JP2010166368A - Device for detecting position where noise is mixed in and image reader - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device for detecting the position where noise is mixed in that can properly find the position, where mixing-in of noise generated in synchronization with a line period is expected, in image data, and an image reader incorporated with the same. <P>SOLUTION: The device for detecting the position where noise is mixed in includes an image data generating means 12, an averaging means 15, and a means 17 for detecting the position where noise is mixed in. The averaging means 15 calculates an average value by sequentially executing processing in the main scanning direction such that gradation values of pixels appearing at the same position in the main scanning direction of each line are respectively averaged in white image data, showing the white color, for the prescribed number of lines. The means 17 for detecting the position where noise is mixed in executes basic processing for detecting the position where noise is mixed in such that a difference between two adjacent average values of the respective average values is sequentially calculated in the main scanning direction so as to detect the position, where noise generated in synchronization with the line period is mixed in the averaged white image data, on the basis of the calculated difference between the two average values as the position, where noise is mixed in, that mixing-in of noise is expected in the image data. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a noise-containing position detecting device and an image reading device that detect a position where noise generated in synchronization with a line cycle is expected to be mixed in image data representing an image.

  An image reading apparatus such as a facsimile machine, an image scanner, or a copier is configured to irradiate a document with light from a light source and read image data with reflected light. An example of this type of image reading apparatus is exemplified in Patent Document 1.

  In this type of image reading apparatus, the illumination lamp emits light with a small cycle. Light generated by such light emission is reflected by the original set on the original table. Then, the reflected light is collected by the lens, reaches a sensor board unit in which CCDs (solid-state imaging devices) are arranged in a row, is photoelectrically converted, and is accumulated as electric charges. The charges accumulated in the sensor board unit in this way are pixels of image data for one line of the document.

  In this manner, the process of accumulating pixels for one line of the document in the sensor board unit is performed in synchronization with the horizontal synchronization signal. As described above, the pixels for one line of the document accumulated in the sensor board unit are converted into electrical signals in time series in synchronization with the transfer clock and sequentially transferred to the subsequent circuit. Such processing is repeated in the sub-scanning direction of the document, and the pixels for one page of the document are converted into electrical signals according to the amount of charge and sequentially transferred to the subsequent circuit, so that image data per page of the document is obtained. Is generated.

JP 2006-74161 A

  Incidentally, in this type of image reading apparatus, noise is generated due to various factors. For example, noise generated from an inverter that emits an illumination lamp, noise generated from a motor that moves the sensor board unit in the sub-scanning direction, and the like can be given.

  In this way, when noise is generated, the generated noise is mixed into the pixels transferred from the sensor board unit. When noise is mixed in the transferred pixels in this way, the charge amount becomes unstable due to the noise, so that the gradation value of a pixel that originally has a large gradation value may be small. Conversely, the gradation value of a pixel that originally has a small gradation value may be large. As described above, a pixel having a gradation value different from the original gradation value of the pixel tends to remain on the recording paper surface as noise when an image is finally formed.

  In particular, when noise is generated in synchronization with the line period, the noise is mixed into a specific position in the main scanning direction in the image data. As a result, when pixels for one page of the document are formed, striped noise remains on the surface of the recording paper in the sub-scanning direction. Here, the line cycle is a period in which image data is generated in units of one line by the sensor board unit.

  In order to prevent the noise generated in synchronization with the line period from remaining on the surface of the recording paper in this way, the position where the noise generated in synchronization with the line period is expected to be mixed in the image data can be appropriately determined. Is desired.

  The present invention has been made to solve the above-described problem. In the image data, a noise-mixing position detection device that can appropriately determine a position where noise generated in synchronization with the line period is expected to be mixed, and An object of the present invention is to provide an image reading apparatus incorporating a noise-mixing position detection device.

  The noise-mixed position detection device according to one aspect of the present invention accumulates, as pixels, charges obtained by photoelectrically converting reflected light generated by reflecting light emitted from a light source, and storing the accumulated pixels. By performing the process of outputting as an electric signal in units of one line, image data generating means for generating image data in units of one line, and white for a predetermined number of lines representing white generated by the image data generating means In the image data, a process of averaging the gradation values of the pixels appearing at the same position in the main scanning direction of each line is sequentially performed in the main scanning direction to obtain an average value, and each average value is calculated in the main scanning direction. Averaging means for generating averaged white image data arranged so as to correspond to pixel positions, and among average values arranged in the averaged white image data generated by the averaging means, A period in which a difference between two average values is sequentially calculated in the main scanning direction, and image data is generated in units of one line by the image data generation unit based on the calculated difference between the two average values. A noise mixing position for detecting a position where noise generated in synchronization with the line period representing the mixed white noise data is mixed into the averaged white image data as a noise mixing position expected to be mixed in the image data generated by the image data generating means And a noise-containing position detecting means for performing basic detection processing (claim 1).

  According to this configuration, the image data generation unit generates white image data representing white for a predetermined number of lines. For such white image data for a predetermined number of lines, the gradation values of pixels appearing at the same position in the main scanning direction of each line are averaged to obtain an average value, and the average value is determined in the main operation direction. The averaged white image data is arranged so as to correspond to each pixel position.

  Using such averaged white image data, a noise mixing position that is expected to include noise generated in synchronization with the line period in the image data is detected. That is, in the averaged white image data, the process of calculating the difference between the two adjacent average values by subtracting the average value located next to the average value on the start side in the main scanning direction from the average value sequentially. This is performed in the main scanning direction. Here, the line cycle is a period in which image data is generated in units of one line by the image data generating means.

  Then, based on the difference between the two average values calculated sequentially, the position where the noise generated in synchronization with the line period is mixed into the averaged white image data is generated by the image data generation means. Is detected as a noise mixing position expected to be mixed.

  For this reason, in the averaged white image data generated by averaging the white image data representing white, the position where the noise generated in synchronization with the line period is mixed can be appropriately determined. Therefore, the position where noise generated in synchronization with the line period is expected to be mixed in the image data can be properly determined.

  The above configuration further includes a storage unit that stores the noise mixing position detected by the noise mixing position detection unit, and the noise mixing position detection unit performs the noise mixing position detection basic process at a preset timing. To detect the noise mixing position and store the detected noise mixing position in the storage means (claim 2). According to this configuration, the noise mixing position detection basic process is performed at a preset timing, and the noise mixing position is detected and stored in the storage unit. Therefore, the noise mixing position is detected without impairing the original productivity of the image reading apparatus that generates image data and removes the pixel appearing at the noise mixing position from the pixels forming the image data.

  In the above-described configuration, the noise mixing position detection means subtracts two adjacent values by subtracting an average value located next to the start side in the main scanning direction of each average value from each average value in the averaged white image data. The process of calculating the difference between the two average values is sequentially performed in the main scanning direction, and in the averaged white image data, the average value difference is equal to or greater than N (where N is a positive numerical value; the same applies hereinafter). Of the two average values, the position where the average value on the start side in the main scanning direction appears is the noise start position, and in the averaged white image data, two adjacent average values whose difference between the average values is N or more appear. The position where the average value on the end side in the main scanning direction appears among two adjacent average values whose difference between the average values is M (where M is a negative numerical value; the same applies hereinafter) and less than 0 appears for the first time after The noise end position In the averaged white image data, the average value including the noise start position and the noise end position and sandwiched between the average value appearing at the noise start position and the average value appearing at the noise end position Can be configured to detect the noise-mixing position.

  According to this configuration, the position at which the average value on the start side in the main scanning direction appears as the noise start position among two adjacent average values whose average value difference is N or more. The main scanning direction end side of two adjacent average values whose average value difference is not less than M and less than 0 appears for the first time after two adjacent average values having an average value difference of N or more appear. The position where the average value of appears is the noise end position.

  And in the averaged white image data, the range including the noise start position and the noise end position, and the average value sandwiched between the average value appearing at the noise start position and the average value appearing at the noise end position Is detected as a noise mixing position.

  Therefore, in the averaged white image data, a range from an average value appearing at the noise start position to an average value appearing at the noise end position is detected as a position where noise generated in synchronization with the line period is mixed. Therefore, a position where noise generated in synchronization with the line period is expected to be mixed in the image data is detected with high accuracy.

  In the above configuration, the apparatus includes a threshold setting unit that receives a value representing the N and a value representing the M, and the noise mixing position detecting unit is a value representing the N set in the threshold setting unit. In addition, the noise mixing position may be detected using a value representing the M.

  According to this configuration, the noise mixing position detection unit detects the noise mixing position using the value representing N and the value representing M set in the threshold setting unit. Therefore, the value representing N and the value representing M used for detecting the noise mixing position can be appropriately changed, and the noise mixing position is detected with high accuracy.

  In addition, the phenomenon in which image data is not properly formed as a result of too many detected noise mixing positions and a large number of pixels removed from the image data is prevented by lowering the numerical value representing the threshold value. .

  In the above configuration, from the storage means for storing the noise mixing position detected by the noise mixing position detection means and the averaged white image data generated by the averaging means, an average value appearing at the noise mixing position is calculated. In the average value removing means to be removed and the noise-removed averaged white image data from which the average value appearing at the noise mixing position has been removed, the setting of the target value of the maximum value of the difference between the two adjacent average values is accepted. Target value setting means, averaged white image data generation processing for generating the averaged white image data for the averaging means, and noise mixing position detection for detecting the noise mixing position for the noise mixing position detection means Basic processing, noise mixing position description for storing the noise mixing position detected by the noise mixing position detection means in the storage means Processing, causing the averaging means to generate the averaged white image data, and causing the average value removing means to store the noise-containing position stored in the storage means among the averaged white image data. In the noise-removed averaged white image data generating process for removing the average value appearing in the above, the process for calculating the difference between the two average values adjacent to each other in the noise-removed averaged white image data is sequentially performed in the main scanning direction. A maximum average value difference determination that determines whether or not the average value difference that is the maximum value is less than or equal to the target value set in the target value setting means among the sequentially calculated average value differences When the difference between the processing and the average value that is the maximum value is not less than or equal to the target value, a value obtained by subtracting a preset value from the value representing N is set in the threshold setting means. Threshold update And a control unit that repeatedly performs the noise-mixed position detection process consisting of the above-described order until the difference between the average values that is the maximum value is equal to or less than the target value (Claim 5). .

  According to this configuration, the control unit detects the noise mixing position in the averaged white image data while sequentially reducing the threshold value represented by N, and calculates the average value appearing at the detected noise mixing position from the averaged white image data. The removal process is repeated until the maximum difference between the two adjacent average values is equal to or less than the target value. When the maximum difference between two adjacent average values becomes equal to or less than the target value, the noise mixing position detected at that time is finally stored in the storage means. Therefore, a position where noise generated in synchronization with the line cycle is expected to be mixed in the image data is detected with high accuracy.

  In the above configuration, a storage unit that stores the noise mixing position detected by the noise mixing position detection unit, and the noise mixing position stored in the storage unit are read out, and the averaging generated by the averaging unit In white image data, an interval in the main scanning direction is sequentially calculated between the noise mixing position and the noise mixing position located immediately on the main scanning direction start side or the main scanning direction end side of the noise mixing position. In the noise interval calculation process, the noise mixture reference period determination process for determining the interval having the highest appearance frequency as the noise mixture reference period among the intervals obtained by the noise interval calculation process, the averaged white image data, Synchronous noise determination processing for determining the noise mixing position synchronized with the noise mixing reference period, and the averaged white image Control means for performing noise mixing position detection processing comprising the order of noise mixing position deletion processing for deleting the noise mixing positions not synchronized with the noise mixing reference period from the storage means. 6).

  According to this configuration, in the averaged white image data, the control unit performs main scanning between a noise mixing position and a noise mixing position located immediately on the main scanning direction start side or main scanning direction end side of the noise mixing position. The interval in the direction is sequentially calculated in the main scanning direction. Then, the interval with the highest appearance frequency is set as the noise mixing reference period, and the noise mixing position that is not synchronized with the noise mixing reference period is deleted from the storage unit in the averaged white image data.

  Therefore, in the averaged white image data, only the noise mixing position synchronized with the noise mixing reference period is finally stored in the storage unit. Therefore, an accurate position where noise generated in synchronization with the line period is expected to be mixed in the image data can be known.

  In the above configuration, the control means can perform the noise-mixing position detection process at a preset timing (claim 7). According to this configuration, the noise mixing position detection process is performed at a preset timing to detect the noise mixing position. Therefore, the noise mixing position is detected without impairing the original productivity of the image reading apparatus that generates image data and removes the pixel appearing at the noise mixing position from the pixels forming the image data.

  An image reading apparatus according to another aspect of the present invention includes a noise mixed position detection apparatus according to any one of claims 1 to 7 and an image data generation unit included in the noise mixed position detection apparatus. And a pixel removing unit that removes pixels appearing at the noise-mixed position detected by the noise-mixed position detection device in the image data generated in units.

  According to this configuration, the pixel appearing at the noise mixed position detected by the noise mixed position detection device is removed from the image data generated by the image data generating unit in units of one line. Therefore, even if noise generated in synchronization with the line cycle is mixed in the image data, the mixed noise does not remain on the surface of the recording paper when the image data is formed.

  In addition, in the image data, pixels appearing at the noise-mixed position detected by the noise-mixed position detection device are removed, so that it is substantially synchronized with the line cycle, but noise generated while gradually deviating from the line cycle is imaged. Even if it is mixed in the data, the mixed noise is appropriately removed.

  In the above configuration, the average value removing unit includes pixels appearing at a noise mixing position detected by the noise mixing position detection device in the image data generated by the image data generating unit in units of one line, and the noise mixing. A predetermined number of pixels close to the start side or the end side in the main scanning direction with respect to the pixel appearing at the position can be removed (claim 9).

  According to this configuration, in the noise that occurs while deviating from the line period, the allowable period that deviates from the line period increases, so even when noise that occurs while deviating within the range of the allowable period from the line period is mixed in the image data, The mixed noise is appropriately removed.

  According to the present invention, in the averaged white image data generated by averaging the white image data representing white, the position where the noise generated in synchronization with the line period is mixed can be appropriately determined. Therefore, the position where noise generated in synchronization with the line period is expected to be mixed in the image data can be properly determined.

It is a block diagram which shows an example of a function structure of the noise mixing position detection apparatus and image reading apparatus which concern on one Embodiment of this invention. 1 is a longitudinal sectional view showing a mechanical configuration of a copying machine incorporating an image reading apparatus according to an embodiment of the present invention. 1 is a perspective view illustrating an example of an appearance of an image reading device viewed from above. It is a figure which shows typically the averaged white image data produced | generated by the averaging part. It is a flowchart which shows an example of the noise mixing position detection process which a control part performs. It is a figure for demonstrating the other example of the noise mixing position detection process which a control part performs. It is a figure for demonstrating the noise removal process which an image reading apparatus performs.

  Hereinafter, a noise mixed position detection apparatus and an image reading apparatus according to an embodiment of the present invention will be described. FIG. 1 is a block diagram illustrating an example of functional configurations of a noise-mixing position detection device and an image reading device according to an embodiment of the present invention.

  An image reading apparatus A shown in FIG. 1 includes a control unit (control unit) 10, a white reference plate 11, a CCD (charge coupled device; image data generation unit) 12, a memory block 13 having a plurality of line memories 14, and an averaging unit. (Averaging unit) 15, Threshold storage unit (threshold setting unit) 16, Noise mixed position detection unit (Noise mixed position detection unit) 17, Target value storage unit (Target value setting unit) 18, Noise mixed position storage unit (Storage) Means) 19, an average value removing unit (average value removing unit) 20, a pixel removing unit (pixel removing unit) 21, an image processing unit 22, and an external I / F (interface) 23.

  In this image reading apparatus A, a noise mixed position detection apparatus 1 is incorporated. 1 includes a control unit 10, a white reference plate 11, a CCD 12, a memory block 13, an averaging unit 15, a threshold storage unit 16, a noise mixing position detection unit 17, a target value storage unit 18, It comprises a noise mixing position storage unit 19 and an average value removal unit 20.

  In the noise mixing position detection apparatus 1 shown in FIG. 1, the control unit 10 includes a CPU (Central Processing Unit) and the like, and transfers control signals and data through a control bus (for example, a CPU bus) and a data bus, The noise-mixing position detection device 1 and the image reading device A are controlled. Moreover, the control part 10 performs the noise mixing position detection process mentioned later.

  The white reference plate 11 is a plate on which white is drawn in order to detect a noise mixing position where noise generated in synchronization with the line cycle is mixed in the image data generated by the CCD 12. Here, the line period is a period in which image data is generated by the CCD 12 in units of one line.

  The CCD 12 stores a charge obtained by photoelectrically converting reflected light generated by reflecting light emitted from a light source (not shown) as a pixel, and outputs the accumulated pixel as an electric signal for one line. By performing in units, image data is generated in units of one line.

  The memory block 13 temporarily and sequentially stores the image data generated by the CCD 12 in units of one line by each of the plurality of line memories 14. That is, each of the plurality of line memories 14... Temporarily stores image data having the number of lines (hereinafter referred to as a predetermined number of lines) that matches the number of line memories 14.

  Further, the image data of the predetermined number of lines temporarily stored in this way is output to the averaging unit 15 and the pixel removing unit 21.

  The averaging unit 15 mainly performs a process of averaging the gradation values of pixels appearing at the same position in the main scanning direction of each line in the white image data for the predetermined number of lines representing white generated by the CCD 12. An average value is obtained by sequentially performing in the scanning direction, and averaged white image data arranged so that each average value corresponds to each pixel position in the main scanning direction is generated. The white image data is image data representing the white color drawn on the white reference plate 11, but may be image data representing the white color of the white plain paper placed on the document table.

  The threshold value storage unit 16 has a value “N (where N is a positive numerical value; the same applies hereinafter)” and a value “M” (note that the noise mixing position detection unit 19 detects a noise mixing position. "M is a negative numerical value; the same applies hereinafter") is received and stored. Values N and N representing such a threshold are set by operating an operation unit (not shown). Further, the values representing the threshold values N and M are automatically set by the control unit 10 when the control unit 10 performs a noise mixing position detection process described later.

  The noise mixing position detection unit 17 detects the noise mixing position in the averaged white image data. Note that a method of detecting a noise mixing position in the averaged white image data will be described later.

  The target value storage unit 18 accepts and stores the setting of the target value of the maximum value of the difference between two adjacent average values in the noise-removed averaged white image data described later. Such a target value is set by operating an operation unit (not shown). The noise mixing position storage unit 19 stores the noise mixing position detected by the noise mixing position detection unit 17. The storage mode of the noise mixing position will be described later.

  The average value removing unit 20 removes the average value appearing at the noise mixing position stored in the noise mixing position storage unit 19 in the averaged white image data. Note that a method of removing the average value appearing at the noise mixing position in the averaged white image data will be described later.

  Further, in the image reading apparatus A shown in FIG. 1, the pixel removing unit 21 removes pixels appearing at the noise mixed position detected by the noise mixed position detecting apparatus 1 from the image data generated by the CCD 12. A method for removing pixels appearing at the noise-mixing position in the image data will be described later.

  The image processing unit 22 performs image processing such as shading, masking processing, gamma correction, edge enhancement, smoothing, and color conversion on the image data generated by the CCD 12 and from which the pixels appearing at the noise-containing positions are removed. I do. The external I / F 23 is connected to an external device and sequentially outputs image processed image data to the external device.

  FIG. 2 is a longitudinal sectional view showing a mechanical configuration of a copying machine incorporating an image reading apparatus according to an embodiment of the present invention. The copying machine 31 generally includes a main body portion 32, a scanner portion 33 (image reading apparatus A) disposed above the main body portion 32, and an ADF 34 disposed above the scanner portion 33. And is configured.

  In the main body 32, any one of the recording papers 41a, 42a, 43a, 44a loaded in one or a plurality of (three in FIG. 2) trays 41, 42, 43 and the manual feed tray 44 is taken-in rollers 41b, 42b, 43b. 44b are taken out one by one, adjusted in timing by registration rollers 45, 46, and then conveyed to the image forming unit 47. The image forming unit 47 includes a charger (not shown), a laser writing unit, a developing unit, a transfer unit 47b, a cleaning unit (not shown), and the like around the photosensitive drum 47a, and forms an image on a recording sheet by electrophotography. . The toner image thus formed on the recording paper is fixed by the fixing unit 48 and discharged from the discharge rollers 49 and 50 onto the paper discharge tray 51.

  Image data given to the laser writing unit is read by the scanner unit 33. The scanner unit 33 includes a scanning unit 53 that irradiates the original with illumination light through the document table 52 and receives reflected light under the document table 52 made of a plate-shaped transparent plate such as a glass plate, and a scanning unit. A mirror unit 54 for further reflecting the reflected light obtained at 53, an imaging lens 55 for condensing the reflected light from the mirror unit 54, and a CCD 12 for photoelectrically converting the reflected light imaged by the imaging lens 55, It is configured with.

  The scanning unit 53 is driven in the sub-scanning direction by a stepping motor (not shown), for example. The scanning unit 53 is displaced at a speed V and the mirror unit 54 is displaced at a speed V / 2 in the left-right direction of FIG. The original image is always formed on the CCD 12 with the same optical path length.

  On the other hand, in the ADF 34 that sequentially takes in sheet originals, the originals 62 stacked on the original tray 61 are taken out one by one by the take-in rollers 63 and supplied to the curved conveyance path 64. And it is conveyed by the conveyance roller 65,66; 67,68 provided in the curved conveyance path 64 to the reading window 72 which consists of plate-shaped transparent plates, such as a glass plate extended in the main scanning direction, and a scanning part is carried out to the reading window 72. The original images are sequentially read in a state where 53 faces, and are then discharged onto the discharge tray 71 by the discharge rollers 69 and 70.

  The scanning unit 53 includes a cold cathode fluorescent lamp 531 (light source), a reflector 532 that irradiates light emitted from the cold cathode fluorescent lamp 531 toward the original table 52 and the reading window 72, and the original table 52 and the reading window 72 from the original. And a mirror 533 for reflecting the reflected light obtained by passing through the mirror unit 54 to the mirror unit 54.

  A white reference plate 11 is disposed between the document table 52 and the reading window 72 on the scanning unit 53 side. The control unit 10 draws the white reference plate 11 by moving the scanning unit 53 by a predetermined distance in the sub-scanning direction when performing the noise mixing position detection basic process and the noise mixing position detection process described later. The white color is scanned. As a result, the CCD 12 generates white image data for a predetermined number of lines representing the white color of the white reference plate 11.

  FIG. 3 is a perspective view illustrating an example of an appearance of the scanner unit 33 as viewed from above. As shown in FIG. 3, a document table 52 and a reading window 72 are disposed on the upper surface of the scanner unit 33. A document stopper 75 is provided between the document table 52 and the reading window 72. A document stopper 76 is provided on one side of the document table 52 adjacent to the document stopper 75. Then, the document stoppers 75 and 76 refer to the document size index to align the document. In this case, the inner edge of the document stopper 75 indicates the position of the end of the document in the sub-scanning direction. As shown in FIG. 3, the white reference plate 11 is disposed on the back side of the document stopper 75.

  The white reference plate 11 has a substantially rectangular shape, the length in the main scanning direction is substantially equal to the length in the main scanning direction of the document table 52, and the length in the sub scanning direction is equal to a predetermined line. Has been. With the white reference plate 11 having such a configuration, the CCD 12 can generate white image data for a predetermined number of lines representing the white color of the white reference plate 11.

  FIG. 4 is a diagram schematically showing the averaged white image data generated by the averaging unit 15, and FIG. 4A schematically shows the averaged white image data D1 in which no noise is mixed, FIG. 4B schematically shows the averaged white image data D2 in which noise is mixed.

  The averaged white image data D1 shown in FIG. 4A and the averaged white image data D2 shown in FIG. 4B correspond to pixel positions in the main scanning direction of white image data for a predetermined number of lines. Are composed of a plurality of average values P. Each of the plurality of average values P is a process of averaging the gradation values of pixels appearing at the same position in the main scanning direction of each line in white image data for a predetermined number of lines representing white. Is an average value obtained sequentially in the main scanning direction.

  The averaged white image data D2 shown in FIG. 4B is used when the noise mixing position detection unit 17 detects the noise mixing position. That is, the noise mixing position detection unit 17 performs the following processing using the averaged white image data D2, and noise generated in synchronization with the line period is mixed in the image data generated by the CCD 12 in units of one line. Then, the expected noise mixing position is detected.

  Note that noise generated without synchronizing with the line period is mixed at random positions in each line of white image data for a predetermined number of lines. Thus, noise mixed at random positions is averaged image data. It has been removed in the process of generating D2.

  In FIG. 4B, X indicates an example of noise mixed in the averaged white image data D2. As an example of the noise X, a process in which the noise mixing position detection unit 17 detects the noise mixing position will be described. Note that the process in which the noise mixing position detection unit 17 detects the noise mixing position is referred to as “noise mixing position detection basic process” in the present specification.

"Noise-mixed position detection basic processing"
That is, the noise mixing position detection unit 17 subtracts the average value P and the average value P located next to the start side in the main scanning direction from the average value P in the averaged white image data D2, thereby calculating the average of two adjacent values. The process of calculating the difference between the values P is sequentially performed in the main scanning direction.

  For example, in the noise X shown in FIG. 4B, the average value P (1) is subtracted from the average value P (2), the average value P (2) is subtracted from the average value P (3), and the average value P Subtracting average value P (3) from (4), subtracting average value P (4) from average value P (5), subtracting average value P (5) from average value P (6), and The average value P (6) is sequentially subtracted from the average value P (7).

  As described above, the noise-mixing position detection unit 17 sequentially performs the process of calculating the difference between the two adjacent average values P and P in the main scanning direction, and in the averaged white image data D2, Determine the noise end position.

[Determination of noise start position]
The noise-mixing position detection unit 17 determines two adjacent average values P and P having an average value difference equal to or greater than N, and of the two adjacent average values P and P thus determined, is on the main scanning direction start side. The position where the average value P appears is defined as the noise start position.

  For example, in the noise X shown in FIG. 4B, adjacent average values P (1) and P (2) that are equal to or greater than the numerical value N represented by “threshold 1” (where N is a positive numerical value; the same applies hereinafter). ) And the position where the average value P (1) on the start side in the main scanning direction appears among the adjacent average values P (1) and P (2) is set as the noise start position.

[Determination of noise end position]
The noise-mixing position detection unit 17 has an average value difference M (where M is a negative numerical value; the same applies hereinafter) that appears for the first time after two adjacent average values P and P having an average value difference of N or more appear. ) Two adjacent average values P and P that are equal to or more than 0 are determined, and a position where the average value P on the end side in the main scanning direction appears among the two adjacent average values P and P is defined as a noise end position.

  For example, in the noise X shown in FIG. 4B, the numerical value M expressed by “threshold 2” is less than 0 and appears for the first time after the two average values P (4) and P (5) described above appear. Adjacent average values P (6) and P (7) are determined, and the position where the average value P (7) on the end side in the main scanning direction of the adjacent average values P (6) and P (7) appears is noise generation. It is considered as a position.

  The noise-mixing position detection unit 17 sequentially calculates the difference between two adjacent average values P and P in the main scanning direction, and then sequentially calculates the two adjacent average values P and P. The noise start position and the noise end position may be determined using the difference.

  Thus, after determining the noise start position and the noise end position in the averaged white image data D2, the noise mixing position detection unit 17 determines the noise mixing position of the noise mixed in the averaged white image data D2. To detect.

  That is, the noise mixing position detection unit 17 includes the noise start position and the noise end position in the averaged white image data D2, and the average value P that appears at the noise start position and the average value P that appears at the noise end position. The range in which the average value P sandwiched between is arranged as a noise mixing position. For example, in the noise X shown in FIG. 4B, the range in which the average value P (1) to the average value P (7) are arranged is the noise mixing position.

  Here, the range in which the average value P sandwiched between the noise start position, the noise end position, and the average value P appearing at the noise start position and the average value P appearing at the noise end position is, for example, It is expressed as That is, the average value P set as the noise start position, the average value P set as the noise end position, and the average value sandwiched between the average value P appearing at the noise start position and the average value P appearing at the noise end position In other words, P is expressed in the average value order indicating the order in which the averaged white image data D2 is arranged in the main scanning direction.

  For example, in the noise X shown in FIG. 4B, the noise start position where the average value P (1) appears is expressed in the average value order (1). Further, the position where the average value P (2) appears is expressed in the order of the average value (2). Further, the position where the average value P (4) appears is expressed in order of average value (3). Further, the position where the average value P (6) appears is expressed in the order of average value (4). Further, the position where the average value P (7) appears is expressed in the order of the average value (5).

  In this way, in the averaged white image data D2 generated by averaging the white image data for the predetermined number of lines representing white, the averaged white image is used by using the numerical value N and the numerical value M that represent preset threshold values. The position of noise mixed in the data D2 is detected.

  Therefore, in the averaged white image data D2 generated by averaging the white image data representing white, the position where the noise generated in synchronization with the line period is mixed can be appropriately known. Therefore, the position where noise generated in synchronization with the line period is expected to be mixed in the image data can be properly determined.

  In this way, the noise mixing position detected by the “noise mixing position detection basic process” is stored in the noise mixing position storage unit 19 in accordance with the control by the control unit 10. Here, as described above, the noise mixing position storage unit 19 includes the noise start position and the noise end position as the noise mixing position, and also appears at the average value P and the noise end position that appear at the noise start position. A range in which the average values P sandwiched between the average values P are arranged is stored. For example, the average value order (1) to the average value order (5) are stored in the noise mixed position storage unit 19 as the noise mixed positions where the noise X shown in FIG.

  Below, the noise mixing position detection process which the control part 10 performs is demonstrated. Such noise-mixed position detection processing is performed for the purpose of accurately detecting, in the image data, a position where noise generated in synchronization with the line period is expected to be mixed.

"Noise mixed position detection process 1"
FIG. 5 is a flowchart illustrating an example of the noise mixed position detection process performed by the control unit 10. First, as preprocessing, the control unit 10 receives a setting of a numerical value N representing a threshold value 1 for noise detection and a setting of a target value T representing an adjacent average value difference target value (step S1). Here, the adjacent average value difference target value means the target value of the maximum value of the difference between two adjacent average values.

  In FIG. 5, “10h” is set as the numerical value N, and “05h” is set as the adjacent average value difference target value. The numerical value N is stored in the threshold value storage unit 16, and the target value T is stored in the target value storage unit 18.

  Next, the control unit 10 causes the averaging unit 15 to generate the averaged white image data D2 (step S2; averaged white image data generation processing), and the noise mixing position detection unit 17 described above “ “Noise mixing position detection basic process” is performed to detect the noise mixing position (step S3; noise mixing position detection processing), and the detected noise mixing position is stored in the noise mixing position storage unit 19 (step S4; noise mixing). Position memory processing).

  Next, the control unit 10 causes the averaging unit 15 to generate new averaged white image data D2, and causes the average value removal unit 20 to generate the newly generated averaged white image data D2. Of these, the average value appearing at the noise mixing position stored in the noise mixing position storage unit 19 is removed (step S5; noise-removed averaged white image data generation process).

  For example, when the noise mixing position where the noise X shown in FIG. 4B is mixed, that is, the average value order (1) to the average value order (5) is stored in the noise mixing position storage unit 19, The average value P (1) to the average value P (7) appearing at the positions represented by the average value order (1) to the average value order (5) are removed from the averaged white data D2.

  As a method for removing the average value P, for example, an electric signal representing the averaged white image data generated by the averaging unit 15 may be masked.

  Next, the control unit 10 sequentially performs a process of calculating a difference between two adjacent average values P and P in the noise-removed averaged white image data in the main scanning direction (step S6). It is determined whether or not the average value difference that is the maximum value among the average value differences is equal to or less than the target value T (step S7). Here, the process composed of steps S6 and S7 is referred to as a maximum average value difference determination process.

  When the difference between the average values that is the maximum value obtained as a result of the maximum average value difference determination process is not less than or equal to the target value T (NO in step S7), the control unit 10 determines in advance from the value representing “N”. A value obtained by subtracting the set value (here, “02h”) is set in the threshold value storage unit 16 (step S8; threshold value update process), and the processes shown in and after step S2 are performed.

  Then, when the difference between the average values that is the maximum value obtained as a result of performing the maximum average value difference determination process is equal to or less than the target value T (YES in step S7), the control unit 10 performs the noise mixing position detection process. finish.

  In this manner, the control unit 10 detects the noise mixing position in the averaged white image data D2 while sequentially reducing the noise detection threshold value represented by “N”, and averages the average value appearing at the detected noise mixing position. The process of removing from the converted white image data D2 is repeated until the maximum value of the difference between the two adjacent average values P and P is equal to or less than the target value T. When the maximum difference between the two adjacent average values P and P becomes equal to or less than the target value T, the noise mixing position detected at that time is finally stored in the noise mixing position storage unit 19. Is done. Therefore, a position where noise generated in synchronization with the line cycle is expected to be mixed in the image data is detected with high accuracy.

"Noise mixed position detection process 2"
FIG. 6 is a diagram for explaining another example of the noise-containing position detection process performed by the control unit 10.

  In the averaged white image data D2 shown in FIG. 6, the vertical direction means an average value P obtained by averaging the gradation values of pixels appearing at the same position in the main scanning direction of the white image data for a predetermined number of lines. The horizontal direction means the main scanning direction in which each of the average values P is arranged corresponding to each pixel position of the white image data for a predetermined number of lines.

  Further, a line drawn in the vertical direction in the averaged white image data D2 shown in FIG. 6 means noise mixed in the averaged white image data. In the averaged white image data D2, each of L1 to L12 indicates a noise mixing position.

[Noise interval calculation processing]
In the averaged white image data D2, the control unit 10 detects the noise mixing positions L1 to L12 detected by the above-described “noise mixing position detection basic process”, and the main scanning direction start side or the main of the noise mixing positions L1 to L12. The intervals in the main scanning direction with the noise mixing positions L1 to L12 located closest to the end side in the scanning direction are sequentially calculated in the main scanning direction.

  For example, in FIG. 6, the interval between the noise mixing position L1 and the noise mixing position L2 is calculated. Similarly, the interval between the noise mixing position L2 and the noise mixing position L3 is calculated. When the calculation of such an interval is sequentially performed in the main scanning direction and the calculation of the interval between the noise mixing position L12 and the noise mixing position L11 is completed, the noise interval calculation process ends.

  For example, FIG. 6A shows the intervals sequentially calculated in the main scanning direction, the interval between the noise mixing position L1 and the noise mixing position L2 is “3”, and the noise mixing position L3 and the noise mixing position. The interval with L2 is “10”.

  An example of the interval in the main scanning direction is a time interval. For example, if the unit time for the CCD 12 to generate white image data for one pixel is set in advance, a value obtained by multiplying the number of pixels representing the interval in the main scanning direction by the unit time is the time interval. It becomes an interval.

[Noise-mixing reference period determination process]
Next, the control unit 10 determines the interval having the highest appearance frequency among the intervals obtained by the noise interval calculation process as the noise mixing reference period. For example, in FIG. 6B, since the interval “10” is the interval with the highest appearance frequency, the interval “10” is determined to be the noise mixing reference period.

[Synchronous noise judgment processing]
Next, the control unit 10 determines the noise mixing positions L1 to L12 synchronized with the noise mixing reference period in the averaged white image data D2. For example, in FIG. 6 (3), the noise mixing positions L2 and L3 that are closest to the start side in the main scanning direction of the averaged white image data D2 and are synchronized with the noise mixing reference period “10” are determined. Such noise mixing positions L2 and L3 are used as a reference point in a noise mixing position deletion process described later.

[Noise mixing position deletion processing]
Next, the control unit 10 deletes the noise mixing positions L1, L5, and L9 that are not synchronized with the noise mixing reference period from the noise mixing position storage unit 19 in the averaged white image data D2. For example, in FIG. 6 (4), noise mixing that does not synchronize with the noise mixing reference period “10” from the noise mixing positions L2 and L3, which are the reference points, toward the start side direction and the end side direction in the main scanning direction. The location is searched. As a result, since the noise mixing positions L1, L5, and L9 that are not synchronized with the noise mixing reference period “10” are determined, the noise mixing positions L1, L5, and L9 are deleted from the noise mixing position storage unit 19.

  As described above, the control unit 10 in the averaged white image data D2 includes the noise mixing positions L1 to L12 and the noise that is located immediately on the main scanning direction start side or the main scanning direction end side of the noise mixing positions L1 to L12. An interval in the main scanning direction from the mixing positions L1 to L12 is sequentially calculated in the main scanning direction. Then, the interval with the highest appearance frequency is set as the noise mixing reference period, and the noise mixing positions L1, L5, and L9 that are not synchronized with the noise mixing reference period are deleted from the noise mixing position storage unit 19 in the averaged white image data D2.

  Therefore, in the averaged white image data D2, only the noise mixing positions L2, L3, L4, L6, L7, L8, L10, L11, and L12 synchronized with the noise mixing reference period are finally sent to the noise mixing position storage unit 19. It will be memorized. Therefore, an accurate position where noise generated in synchronization with the line period is expected to be mixed in the image data can be known.

  As described above, the noise-mixing position detection device 1 can store, in the noise-mixing position storage unit 19, an accurate position where noise generated in synchronization with the line cycle is expected to be mixed in the image data. Therefore, the image reading apparatus A in which such a noise-containing position detection apparatus 1 is incorporated can perform the following noise removal processing.

  The noise-mixed position detection device 1 performs the “noise-mixed position detection basic process”, “noise-mixed position detection process 1”, and “noise-mixed position detection process 2” when the power is turned on and when the sleep mode is released. It is desirable to carry out at a fixed timing such as when changing the document reading speed. If the process of detecting the noise-mixing position is frequently performed, the original productivity of the image reading apparatus that generates image data and removes the pixel appearing at the noise-mixing position from the pixels forming the image data may be impaired. Because there is.

  FIG. 7 is a diagram for explaining the noise removal processing performed by the image reading apparatus A. FIG. 7A shows an example of the horizontal synchronization signal, and image data is generated in units of one line by the CCD 12 based on the horizontal synchronization signal. FIG. 7B shows an example of the mask period, and the pixel removal unit 21 masks the image data during the mask period. Such a mask period is a period for removing pixels appearing at the noise-mixed position detected by the noise-mixed position detection device 1.

  FIG. 7C shows an example of a noise mixing position of noise mixed in pixels arranged in a line corresponding to the leading end side of the document in the image data. FIG. 7D shows an example of a noise mixing position of noise mixed in pixels arranged in a line corresponding to the document trailing edge side in the image data.

  FIG. 7E shows an example of a formed image obtained by printing image data from which noise has not been removed on paper. FIG. 7F shows an example of a formed image obtained by printing image data from which noise has been removed on paper.

  The formed image shown in FIG. 7 (e) gradually deviates from the line cycle as a result of mixing noise that occurs periodically while gradually deviating from the line cycle every time image data is generated by the CCD 12 in units of one line. However, the periodically generated noise remains slightly oblique from the front end of the paper to the rear end of the paper in the main scanning direction (rightward in FIG. 7E).

  As described above, the average value removing unit 21 removes pixels appearing at the noise-mixing position from the image data during the mask period for removing the pixels appearing at the noise-mixing position detected by the noise-mixing position detection device 1. To do. As a method for removing the pixels, for example, image data generated by the CCD 12 and output from the memory block 13 is received in units of lines, and an electrical signal representing the received image data is masked.

  The noise generated while gradually deviating from the line period is substantially synchronized with the line period as shown in FIG. 7E, so that the pixel removing unit 21 appropriately removes such noise during the mask period. it can. Therefore, the image data from which noise has been removed is printed on a sheet, and a formed image as shown in FIG. 7F is obtained.

  Note that the pixel removal unit 21 applies to the pixel appearing at the noise mixing position detected by the noise mixing position detection device 1 and the pixel appearing at the noise mixing position in the image data generated by the CCD 12 in units of one line. A predetermined number of pixels close to the start side or end side in the main scanning direction may be removed. In this way, in the noise that occurs while deviating from the line period, the allowable period that deviates from the line period increases, so even when noise that deviates from the line period within the allowable period is mixed into the image data. Noise is removed appropriately.

A Image reading device 1 Noise mixing position detection device 10 Control unit 12 CCD
DESCRIPTION OF SYMBOLS 15 Average part 16 Threshold value memory | storage part 17 Noise mixing position detection part 18 Target value storage part 19 Noise mixing position memory | storage part 20 Average value removal part 21 Pixel removal part D1, D2 Average white image data P Average value X Noise

Claims (9)

A process of accumulating charges obtained by photoelectrically converting reflected light generated by reflecting light emitted from a light source as pixels and outputting the accumulated pixels as electric signals is performed in units of one line. An image data generating means for generating image data in units of one line;
A process of averaging the gradation values of each pixel appearing at the same position in the main scanning direction of each line in the white image data for a predetermined number of lines representing white generated by the image data generating means. Averaging means for sequentially calculating in the direction, obtaining average values, and generating average white image data arranged so that each average value corresponds to each pixel position in the main scanning direction;
Of the average values arranged in the averaged white image data generated by the averaging means, the difference between two adjacent average values is sequentially calculated in the main scanning direction, and the two calculated average values are calculated. On the basis of the difference between the averaged white image data and the position where the noise generated in synchronization with a line period representing a period in which the image data is generated by the image data generation unit in one line unit is included in the image data. Noise mixing position detection means for performing noise mixing position detection basic processing for detecting as a noise mixing position expected to be mixed in the image data generated by the generation means;
A noise-containing position detection device comprising: And further comprising storage means for storing the noise mixing position detected by the noise mixing position detection means,
The noise contamination position detection means performs the noise contamination position detection basic processing at a preset timing to detect the noise contamination position, and stores the detected noise contamination position in the storage means. The noise-mixing position detection device according to claim 1. The noise mixing position detecting means is
In the averaged white image data, a process of calculating a difference between two adjacent average values by subtracting an average value located next to the start side in the main scanning direction of each average value from each average value. Sequentially in the scanning direction,
In the averaged white image data, a position where an average value on the start side in the main scanning direction appears among two adjacent average values where the difference between the average values is N (where N is a positive numerical value; the same applies hereinafter). Is the noise start position,
In the averaged white image data, the difference between the average values, which appears for the first time after two adjacent average values with the average value difference equal to or greater than N appear, is M (where M is a negative numerical value; the same applies hereinafter) ) The position where the average value on the end side in the main scanning direction among the two adjacent average values which is less than 0 is equal to the noise end position.
In the averaged white image data, an average value including the noise start position and the noise end position and sandwiched between an average value appearing at the noise start position and an average value appearing at the noise end position is arranged. The noise-mixed position detection device according to claim 1, wherein the range thus detected is detected as the noise-mixed position. A threshold value setting means for receiving settings of a value representing the N and a value representing the M;
4. The noise mixing position according to claim 3, wherein the noise mixing position detecting unit detects the noise mixing position using a value representing the N and a value representing the M set in the threshold setting unit. Position detection device. Storage means for storing the noise mixing position detected by the noise mixing position detection means;
From the averaged white image data generated by the averaging means, an average value removing means for removing the average value appearing at the noise mixing position;
In the noise-removed averaged white image data from which the average value appearing at the noise mixing position has been removed, target value setting means for receiving setting of a target value of the maximum value of the difference between the two adjacent average values;
Averaged white image data generation processing for causing the averaging means to generate the averaged white image data;
Noise mixing position detection basic processing for causing the noise mixing position detection means to detect the noise mixing position;
A noise mixing position storage process for storing the noise mixing position detected by the noise mixing position detection means in the storage means;
The averaging unit generates the averaged white image data, and the average value removing unit causes the averaged white image data to appear at the noise mixing position stored in the storage unit. Noise-removed averaged white image data generation process that removes the average value
In the noise-removed averaged white image data, the process of calculating the difference between the two adjacent average values is sequentially performed in the main scanning direction, and the average that is the maximum among the sequentially calculated average value differences A maximum average value difference determination process for determining whether or not a value difference is equal to or less than the target value set in the target value setting means;
as well as,
A threshold update process for setting a value obtained by subtracting a preset value from the value representing N to the threshold setting means when the difference between the average values that is the maximum value is not less than or equal to the target value;
Control means for repeatedly performing the noise mixing position detection processing consisting of the above order until the difference between the average values that is the maximum value is equal to or less than the target value;
5. The noise-containing position detection device according to claim 1, comprising: Storage means for storing the noise mixing position detected by the noise mixing position detection means;
The noise mixing position stored in the storage unit is read, and in the averaged white image data generated by the averaging unit, the noise mixing position and the main scanning direction start side or main scanning direction of the noise mixing position A noise interval calculation process for sequentially calculating in the main scanning direction the interval in the main scanning direction with the noise mixing position located closest to the end side;
Among the intervals obtained by the noise interval calculation process, a noise mixing reference period determination process for determining an interval having the highest appearance frequency as a noise mixing reference period,
In the averaged white image data, a synchronous noise determination process for determining the noise mixing position synchronized with the noise mixing reference period,
as well as,
In the averaged white image data, the noise mixing position deletion process for deleting the noise mixing position that is not synchronized with the noise mixing reference period from the storage unit,
A control means for performing a noise mixing position detection process consisting of:
5. The noise-containing position detection device according to claim 1, comprising: The control means includes
The noise-containing position detection apparatus according to claim 5 or 6, wherein the noise-containing position detection processing is performed at a preset timing. A noise-containing position detection device according to any one of claims 1 to 7,
Pixel removal means for removing pixels appearing at the noise mixing position detected by the noise mixing position detection apparatus in the image data generated by the image data generation means included in the noise mixing position detection apparatus;
An image reading apparatus comprising: The pixel removing means includes
In image data generated in units of one line by the image data generating means, a main scanning direction with respect to a pixel appearing at a noise mixing position detected by the noise mixing position detecting device and a pixel appearing at the noise mixing position The image reading apparatus according to claim 8, wherein a predetermined number of pixels adjacent to a start side or an end side of the image are removed.

Images