D7 corresponding to C7 is removed. From brightness characteristic F1
One of the selection conditions of F7 is selected as follows: F1: D (Ave.)> T7 F2: T7> D (Ave.)> T6 F3: T6> D (Ave.)> T5 F4: T5> D ( Ave.)> T4 F5: T4> D (Ave.)> T3 F6: T3> D (Ave.)> T2 F7: T1> D (Ave.) F1 to F7: Lightness characteristics T1 to T7: Comparison table , T1 to T7 are empirically determined constants. Said
T1 to T7 are constants set on the assumption that the slice level SL is binarized by C1 to C7. According to the lightness characteristics selected by the comparison, the corresponding numbers are lit on the display units 13a to 13g, and the setting of the lightness characteristics is completed. If the binarization is performed by the binary processing unit 6b in accordance with this setting, the image information can be copied more faithfully without outputting the unnecessary background of the color document 41 to the printer 16. If it is desired to further change the brightness, the brightness can be reset by using the operation button 14a or the operation button 14b. In this case, the display units 13a to 13g move from the current setting position, and resetting is performed. The same brightness characteristic is used whether the automatic brightness adjustment is performed by pressing the auto button or the brightness adjustment is performed manually, so that the processing is not extremely different between the automatic brightness adjustment and the manual brightness adjustment. As a result, the automatic brightness adjustment and the manual brightness adjustment can be shifted and operated without discomfort.

In this embodiment, eight levels of background removal density D1 to D8 are prepared, and D8 is not removed. D1 to D7 are set corresponding to the input levels C1 to C7 for performing the binary processing. Normally D8 is set and the input level is not removed,
In the automatic operation, the CPU 12 automatically sets the background removal bit except when sampling the brightness data. Although the background removal bit is 3 bits in this embodiment,
It is naturally possible with this configuration to increase the number of bits and set a more accurate background removal level. In this case, the range and accuracy of the change of the average density D (Ave.) can be extended at the maximum. The maximum setting is the one-to-one correspondence between the average density D (Ave.) and the background removal level. By automatically setting the background removal level, even when multi-value data is processed by the halftone processing unit 6a, binarization processing can be performed while removing unnecessary background. As a result, the image information can be copied more faithfully without outputting the unnecessary base of the color document 41 to the printer 16.

 FIG. 4 is a block diagram of the binarization processing section 6.

Reference numeral 6a denotes a halftone processing unit that generates binary pseudo halftone data from the multivalued data, and 6b denotes a binary processing unit that binarizes the multivalued data at a certain slice level. As the halftone processing unit, a dither processing or an error diffusion processing unit is used. Reference numeral 6c denotes a selector for selecting the binary output of the halftone processing section and the binary processing section. The setting parameters of the halftone processing unit and the binary processing unit are set by the CPU 12 via the data bus 15 and the contents of the ROM 10. In the present embodiment, the slice level of the binary processing unit is set to a fixed value.

In the automatic operation place, the color document 41 is scanned once, and the density detecting unit 9 generates the background agricultural degree distribution of the color document 41 and performs data processing. The setting of the gamma correction unit 3 is performed based on the data processing, and a copying operation is performed.

The masking processing section 4 outputs the output signal of the gamma correction section 3
Perform color correction etc. from R3, G3, B3 and print image data Y, M, C, K
Are generated simultaneously.

Since the printer 16 can process only one of the colors Y, M, C, and K, the sensor 45 returns to the position before the sensor 46 after reading the area of the color document 41, and performs a plurality of reciprocating operations. Will do. The selector 5 selects one of the print image data of Y, M, C, K and brightness data L generated by one scan, outputs the data GX to the binarization processing unit 6, and performs the binarization processing. The unit 6 performs binarization processing and outputs data GY to the printer 16. The printer 16 uses, for example, a color laser beam printer by an electrophotographic process, and in color copying, print image data sent each time a color document 41 is read is converted into, for example, Y, M, C, K It is formed on the photosensitive drum in order, and is developed by the developing units Y, M, C, and K, so that multi-color superposition is performed to reproduce a color image. In monochromatic copying, an image is formed on a photosensitive drum using print image data of lightness data L a plurality of times, and developing devices Y, M, C,
Since each is developed by K, multi-color superposition is performed to reproduce a color monochromatic image.

As described above, according to this embodiment, the color original image is
An image reading device that separates colors into colors R, G, and B; and an area detection unit that detects the color original image area from the output {{R, G, B} of the three colors R, G, and B of the image reading device. And a density detector for detecting the background density before the color original image is emphasized based on the area output of the area detector, so that accurate density detection is performed for the color original image area. be able to. Also, since the image reading device has a processing unit for varying the background removal level based on the density level of the density detection unit based on the density level of the density detection unit, unnecessary background portions are subjected to binary processing and halftone processing. Can also be removed. Furthermore, a correction unit having a plurality of brightness characteristics for the image signal of the image reading device, and a processing unit for binarizing the output of the correction unit is provided. Since the binarization level is equivalent to the binarized image density, the same characteristic correction output can be used for the binary processing and the halftone processing.

In the present embodiment, the binary slice level of the binary processing unit 6b is set to SL as a constant. However, the binary slice level may be changed in conjunction with the brightness adjustment buttons 14a and 14b during manual operation. Is obtained.

As described above, the present invention sequentially detects the start position and the end position of an image area on a scan line of a rectangular document arbitrarily arranged, and detects the image area from the start position to the end position on the scan line. , A background density level is detected using a plurality of background densities detected from each scanning line, and a background density removal process can be performed in accordance with the background density level. An effect is obtained that the background density of the document can be removed.

[Brief description of the drawings]

FIG. 1 is a block diagram of an image processing apparatus according to an embodiment of the present invention, FIGS. 2 (a) and 2 (b) are block diagrams and explanatory diagrams of an area detecting unit, FIG. 3 is a block diagram of a density detecting unit, FIG. 4 is a block diagram of a binarization processing unit, and FIG.
6B is an explanatory diagram of a density detecting unit and a block diagram of an edge emphasizing unit, FIGS. 6A and 6B are explanatory diagrams and a configuration diagram of a lightness characteristic, FIG. 7 is a configuration diagram of an operation unit, FIG. Figures (a) and (b)
FIG. 9 is an enlarged view of a part of the image reading apparatus and an explanatory view, and FIG.
FIG. 10 is an explanatory view of a conventional example. 2... Edge emphasis section, 3... Gamma correction section, 7... Area detection section, 8... Lightness generation section, 9... Density detection section, 6.
Binarization processing unit, 13 display unit, 14 operation unit.

Claims (2)

(57) [Claims] An image reading means for sequentially reading a document image by a line sensor; a comparing means for comparing an image signal from the image reading means with a predetermined judgment level and outputting a predetermined judgment value; Area detection means for detecting a starting end position and an ending position of the original image on a read scanning line from the predetermined determination value; and sequentially determining a background density of the original image from the starting end position to the ending position on a scanning line. Density detecting means for detecting and detecting a background density distribution of the original image using a plurality of the background densities detected in each scanning line; and the image reading means based on the background density distribution detected by the density detecting means. And a processing unit for removing a predetermined density from the image signal. 2. The image processing apparatus according to claim 1, wherein said comparing means receives a signal from the image reading means.
The image signal separated into colors is compared with the predetermined determination level, and a predetermined determination value is output when any of the color signals satisfies the predetermined determination level condition. 2. The image processing device according to 1.