JP7155706B2 - Image reading device and image forming device - Google Patents

Description

The present invention relates to an image reading device and an image forming device, and more particularly to technology for speeding up document free erase processing.

Image reading devices that read documents and generate image data either stand alone or installed in an image forming apparatus include a sheet-through method in which documents are conveyed one by one on a slit glass while being read, and a device placed on a platen glass. A reading method such as a platen set method for reading a placed document is known.

When scanning a document using the platen setting method, there are cases where the document cover is used to cover the document from above, and cases where the document is not covered. . For example, as shown in FIG. 22, a case where a book 2202 is pressed against a platen glass 2203 with the document cover 2201 open and the page is read corresponds to sky shot.

An image reader reads a document by illuminating the surface of the document to be read from below the platen glass and detecting the amount of light reflected from the surface to be read. The illumination light is diffused as it is without being reflected. Therefore, only weak external light is detected outside the surface to be read, and the surface becomes dark. Focusing on such characteristics, a technique of determining a dark portion as the outside of the document and correcting the outer portion to white to generate image data is called a document free erase function.

However, when the peripheral portion of the reading target surface of the document is dark, it becomes difficult to distinguish between the inside of the document and the outside of the document.

Some models of image reading apparatuses having a low reading speed have a long processing (light receiving) time per scanning line, and thus the amount of light emitted from the light source is suppressed in order to reduce the cost. In such a low-speed model, the time for receiving external light is long, but the amount of light from the light source is suppressed, so the difference in the amount of light received between the inside and outside of the document becomes small, making it difficult to distinguish between the inside and outside of the document.

Furthermore, in models that use a contact-type CIS (Contact Image Sensor), since the light-receiving sensor is in close contact with the platen glass, external light is likely to enter the light-receiving sensor, which tends to reduce identification accuracy.

To solve this problem, for example, by detecting external light with the illumination from below the platen glass turned off, mask data for masking the outside light is generated, and the illumination from below the platen glass is turned on. Techniques have been proposed in which the mask data is applied to image data generated by reading a document in a state (see Patent Documents 1 and 2).

There is also proposed a technique for determining the size of a document by comparing image data generated by reading the document with both the platen cover open and closed (see Patent Document 3).

By adopting these techniques, it is possible to accurately distinguish between the surface of the document to be read and the outside thereof, and correct only the outside to be white.

JP 2007-028413 A JP 2006-313967 A JP-A-11-075025

However, when the above conventional technology is adopted, reading must be performed twice, with the illumination from below the platen glass turned off and on, or with the platen cover open and closed. It doubles the reading time of the document. For this reason, especially in low-speed models, user convenience is inevitably reduced.

Also, if the document is read using separate sensors when the illumination from below the platen glass is turned off and when it is turned on, it is possible to avoid doubling the reading time, but the sensor and its peripheral circuits are required. Since two sets are required, an increase in part cost and an increase in size cannot be avoided.

SUMMARY OF THE INVENTION The present invention has been made in view of the problems described above, and provides an image reading apparatus and an image forming apparatus that improve identification accuracy in document-free erasing without increasing reading time or part costs. intended to

In order to achieve the above object, an image reading apparatus according to the present invention includes a translucent document placing table on which a document is placed, an openable and closable cover covering the document placing table, and a document placed on the document placing table. a light source for irradiating a document with the cover open, and an amount of light emitted from the light source in units of scanning lines according to a predetermined switching pattern along the sub-scanning direction when reading the document with the cover open. document reading means for detecting the amount of incident light including the amount of reflected light from the document for each pixel; and correction means for correcting the detected value of the incident light amount in accordance with the switching pattern of the amount of irradiation light. and, as a result of the correction by the correcting means, a group of pixels that are continuous in the sub-scanning direction and in which the detected values after the correction vary along the sub-scanning direction in accordance with the switching pattern of the irradiation light amount. edge pixel detection means for detecting pixels included in the pixel group as edge pixels when there is a boundary; characterized by

In this case, a low-brightness pixel detecting means for detecting low-brightness pixels whose reflected light amount is smaller than a predetermined light amount is provided, and the boundary specifying means detects edge pixels, if they are not low-brightness pixels, outside the document. Instead, it is desirable to specify the boundaries of the document to be internal.

Further, it is more preferable if the correcting means also serves as shading correction.

Further, the image reading apparatus according to the present invention includes a document placing table on which a document is placed, a cover that can be opened and closed to cover the document placing table, and a light source that irradiates the document placed on the document placing table with light. light amount control means for switching the irradiation light amount of the light from the light source in units of scanning lines according to a predetermined switching pattern along the sub-scanning direction when reading the document with the cover open; document reading means for detecting the amount of reflected light for each pixel; edge pixel detection means for detecting edge pixels whose detected value of the amount of reflected light corresponds to the switching pattern of the amount of irradiation light; and pixels other than edge pixels outside the document. and boundary identifying means for identifying the boundary of the document such that:

In this case, low-lightness pixel detection means for detecting low-lightness pixels whose amount of reflected light is less than a predetermined amount of light is provided, and the boundary specifying means further detects pixels of the document if they are not low-lightness pixels, even if they are not edge pixels. It is desirable to specify the boundaries of the document to be internal.

In either case, the light amount control means may periodically switch the irradiation light amount along the sub-scanning direction.

Further, the light amount control means may switch between two types of irradiation light amounts, large and small, for each scanning line.

Further, smoothing means for smoothing the detected value of the amount of reflected light along the main scanning direction may be provided, and the edge pixel detecting means may detect the edge pixel using the smoothed detected value.

Further, the light source may be composed of a plurality of partial light sources having different light colors, and the light amount control means may switch the irradiation light amount only for some of the plurality of partial light sources.

Further, the edge pixel detection means detects pixels by determining whether the difference in the detected amount of reflected light between pixels adjacent to each other in the sub-scanning direction is within a predetermined range. good too.

Further, the predetermined range may be designated by a value corresponding to the switching pattern of the irradiation light amount.

Further, an image forming apparatus according to the present invention is characterized by comprising the image reading apparatus according to the present invention. In this way, when processing such as copying or facsimile transmission involves sky shots, it is possible to increase the speed and accuracy of the processing.

In this way, even when scanning a document with a high density at the periphery by Sky Shot under conditions where a large amount of unnecessary light is incident on the outside of the document, the boundary between the inside and outside of the document can be accurately detected in one reading. can be well identified. It is also advantageous in that it does not involve a significant increase in cost or decrease in productivity due to the addition of parts.

Manuscripts with dark edges are relatively common in magazines and books. However, the probability of existence of a document having a density level difference that is approximately equal to the amount of light amount switching of the image forming apparatus for each line is almost zero, and it is difficult to place the document on the platen glass while maintaining the parallelism accurately. is. Therefore, it is believed that the document boundary discrimination capability of the present invention is very high.

1 is an external perspective view showing a configuration of a multifunction machine 1 according to a first embodiment of the invention; FIG. 4 is an external perspective view showing a state in which the document cover 102 is opened; FIG. 3A is a diagram showing the main configuration of the main body 101 of the image reading unit 100, and FIG. 4B is a diagram showing the main configuration of the reading unit 301. FIG. 2 is a block diagram showing the main configuration of a control unit 203; FIG. 9 is a flow chart showing processing for determining a shading correction coefficient; 4 is a timing chart illustrating main scanning synchronization signals HSYNC and lighting signals (RGB) when determining shading correction coefficients; 4 is a flowchart showing document reading processing; 10 is a flow chart showing reading processing for original free erase. 4 is a timing chart exemplifying a main scanning synchronization signal HSYNC and lighting signals (RGB) during reading processing for original free erase; (a) illustrates an image before shading correction processing for document-free erasure, (b) illustrates an image after shading correction processing for document-free erasure, and (c) illustrates the brightness on the EE line in (b). (d) shows the brightness value on the FF line in (b). 9 is a flowchart showing shading correction processing for original free erase. 4 is a flowchart showing brightness generation processing; 5 is a flowchart showing main scanning direction smoothing processing; 4 is a flowchart showing sub-scanning direction brightness difference processing. 4 is a flowchart showing brightness determination processing; 4 is a flowchart showing boundary coordinate determination processing; 4 is a flowchart showing erase processing; 9 is a flowchart showing shading correction processing according to the second embodiment of the present invention; 9 is a flowchart showing boundary coordinate determination processing according to the second embodiment of the present invention; 9 is a timing chart illustrating a main scanning synchronization signal HSYNC and lighting signals (RGB) according to a modification of the invention; FIG. 11 is an external perspective view showing the configuration of a multifunction machine 1 according to a modified example of the present invention; It is an external appearance perspective view explaining a sky shot.

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of an image reading apparatus according to the present invention will be described below with reference to the drawings.
[1] First Embodiment An image reading device according to the present embodiment is installed in a multifunction peripheral (MFP: Multi-Function Peripheral).
(1-1) Configuration of MFP As shown in FIG. 1, the MFP 1 includes an image reading unit 100, an image forming unit 110, and a paper feeding unit 120, and has a print function, a copy function, a fax function, and a scan function. It has multiple functions such as functions. The image reading section 100 is composed of a body section 101 and a document cover 102, and reads a document to generate image data.

The image forming unit 110 receives image data generated by the image reading unit 100 and image data obtained from a PC (Personal Computer) or other device via a facsimile line or an IP (Internet Protocol) communication network such as a LAN (Local Area Network). An image forming process is executed based on the image data. In this image forming process, the image forming section 110 receives a recording sheet from the paper feeding section 120, transfers the toner image onto the recording sheet, and thermally fixes the toner image. The recording sheet on which the toner image has been fixed is discharged onto the discharge tray 111 .

An operation panel 112 having a plurality of keys (buttons) is provided on the front side (front side) of the image forming section 110 , and a function selected from various functions of the multifunction machine 1 can be operated. The image forming unit 110 can also transmit image data and the like to another device via a facsimile line or an IP communication network.

The paper feeding section 120 is arranged below the image forming section 110 and accommodates a bundle of recording sheets. The paper feeding unit 120 supplies recording sheets one by one in accordance with the image forming process of the image forming unit 110 .
(1-2) Configuration of Image Reading Unit 100 As shown in FIG. can be opened and closed in the direction of arrow A. A platen glass 201 is provided on the upper surface of the main body 101, and a document is placed on the platen glass 201 when reading an image.

The main body 101 also includes an open/close sensor that detects the open/closed state of the document cover 102 depending on whether or not the pin 202 is pressed by the document cover 102 . A white reference plate 211 is arranged at a position facing the platen glass 201 with the document cover 102 closed. The white reference plate 211 is uniformly white on its entire surface, and is used for determining shading correction coefficients used for shading correction, as will be described later.

Furthermore, the body portion 101 incorporates a control portion 203 that controls the operation of the body portion 101 .

As shown in FIG. 3A, the main body 101 incorporates a reading unit 301 elongated in the main scanning direction, a traveling unit 302 and a guide rail 303 . The reading unit 301 is attached to a traveling unit 302, and the traveling unit 302 travels along the guide rails 303 from a home position 310 to a travel completion position 311, thereby reading an image from a document while moving in the sub-scanning direction. .

As shown in FIG. 3(b), the reading unit 301 is a so-called 1-line CIS (Contact Image Sensor), the image reading sensor 314 is only a single row of monochrome line sensors, and light sources 311R and 311G for RGB colors are used. , and 311B are sequentially switched and turned on to obtain line data of each color of RGB. Since there is no mechanism for separating wavelengths when acquiring line data, RGB are temporally separated on the assumption that light having the same wavelength as that of a lighted light source is received.

The light sources 311R, 311G, and 311B are arranged at one or both ends in the main scanning direction of the light guide 312 elongated in the main scanning direction, and emit reading light for illuminating the surface of the document D to be read. It exits toward the end. The light guide 312 diffuses the reading light emitted from the light sources 311R, 311G and 311B substantially uniformly in the main scanning direction. As a result, the reading surface of the document D is illuminated over the entire width in the main scanning direction.

Reflected light from the surface to be read enters the image reading sensor 314 via the lens array 313 . The lens array 313 focuses reflected light from the surface to be read onto the image reading sensor 314 . In this embodiment, SLA (SELFOC Lens Array. SELFOC is a registered trademark of Nippon Sheet Glass Co., Ltd.) is used as the lens array 313. However, other optical systems such as MLA (Micro Lens Array) may be used. good.

The image reading sensor 314 is composed of a plurality of photoelectric conversion elements arranged along the main scanning direction, and outputs an output value corresponding to the amount of incident light (the product of the luminance of the incident light and the incident time). That is, the image reading sensor 314 converts an optical reflected image generated on the object to be read into an electrical image signal for each scanning line and outputs the image signal.

A configuration other than the one-line CIS may be used as the reading unit 301 . Further, the image reading sensor is fixed inside the main body 101, and the light source and the mirror are moved in the sub-scanning direction along the surface to be read, and the reflected light from the surface to be read is made incident on the image reading sensor. A mirror scan method for generating an image signal may be employed.

The body portion 101 incorporates a control portion (not shown). This controller controls the operations of the reading unit 301 and the traveling unit 302, and also determines shading correction coefficients to be used for shading correction using the image signal output from the image reading sensor 314, and executes shading correction.
(1-3) Configuration of Control Unit Next, the configuration of the control unit built in the main unit 101 will be described.

As shown in FIG. 4 , the control unit 203 includes a CPU (Central Processing Unit) 401 , ROM (Read Only Memory) 402 , RAM (Random Access Memory) 403 and HDD (Hard Disk Drive) 404 . When the MFP 1 is powered on, the CPU 401 reads a boot program from the ROM 402 to start up, and uses the RAM 403 as a working storage area to execute an OS (Operating System) and control programs read from the HDD 404 .

A clock circuit 405 generates a main scanning synchronization signal HSYNC and inputs it to the CPU 401 . A main scanning synchronization signal HSYNC is a signal for synchronizing one scanning line. The main scanning synchronization signal HSYNC according to the present embodiment turns on the light sources in order of R→G→B. CPU 401 receives a detection signal from open/close sensor 411 that detects opening/closing of document cover 102 using pin 202 .

The CPU 401 inputs lighting signals (RGB) to the light sources 311R, 311G and 311B in synchronization with the main scanning synchronization signal HSYNC. Each of the lighting signals (RGB) is a signal that takes two states of H for turning on the light source and L for turning off the light source. , simply referred to as “lighting period”), the light continues to be lit at a predetermined luminance. The CPU 401 refers to the main scanning synchronization signal HSYNC output by the clock circuit 405 to determine the timing of turning on the light sources 311R, 311G and 311B.

Upon receiving an image signal from the image reading sensor 314 , the CPU 401 corrects the image signal as described later, and then outputs the corrected image signal to the image forming unit 110 . Further, CPU 401 receives operation signals related to user operations from operation panel 112 provided in image forming unit 110 . Furthermore, the CPU 401 outputs control signals to control the operation of the traveling unit 302 .
(1-4) Processing for Calculating Shading Correction Coefficient Next, processing for calculating a shading correction coefficient will be described.

When calculating the shading correction coefficient, as shown in FIG. 5, the control unit 203 first refers to the detection signal output from the open/close sensor 411, and if the document cover 102 is closed (S501: YES). ), the traveling unit 302 starts traveling in the sub-scanning direction (S502). Next, the control unit 203 executes the reading process from step S503 to step S507 for each scanning line.

As shown in FIG. 6, the main scanning synchronization signal HSYNC output by the clock circuit 405 sequentially generates pulses corresponding to RGB colors. All the pulses have the same pulse width, and therefore the amounts of light emitted by the light sources 311R, 311G and 311B for each scanning line are equal to each other.

When the CPU 401 detects the light source lighting timing by detecting the rising edge of the pulse of the main scanning synchronization signal HSYNC (S504: YES), the CPU 401 turns on the light source of the light color corresponding to the pulse among the light sources 311R, 311G and 311B. (S505), and the line sensor 314 is used to detect the amount of reflected light from the white reference plate 211 for each pixel (S506).

Thereafter, shading correction coefficients are calculated for each light color for all pixels (shading correction coefficient calculation processing: from S508 to step S510). The shading correction coefficient is calculated by dividing the reference value by the read value of the amount of reflected light detected for each pixel for all pixels and all light colors.

Shading correction coefficient = standard value / read value of reflected light amount (1)
When the document is read with the document cover 102 closed, the shading correction is performed by reading the document with the same amount of light emission as when determining the correction value. In this case, the pixel value before shading correction is multiplied by the shading correction coefficient of the pixel for each pixel.

Pixel value after correction = shading correction coefficient x pixel value before correction (2)
Note that the reference value used to calculate the shading correction coefficient may be different for each light color in consideration of color balance. Further, when the same reference value is used for each light color, processing for adjusting the color balance may be performed on the corrected pixel value calculated using the above equation (2).
(1-5) Document-Free Erase Processing Next, the document-free erase processing executed when the document is read with the document cover 102 open will be described.

As shown in FIG. 7, when the image reading unit 100 receives an instruction to read a document from the operation panel 112 from the user of the multifunction machine 1 (S701: YES), the image reading unit 100 refers to the detection signal from the open/close sensor 411 and reads the document cover. 102 is closed (S702: NO), normal reading processing is executed (S711). On the other hand, if the document cover 102 is open (S702: YES), reading processing for document free erase is executed (S703).

In the document free erase reading process, the light emission amounts of the light sources 311R, 311G and 311B are switched for each scanning line to read the document and generate color image data (hereinafter referred to as "read image"). Next, shading correction processing for original free erase is applied to the read image (S704). In this document free erase shading correction process, the correction coefficients are switched according to the light emission amounts of the light sources 311R, 311G and 311B for each scanning line.

In the brightness generation process, single-color brightness image data (hereinafter referred to as "brightness image") is created from three-color RGB image data (S705). The smoothing process in the main scanning direction is a process of carrying out a weighted moving average of brightness for each scanning line in a brightness image (S706).

The sub-scanning direction brightness difference detection process determines whether or not the brightness difference between three pixels adjacent to each other in the sub-scanning direction is within a predetermined range. is a process of detecting pixels belonging to a striped pattern area (S707). The lightness determination process is a process of comparing the lightness of each pixel with a predetermined threshold to detect pixels with low lightness (S708).

The boundary coordinate specifying process is a process of specifying boundary coordinates inside and outside the document using the processing results of the sub-scanning direction brightness difference detection process and the brightness determination process (S709). The erase process is a process of whitening the outside of the boundary coordinates (S710).
(1-5-1) Reading process for original free erase (S703)
In the document free erase reading process, as shown in FIG. 8, first, the traveling unit 302 starts traveling from the home position 310 in the sub-scanning direction (S801). The traveling unit 302 travels from the home position 310 to the travel completion position 311 at a predetermined speed determined by the pulse rising interval of the main scanning synchronization signal HSYNC and the resolution of the reading device 100 .

After that, the processing from step S802 to step S808 is repeated until the travel unit 302 reaches the travel completion position 311 . That is, the main scanning synchronizing signal HSYNC output by the clock circuit 405 is monitored, and each time a rise of the pulse is detected (S803: YES), the lighting signal (RGB) for lighting the RGB light sources is sequentially set to H. (S804).

FIG. 9 is a timing chart illustrating main scanning synchronization signal HSYNC and lighting signals (RGB). As shown in FIG. 9, every time the pulse of the main scanning synchronization signal HSYNC rises, the lighting signals (RGB) of the light colors RGB are turned to H in the order of RGB. Further, the lighting period of the lighting signal (RGB) repeats a long (Tl) and a short (Ts) for each scanning line. Also, the lighting period is the same between the lighting signals (RGB) for each scanning line.

The lighting period of the lighting signal, in other words, the lighting period of the light source is alternately switched between Tl and Ts for each scanning line. Since the lighting period Tl is longer than the lighting period Ts, the amount of light emitted from the light source is large. After the lighting period of the light source has elapsed, the amount of reflected light for one scanning line from the reading target surface of the document is read out for each pixel from the line sensor 314 (S805).

In this embodiment, the ratio of Ts to T1 is about several tens of percent. If the ratio of Ts to Tl is too large, it becomes difficult to detect the boundary between the document and the outside of the document. On the other hand, if the ratio of Ts to Tl is too small, the amount of light will be too small and the reading accuracy of the document may be lowered. Therefore, it is desirable to set an appropriate time length as Ts.

After that, when reading of three colors of RGB of one scanning line is completed (S806: YES), the light emission amount is switched (S807). Specifically, if the current lighting period is long (Tl), it is shortened (Ts), and if it is short (Ts), it is lengthened (Tl).

As described above, the reading unit 301 is a 1-line CIS, and the 1-line CIS generally reads each color continuously without stopping the movement of the original or the reading unit when reading each dot. Also, the rise timing of the pulse of the lighting signal (RGB) is shifted by ⅓ of the main scanning period Th. For this reason, positional deviation of 1/3 dot in the sub-scanning direction occurs on the reading surface of the document between the light sources adjacent in order of light emission. This positional deviation is corrected by image processing.

When the traveling unit 302 reaches the travel completion position 311, reading of the document is finished, and the traveling unit 302 is returned to the home position 310 (S809) to complete the document free erase reading process.

FIG. 10A is a diagram illustrating a read image. The inside of a rectangle 1001 indicated by broken lines is the document, and the outside is the outside of the document. Inside the rectangle 1001, since the document reflects the light emitted from the light sources 311R, 311G and 311B, a striped pattern corresponding to the switching of the light emission amounts of the light sources 311R, 311G and 311B is recognized. On the other hand, since the light emitted from the light sources 311R, 311G, and 311B is not reflected outside the rectangle 1001, no striped pattern is generated, and the area is a black, low-brightness region with approximately uniform brightness.

It should be noted that unevenness in the light amount distribution in the main scanning direction of the light emitted from the light sources 311R, 311G, and 311B guided by the light guide 312, unevenness in the imaging efficiency of the lens array 131, and variations in the photoelectric conversion elements constituting the line sensor 314 Variations in lightness caused by variations in sensitivity are not shown in FIG. 10A, but are corrected by the document-free erase shading correction process described below.
(1-5-2) Original Free Erase Shading Correction Processing (S704)
In the document free erase shading correction process, as shown in FIG. 11, the process from step S1101 to step S1104 is repeated for all scanning lines of the read image. That is, first, a light amount correction coefficient is set according to the length of the lighting period when reading the scanning line (S1102). If the lighting period of the lighting signal when reading the scanning line is Tl, the light intensity correction coefficient is
Light intensity correction coefficient = T1/Tl = 1 (3)
and when the lighting period is Ts,
Light intensity correction coefficient = T1/Ts (4)
is.

Next, the pixel values of all the pixels of the scanning line are corrected using the following formula (S1103).

Pixel value after correction = shading correction coefficient x light amount correction coefficient x pixel value before correction (5)
A shading correction coefficient that is calculated for each pixel in advance is used.

FIG. 10(b) shows the image of FIG. 10(a) subjected to shading correction processing for original free erase. Since the light quantity correction coefficient is multiplied in the shading correction process for document free erase, the striped pattern is eliminated in the document portion inside the rectangle 1002 indicated by the dashed line. On the other hand, since there is no striped pattern outside the rectangle 1002 before correction, a striped pattern is generated by multiplication by the light amount correction coefficient.
(1-5-3) Brightness generation processing (S705)
In the brightness generation process, as shown in FIG. 12, the processes from step S1201 to step S1203 are repeated for all scanning lines. Specifically, the brightness V is generated from the pixel values of RGB colors using the following formula (S1202).

V = (0.3 x R) + (0.6 x G) + (0.1 x B) (6)
By generating a brightness image from a read image (color image) in this way and using the brightness image to detect the boundaries between the inside and outside of the document, the color dependence of the edges of the document can be eliminated as much as possible. Note that the brightness V may be generated using a formula other than the above. Further, in the present embodiment, the brightness of a pixel is represented by 256 gradations, with a value of 0 representing black and a value of 255 representing white.
(1-5-4) Main scanning direction smoothing process (S706)
In the main scanning direction smoothing process, as shown in FIG. 13, the processes of steps S1301 to S1303 are executed for all pixels of the brightness image. That is, the weighted sum of five pixels including the i-th pixel on the scanning line is calculated and divided by the total value of the weights to determine the smoothed lightness of the i-th pixel. (S1302).

SV(i)={V(i-2)+3*V(i-1)+8*V(i)+3*V(i+1)+V(i+2)}/(1+3+8+3+1) (7)
Here, SV(i) represents the brightness of the i-th pixel after smoothing, and V(k) represents the brightness of the k-th pixel before smoothing. For pixels near the end of a scanning line, if the weighted average value of brightness cannot be calculated from five pixels adjacent in the main scanning direction, the number of pixels is reduced and the weighted average value is calculated. you may ask.

For example, when smoothing is performed on the first pixel of a scanning line, since there is no -1-th pixel or 0-th pixel, the weighted average value of the brightness of the first to third pixels is The lightness SV(1) after smoothing of the first pixel may be used. Moreover, it goes without saying that the weighting method is not limited to the above, and the weighting method may be changed as necessary.

The pixel values (brightness level) outside the original are originally low, and the difference in brightness in the sub-scanning direction caused by switching the amount of light is not so large, so that it is easily affected by noise. On the other hand, if the main scanning direction smoothing process is performed to remove the noise component, it is possible to improve the detection accuracy of the brightness step outside the document.
(1-5-5) Sub-scanning direction brightness difference detection process (S707)
In the sub-scanning direction brightness difference detection process, it is determined whether or not the brightness difference between pixels adjacent in the sub-scanning direction in the brightness image is within a predetermined range, so that the pixels form a striped pattern. Determine whether or not it is a pixel. Pixels forming a striped pattern are hereinafter referred to as "edge pixels".

As shown in FIG. 14, the processing from step S1401 to step S1404 is executed for all pixels.

That is, the brightnesses of the three pixels adjacent in the sub-scanning direction across the pixel on the j-th scanning line are INj-1, INj and INj+1 in order in the sub-scanning direction, and the brightness difference threshold value Vth1 is hand,
INj-INj+1>Vth1 (8)
and INj - INj-1 > Vth1 (9)
, the brightness of the central pixel in the sub-scanning direction is higher than the brightness of the neighboring pixels by a brightness threshold or more, or
INj+1-INj>Vth1 (10)
and INj-1 - INj > Vth1 (11)
When the brightness of the central pixel in the sub-scanning direction is lower than the brightness of the pixels on both sides by the brightness threshold or more (S1402: YES), these three pixels are determined to be edge pixels ( S1403).

If the brightness of three pixels adjacent in the sub-scanning direction does not satisfy the above conditional expression (S1402: NO), these three pixels are not determined to be edge pixels. In this embodiment, in the default state prior to the start of the loop processing from steps S1401 to S1405, it is determined that none of the pixels is an edge pixel, and the determination that the pixel is not an edge pixel is the above loop processing. modified by In addition, since it is not determined that the pixel is not an edge pixel, once it is determined that the pixel is an edge pixel, the determination will not be reversed and the pixel will not be determined to be an edge pixel.

For the brightness difference threshold value Vth1, an appropriate value corresponding to the light amounts of the light sources 311R, 311G, and 311B is obtained in advance by experiment or the like. Also, if the pixel is on the first scanning line or the last scanning line, there is only one adjacent pixel in the sub-scanning direction, and the above conditional expression cannot be satisfied. I do not judge that there is.

Furthermore, after determining that three pixels, that is, the pixel Pj-1 on the j-1th scanning line and the pixels Pj-2 and Pj adjacent to this pixel in the sub-scanning direction, are edge pixels, the j-th and the pixel Pj+1 on the j+1-th scanning line adjacent to the pixel in question in the sub-scanning direction do not satisfy the equations (8) and (10), the j-th scanning If the pixel Pj on the line and the pixels Pj-1 and Pj+1 adjacent to the pixel in the sub-scanning direction are not determined to be edge pixels, the pixel Pj+1 is not an edge pixel. However, the determination that the pixels Pj-1 and Pj are edge pixels is maintained.

For example, when the brightness of pixels on line EE in FIG. 10B is as shown in FIG. If the signs of the lightness differences are different from each other when the above conditional expression is applied, it is determined that the center pixel is not an edge pixel.

Also, if one brightness difference is D2 and the other brightness difference is 0, the brightness difference 0 is smaller than the brightness difference threshold Vth1, so these pixels are determined not to be edge pixels. Furthermore, if the brightness difference between the center pixel and the pixels on both sides is D3, and if the brightness difference D3 is greater than the brightness difference threshold Vth1 and the signs match, these pixels are edge pixels. is determined.

Accordingly, on line EE, pixels outside the rectangle 1002 are determined to be edge pixels, while pixels inside the rectangle 1002 are determined not to be edge pixels. Also, all pixels on the FF line are determined to be edge pixels.
(1-5-6) Brightness determination process (S708)
In the brightness determination process, the brightness of each pixel of the brightness image is compared with a predetermined threshold to determine whether or not the pixel is a pixel with low brightness (hereinafter referred to as a "low brightness pixel").

As shown in FIG. 15, the processing from step S1501 to step S1505 is executed for all pixels. First, when the brightness IN of the pixel is less than the brightness threshold Vth2:
IN < Vth2 (12)
(S1502: YES), the pixel is determined to be a low brightness pixel (S1503). If the brightness of the pixel is equal to or higher than the brightness threshold value Vth2 (S1502: NO), it is determined that the pixel is not a low brightness pixel (S1504). As the low brightness threshold Vth2, a common value set in advance is used for all pixels.

Since the light emitted from the light sources 311R, 311G and 311B is not reflected outside the document, the brightness is lower than inside the document. Therefore, by extracting pixels whose lightness is lower than a predetermined level by lightness determination processing, it is possible to detect pixels outside the document.
(1-5-7) Boundary coordinate determination process (S709)
In the boundary coordinate specifying process, as shown in FIG. 16, the process from step S1601 to step S1616 is repeated for each scanning line of the brightness image. First, the value of a working variable i representing the pixel number on the scanning line is initialized to 1 (S1602). It is assumed that the total number of pixels on one scanning line is N.

If the i-th pixel is determined to be a low brightness pixel in the brightness determination process (S708) (S1603: YES) and is determined to be an edge pixel in the sub-scanning direction brightness difference detection process (S707), ( S1604: YES), it is determined that it is not a boundary pixel (S1605), and the value of working variable i is incremented by 1 (S1606). If the value of the working variable i is equal to or less than the number of pixels N on one scanning line (S1607: YES), the process advances to step S1603 to repeat the above processing.

If the value of the working variable i is greater than N (S1607: NO), there is no pixel on the scan line that is determined to be a boundary pixel. This case corresponds to the scanning line on line GG in FIG. 10(b). In this case, the processing for the scan line is terminated, and the process advances to step S1602 to execute processing for the next scan line.

If it is determined that the i-th pixel is not a low-lightness pixel (S1603: NO) or if it is determined that it is not an edge pixel (S1604: NO), the i-th pixel is a boundary pixel (hereinafter referred to as this pixel is referred to as a "left boundary pixel") (S1608). In the scanning line on line HH in FIG. 10B, pixel 1011 corresponds to the i-th pixel. This is because the 1st to i−1th pixels are outside the rectangle 1002 while the pixel 1011 is on the boundary of the rectangle 1002 .

Next, the value of working variable j is initialized to N (S1609). If it is determined that the j-th pixel is a low-lightness pixel (S1610: YES) and that it is an edge pixel (S1611: YES), it is determined that the j-th pixel is not a boundary pixel. Then (S1612), the value of the working variable j is decremented by 1 (S1613). If the value of working variable j is greater than the value of working variable i (S1614: YES), the process advances to step S1609 to repeat the above processing.

If the value of the working variable j is equal to or less than the value of the working variable i (S1614: NO), the j-th pixel has already been processed in steps S1602 to S1608. has been completed, the flow advances to step S1602 to process the next scan line.

If it is determined that the j-th pixel is not a low-lightness pixel (S1610: NO) or if it is determined that it is not an edge pixel (S1611: NO), the j-th pixel is a boundary pixel (hereinafter referred to as this pixel is referred to as a "right boundary pixel") (S1615). In the scanning line on line HH in FIG. 10B, pixel 1012 corresponds to the j-th pixel. This is because the Nth to j+1th pixels are outside the rectangle 1002 , whereas the pixel 1012 is on the boundary of the rectangle 1002 .

After identifying the two boundary pixels corresponding to both ends of the document on the scanning line in this way, the flow advances to step S1602 to process the next scanning line.

After completing the processing for all scanning lines, the boundary coordinates are smoothed (S1617). Boundary coordinate smoothing may be determined, for example, by a weighted average as follows. L(n) represents the position of the left boundary pixel of the n-th scanning line on the scanning line (the number of pixels from the left end), and similarly, the position on the n-th scanning line after smoothing. is represented by SL(n),
SL(n) = {L(n-2)+3*L(n-1)+8*L(n)+3*L(n+1)+L(n+2)}/(1+3+8+3+1) (13)
If it is not possible to calculate the weighted average value of the boundary pixels near the top or bottom of the document from five scanning lines adjacent to each other in the sub-scanning direction, the number of boundary pixels may be reduced. A weighted average may be determined.

For example, when the smoothing process is performed on the boundary pixels at the upper edge of the document, since there are no boundary pixels in the upper scanning line, the positions of the boundary pixels of the next scanning line and the positions of the boundary pixels of the further next scanning line are calculated. may be used as the smoothed position SV(1) of the top boundary pixel. Moreover, it goes without saying that the weighting method is not limited to the above, and the weighting method may be changed as necessary.
(1-5-8) Erase processing (S710)
In the erase process, as shown in FIG. 17, the process from step S1701 to step S1711 is repeated for all scanning lines of the original read image (color image). That is, the value of the working variable i representing the position of the pixel on the scanning line is initialized to 1 (S1702), and if the i-th pixel is not a boundary pixel (S1703: NO), the i-th pixel is Since it is considered to be outside, the i-th pixel is made white (S1704). Specifically, the gradation value of each color of RGB of the pixel is set to 255.

After that, the value of the working variable i is incremented by 1 (S1705), the process proceeds to step S1703, and if the i-th pixel is a boundary pixel (S1703: YES), it is considered that the pixel has entered the document. The value of the working variable i is increased by 1 without changing the gradation value of the i-th pixel (S1706). After that, if the i-th pixel is not a boundary pixel (S1707: NO), it is still considered to be within the document, so the process advances to step S1706.

If the i-th pixel is a boundary pixel (S1707: YES), it is considered that the pixel has escaped outside the document. (S1709). If the value of the working variable i is smaller than the number of pixels N on the scanning line (S1710: YES), the process advances to step S1708 to repeat the above processing.

When the value of the working variable i becomes equal to or greater than N (S1710: NO), it means that processing has been completed up to the end of the scanning line, and processing of the next scanning line is executed.

In this way, the document-free erase process can be executed with high accuracy only by reading the document once, so that the document reading time can be shortened and the user's convenience can be improved.
[2] Second Embodiment Next, a second embodiment of the present invention will be described. The MFP 1 according to the present embodiment has substantially the same configuration as the MFP 1 according to the first embodiment. In this embodiment, instead of performing normal shading correction processing, the method of determining boundary pixels is different. The following description will focus mainly on the differences.

In this specification, common reference numerals are assigned to members common to the embodiments.
(2-1) Shading Correction Processing The shading correction processing according to the present embodiment is normal shading correction processing, and as shown in FIG. repeat. That is, the pixel values of RGB colors are corrected using the following equations (S1802).

Pixel value after correction = shading correction coefficient x pixel value before correction (14)
A shading correction coefficient that is calculated for each pixel in advance is used.
(2-2) Sub-scanning Direction Brightness Difference Detection Processing As described above, in the present embodiment, in the shading correction processing, the light amount correction coefficients corresponding to the light emission amounts of the light sources 311R, 311G, and 311B are not multiplied. As illustrated in FIG. 10(a), the pixels in the document are edge pixels because a striped pattern appears in the document indicated by rectangles 1001 reflecting the light emitted from the light sources 311R, 311G and 311B. is determined. On the other hand, since no striped pattern appears outside the document, pixels outside the document are determined not to be edge pixels.
(2-3) Boundary Coordinate Determination Processing In the first embodiment, pixels outside the document are determined to be edge pixels. It is determined to be a pixel. For this reason, in the boundary coordinate determination processing according to the present embodiment, as shown in FIG. S1911: NO), it is determined that it is not a boundary pixel (S1905, S1912).

By performing such determination processing, the boundary pixels can be identified as in the first embodiment. Therefore, since the document-free erase process can be executed with high accuracy only by reading the document once, it is possible to shorten the document reading time and improve the user's convenience.
[3] Modifications Although the present invention has been described above based on the embodiments, the present invention is of course not limited to the above-described embodiments, and the following modifications can be implemented. .
(3-1) In the sub-scanning direction brightness difference detection processing according to the above-described embodiment, an example will be described in which the magnitude of the brightness difference between adjacent pixels in the sub-scanning direction is determined using the brightness difference threshold value Vth1. However, it goes without saying that the present invention is not limited to this, and instead of this, the following may be done.

For example, the determination may be made based on the brightness obtained by multiplying the brightness of the central pixel among three pixels adjacent in the sub-scanning direction by the light amount ratio when the light emission amounts of the light sources 311R, 311G, and 311B are switched. Specifically, the brightness of three pixels adjacent in the sub-scanning direction is INj-1, INj, and INj+1 in order in the sub-scanning direction, and the coefficients α and β calculated in advance from the light amount ratio (however, α > β > 0),
INj×α>INj−INj+1>INj×β (15)
and INj × α > INj - INj-1 > INj × β (16)
, the brightness of the central pixel in the sub-scanning direction is higher than the brightness of the neighboring pixels by a brightness threshold or more, or
INj×α>INj+1−INj>INj×β (17)
and INj × α > INj-1 - INj > INj × β (18)
, the center pixel may be determined to be an edge pixel when the brightness of the center pixel in the sub-scanning direction is lower than the brightness of the adjacent pixels on both sides by a brightness threshold or more.

The coefficients α and β are, for example,
Ts = Tl x 70% (19)
In addition, if the brightness is uniform on the document, it is assumed that there will be a brightness difference of about 30% between adjacent pixels in the sub-scanning direction after the shading correction process. Therefore, in order to detect the brightness difference (edge) caused by switching the light emission amounts of the light sources 311R, 311G, and 311B,
α = 0.4, β = 0.2 (20)
should be set as

In this way, edges that are too large, such as those that exist inside the document or occur at the document boundary, can be excluded. In addition, since the level of unnecessary light incident outside the document varies depending on the environment in which the multifunction device 1 is installed, by multiplying the brightness level of the read pixel by a coefficient, the influence of this variation due to the installation environment can be reduced. is also valid.

Furthermore, it is desirable to select the conditional expression according to whether the amount of light emitted from the light sources 311R, 311G, and 311B is large or small when reading the scanning line to which the central pixel among the three pixels adjacent in the sub-scanning direction belongs. , the detection accuracy can be further improved by selecting the conditional expression in this way.

For example, in the first embodiment, only conditional expressions (8) and (9) are used when the amount of light emitted from the light sources 311R, 311G, and 311B when reading the scanning line to which the central pixel belongs is large. Then, conditional expressions (10) and (11) are satisfied by the lightness of the document image itself, and erroneous determination as an edge pixel can be prevented.

Similarly, when the amount of light emitted from the light sources 311R, 311G, and 311B when reading the scanning line to which the central pixel belongs is small, edge pixel detection accuracy can be improved by using only conditional expressions (10) and (11). can.
(3-2) In the first embodiment, the shading correction process for document free erase (S704) is performed after the reading process for document free erase (S703) and before the brightness generation process (S705). has been described as an example, but the present invention is not limited to this, and the following alternatives may be used.

For example, after the document free erase reading process (S703) and before the brightness generation process (S705), only the correction process for switching the light emission amounts of the light sources 311R, 311G, and 311B is executed, and the boundary A normal shading correction process may be executed after the coordinate determination process (S709) and before the erase process (S710).

In this case, in the correction processing for switching the light emission amounts of the light sources 311R, 311G, and 311B, for all scanning lines, similarly to S1102 according to the first embodiment, the lighting period when reading the scanning line is Set the light amount correction coefficient according to the length of If the lighting period of the lighting signal when reading the scanning line is Tl, the light intensity correction coefficient is
Light intensity correction coefficient = T1/Tl = 1 (21)
and when the lighting period is Ts,
Light intensity correction coefficient = T1/Ts (22)
is.

Next, the pixel values of all the pixels in the scan line are corrected using the following formula.

Pixel value after correction = light amount correction coefficient x pixel value before correction (23)
Even in this way, the same effect of the present invention can be obtained.
(3-3) In the second embodiment, the shading correction process for document-free erase (S704) is performed after the reading process for document-free erase (S703) and before the brightness generation process (S705). has been described as an example, but the present invention is not limited to this, and the following alternatives may be used. For example, normal shading correction processing may be executed after the boundary coordinate determination processing (S709) and before the erase processing (S710).

Also in this way, the shading of the finally generated read image can be corrected. It should be noted that the above-described second embodiment is effective in that the influence of shading on the processing for detecting boundary pixels can be reduced.
(3-4) In the above embodiment, the case where the light emission amounts of all the light sources 311R, 311G and 311B are switched has been described as an example. , 311G and 311B may be switched, and the other light sources may be kept constant.

For example, as shown in FIG. 20, only the light emission amount of the light source 311G may be switched to keep the light emission amounts of the light sources 311R and 311B constant. By switching only the light emission amount of G color, compared with the case of switching the light emission amount of other light colors, it becomes difficult to visually recognize the influence of the light emission amount switching on the image quality of the read image, thereby suppressing deterioration of the read image quality. It is suitable in the sense that it is possible.

If the amount of light emitted from the light sources 311R, 311G, and 311B is reduced, the sense of noise in the read image may deteriorate slightly. To address such a problem, if the number of light sources for switching the amount of emitted light is reduced, it is possible to suppress deterioration in the image quality of the read image.
(3-5) In the above embodiments, the case where a color image is read from a document has been described as an example, but the present invention is not limited to this, and the present invention is applied to an image reading apparatus that reads a monochrome image. The same effect can be obtained even if
(3-6) In the above embodiment, the reading unit 301 is a 1-line CIS. Alternatively, a CCD (Charge Coupled Device) may be used. The present invention can achieve similar effects regardless of the configuration of the reading unit 301 .
(3-7) In the above embodiment, the case where the light emission amounts of the light sources 311R, 311G and 311B are switched for each scanning line has been described as an example. Alternatively, for example, switching may be performed every two scanning lines, or switching may be performed alternately between switching every one scanning line and switching every two scanning lines.

Moreover, the switching pattern of the light emission amount is not limited to periodic switching, and may be switched aperiodically. For example, the switching may be performed in a pattern in which the switching every one scanning line and the switching every two scanning lines appear aperiodically.

Regardless of the switching pattern, if a pixel having a brightness difference corresponding to the switching pattern can be determined as an edge pixel, the effect of the present invention can be obtained.

Note that when switching the light emission amounts of the light sources 311R, 311G, and 311B every M scanning lines,
The brightness of 3M pixels adjacent in the sub-scanning direction is IN1, . For 1, 2M+1≤j≤3M-1,
INj-INj+1<Vth1 (24)
And,
INM+1 - INM > Vth1 (25)
And,
IN2M - IN2M+1 > Vth1 (26)
, the brightness of the center pixel in the sub-scanning direction is higher than the brightness of the pixels on both sides by the brightness threshold or more, or instead of equations (25) and (26),
INM-INM+1 > Vth1 (27)
and IN2M+1 - IN2M > Vth1 (28)
, when the brightness of the central pixel in the sub-scanning direction is lower than the brightness of the pixels on both sides by a brightness threshold or more, these 3M pixels may be determined to be edge pixels. Also, it goes without saying that other determination methods may be used.
(3-8) In the above embodiment, an example of switching the light emission amount of the light sources 311R, 311G, and 311B in two steps was explained, but it goes without saying that the present invention is not limited to this. may be switched in three or more steps.

Further, in the above embodiment, the case where the light emission amount of the light source is switched according to the length of the light emission period of the lighting signal has been described as an example, but it goes without saying that the present invention is not limited to this, and the lighting signal is a voltage signal. In some cases, the amount of light emitted from the light source may be switched by switching the voltage, or by switching the amount of current when the lighting signal is a current signal.

Further, a shutter is provided between the light sources 311R, 311G and 311B and the light guide member 312, and the surface of the document to be read is illuminated by opening and closing the shutter while selectively lighting the light sources 311R, 311G and 311B. By controlling the period, a similar effect can be obtained even if the amount of illumination light is switched.
(3-9) In the above embodiment, the case where the read image has 256 gradations for each color of RGB has been described as an example, but it goes without saying that the present invention is not limited to this, and the number of gradations is 256 gradations. It may be other than In addition, the light color of the light source may be other than RGB, and the number of types of light colors may be four or more. A similar effect can be obtained by applying the invention.
(3-10) In the above embodiment, the case where the image reading unit 100 is installed in the multifunction machine 1 has been described as an example. Alternatively, it may be installed in a single-function machine such as a facsimile machine or a copier. Also, the same effect can be obtained by applying the present invention to a single image reading device that is not part of another device.
(3-11) In the above embodiment, the image reading unit 100 includes the document cover 102, and the case where the image is read from the document only by the platen setting method has been described as an example, but the present invention is not limited to this. Needless to say, as shown in FIG. 21, an automatic document feeder (ADF) 2101 that also serves as a document cover 102 is provided, and an image can be read from the document by the sheet-through method in addition to the platen setting method. good too. The present invention can also be applied to the image reading unit 100 having the automatic document feeder 2101, and similar effects can be obtained.

INDUSTRIAL APPLICABILITY The image reading device and image forming device according to the present invention are useful as devices that perform free erase processing at high speed and with high accuracy.

1 ……………………Multifunction machine 100 ………………Image reading unit 110 ………………Image forming unit 120 .................Paper supply unit 101 ..................Main unit 102 ..Document cover 201 .................. ……………… Platen glass 203 ……………… Control unit 301 ……………………………… Reading units 311R, 311G, 311B … Light source

Claims (12)

a translucent document placement table for placing the document;
a cover that can be opened and closed to cover the document table;
a light source that emits light toward the document placed on the document table;
light amount control means for switching the irradiation light amount of the light from the light source in units of scanning lines according to a predetermined switching pattern along the sub-scanning direction when reading a document with the cover open;
document reading means for detecting the amount of incident light including the amount of reflected light from the document for each pixel;
a correction means for correcting the detected value of the amount of incident light in accordance with the switching pattern of the amount of irradiation light;
As a result of the correction by the correcting means, there is a group of pixels that are continuous in the sub-scanning direction and in which the detected value after the correction varies along the sub-scanning direction in accordance with the switching pattern of the amount of irradiation light. edge pixel detection means for detecting pixels included in the pixel group as edge pixels;
and boundary identifying means for identifying a boundary of the document so that the edge pixels are located outside the document. Low-lightness pixel detection means for detecting low-lightness pixels whose reflected light amount is less than a predetermined light amount,
2. The image according to claim 1, wherein said boundary identifying means identifies the boundary of the document such that even if it is an edge pixel but not a low-lightness pixel, it is located inside the document instead of outside the document. reader. 3. The image reading apparatus according to claim 1, wherein said correction means also performs shading correction. a manuscript placing table for placing a manuscript;
a cover that can be opened and closed to cover the document table;
a light source for irradiating the document placed on the document table;
light amount control means for switching the irradiation light amount of the light from the light source in units of scanning lines according to a predetermined switching pattern along the sub-scanning direction when reading a document with the cover open;
document reading means for detecting the amount of reflected light from the document for each pixel;
edge pixel detection means for detecting an edge pixel corresponding to the switching pattern of the irradiation light amount, the detection value of the reflected light amount;
an image reading apparatus, comprising: a boundary identifying unit that identifies a boundary of a document so that the pixels other than the edge pixels are located outside the document. Low-lightness pixel detection means for detecting low-lightness pixels whose reflected light amount is less than a predetermined light amount,
5. The image reading apparatus according to claim 4, wherein said boundary identifying means further identifies the boundary of the document such that even if the pixel is not an edge pixel, it is located inside the document unless the pixel is a low-brightness pixel. 6. The image reading apparatus according to claim 1, wherein said light amount control means periodically switches said irradiation light amount along the sub-scanning direction. 7. The image reading apparatus according to claim 1, wherein said light amount control means switches between two types of irradiation light amounts, large and small, for each scanning line. Smoothing means for smoothing the detected value of the amount of reflected light along the main scanning direction,
8. The image reading apparatus according to claim 1, wherein said edge pixel detection means detects edge pixels using smoothed detection values. the light source comprises a plurality of partial light sources having different light colors;
9. The image reading apparatus according to claim 1, wherein said light amount control means switches said irradiation light amount only for some of said plurality of partial light sources. The edge pixel detection means detects a pixel by determining whether a difference in the detected amount of reflected light between adjacent pixels in the sub-scanning direction is within a predetermined range. 10. The image reading device according to any one of claims 1 to 9. 11. The image reading device according to claim 10, wherein the predetermined range is designated by a value corresponding to the switching pattern of the irradiation light amount. An image forming apparatus comprising the image reading apparatus according to claim 1 .

Images