JP6362075B2 - Image reading apparatus, image processing apparatus, and image processing method therefor - Google Patents

Description

  The present invention relates to an image reading apparatus, an image processing apparatus, and an image processing method thereof, and more particularly to an image reading apparatus, an image processing apparatus, and an image processing method for optically reading an image original.

  Conventionally, there is a scanner device (image reading device) that optically reads an object such as a document printed on plain paper or a photograph printed on photographic paper, and converts the read image into electronic data. Further, recent scanner apparatuses are not limited to a single independent form, but may take the form of a multi-function apparatus integrated with a recording apparatus and a communication function. In such a configuration, it is common for the multifunction device to have many functions such as a copying function and a fax in addition to digitization of a reading target.

  On the other hand, hard resource environments such as CPUs, memories, and LCDs used to realize these functions have become very high performance. Therefore, it is also an important factor to fully utilize them and provide a more comfortable use environment for the user. For example, in the case of image processing, it is an important requirement to realize high-level image processing without the user having advanced knowledge or performing complicated operations. A typical example is automation of image processing. As a result, it is possible to automatically analyze the target image and execute an optimum process corresponding to the analysis result.

  Conventionally, a technique disclosed in Patent Document 1 has been proposed as a technique for specifying a target area of a read image. Japanese Patent Application Laid-Open No. 2004-228561 proposes a technique for analyzing an image data obtained by scanning and reading an image original to detect an edge and determining the position of the original.

  By the way, in many cases, a document to be scanned is generally a predetermined size. For example, the size is A4 or A5. Such a document of a predetermined size is hereinafter referred to as a standard document.

  Here, a general reading method of a scanner apparatus that optically reads a document will be outlined.

  The scanner device includes a sensor unit in which a light emitting unit and a light receiving element are arranged, a document table on which a document is placed, and a pressure plate that presses the document. The scanner device moves the sensor unit relative to the document and reads the document. The relative movement direction of the sensor and the document is generally called a sub-scanning direction. A direction perpendicular to the moving direction is called a main scanning direction. A plurality of light receiving elements are arranged in the main scanning direction, and an image is read in the main scanning direction by electronically scanning these light receiving elements. In the reading, the original is irradiated with light from the light emitting unit, a part of the reflected light is received by the light receiving element, and the light received by the light receiving element is photoelectrically converted to generate an image signal. .

Japanese Patent Laid-Open No. 11-150655 (Japanese Patent Application No. 9-317453)

  When scanning a document, the end of the document tends to be an edge among components parallel to the main scanning direction of the document (the width direction of the document). This is because the irradiation light from the light source has directivity and a shadow due to the thickness of the document is formed. Therefore, if the shadow portion can be detected from the scanned image, the edge can be recognized. On the other hand, a component parallel to the leading end portion of the document and the sub-scanning direction (length direction of the document) in the width direction of the document does not easily become an edge. This is because the side surface of the document scatters light at the leading edge of the document, making it difficult to produce a shadow, and the directivity of the irradiation light is parallel to the width direction of the document in the length direction of the document, making it difficult to produce a shadow. As a result, the difference between the value of the image signal obtained by reading the pressure plate and the value of the image signal obtained by reading the document is small at the leading edge of the document and the edge portion in the length direction of the document, and it is difficult to detect the edge. .

  For this reason, in Patent Document 1, edge detection may be successful, but an erroneous determination may occur at the document leading edge or the document length end. Therefore, if an erroneous determination is made, the length of the original and the width of the original cannot be detected correctly, and finally an appropriate original size cannot be determined. For example, there may occur a case where an A5 size document is cut out even though an A4 size document is scanned.

  SUMMARY An advantage of some aspects of the invention is that it provides an image reading apparatus, an image processing apparatus, and an image processing method capable of accurately detecting a document size.

  In order to achieve the above object, an image reading apparatus of the present invention includes the following configuration.

That is, an image reading apparatus that reads an image of a document placed on a document table, a storage unit that stores the sizes of a plurality of standard documents, and a reading unit that optically reads the image of the document in the main scanning direction. the moving means for the reading means is moved in the sub-scanning direction, wherein while moving the reading means by the moving means based on the image data obtained by reading the document, a sub-scanning direction of the end of the document Acquisition means for acquiring the length of the edge along the main scanning direction, and a plurality of standard originals stored in the storage means based on the edge length of the original acquired by the acquisition means. of size, a plurality of sizes of candidates to be estimated as the size of the document, extracting means for extracting, area identified from the difference in the size of a plurality of candidates of the document extracted by the extraction means It relates, and if determining means for determining a document to the analyzing the image data area exists, when it is determined that the document on the region is present by the determining means, among the size of the plurality of candidates, When the size of the standard document having the edge length as the short side is determined as the size of the document, and the determination unit determines that the document does not exist in the area, And determining means for determining a size of a standard document having an edge length as a long side as the size of the document .

  Therefore, according to the present invention, when the original is, for example, a standard size, an original size candidate is estimated, and the size can be automatically and accurately determined from image processing.

1 is a schematic perspective view of an image reading apparatus according to an embodiment of the present invention. It is a block diagram which shows the internal structure of an image reading apparatus. 3 is a flowchart illustrating an outline of image processing executed by an image processing unit of the image reading apparatus. 4 is a flowchart illustrating image processing after image data is transferred to a PC. , FIG. 6 is a schematic diagram of an image obtained by placing an original having the same width and different length on an original table and reading an image. 6 is a flowchart showing image analysis of a document determination area. It is a figure which shows the image obtained by image analysis. It is a flowchart which shows the detail of the trimming process shown to step S408. It is a figure which shows a coupling | bonding process typically. It is a figure which shows an example of a small area removal process.

  Hereinafter, embodiments of the present invention will be described more specifically and in detail with reference to the accompanying drawings. In addition, the same code | symbol is attached | subjected to the already demonstrated part and duplication description is abbreviate | omitted. Also, the drawings cited below and the reference numerals of the drawings are shown as examples of the embodiments, and the variations of the embodiments according to the present invention are not limited thereby.

  In this specification, “recording” (sometimes referred to as “printing” or “printing”) is not limited to the case where significant information such as characters and graphics is formed, and is not significant. It also represents the case where an image, a pattern, a pattern, etc. are widely formed on a recording medium, or the medium is processed, regardless of whether it is manifested so that humans can perceive it visually. .

  In the following embodiments, a single-function image reading apparatus (scanner apparatus) will be described as an example, but the present invention is not limited thereto. For example, the present invention can be applied to a multi-function device (multifunction printer device, MFP device) having a recording device, a facsimile communication function, and the like.

<Description of Configuration of Image Reading Apparatus (Scanner Apparatus) (FIGS. 1-2)>
FIG. 1 is a schematic perspective view of an image reading apparatus according to an embodiment of the present invention.

  In FIG. 1, reference numeral 101 denotes a pressure-bonding plate for pressing a set document, 102 a document table on which a document is placed, 103 a document reading unit for reading a set document, and 104 an operation button.

  When the user places a document on the document table 102 and presses the operation button 104, the document reading unit 103 is driven by a motor (not shown) to optically read the document. At the time of reading a document, light is irradiated to the document from a light source installed in the document reading unit 103, and the reflected light is received by a light receiving element (not shown) of the document reading unit 103, and the received light is photoelectrically converted. Thus, an image signal is generated.

  FIG. 2 is a block diagram illustrating an internal configuration of the image reading apparatus.

  In FIG. 2, the CPU 201 controls the overall operation of the image reading apparatus by executing a control program stored in the ROM 202. Further, the CPU 201 executes the image processing program stored in the ROM 202 by using various values and mathematical formulas used for processing such as shading and size information such as portrait and landscape orientations of the standard document. The scan unit 203 reads the original image and outputs red (R), green (G), and blue (B) analog luminance data.

  The image processing unit 204 executes image processing such as correction of the value of the image signal digitized by A / D conversion and image encoding. If image data needs to be stored during image processing, the RAM 208 is temporarily used. The operation unit 205 detects a pressed state of the operation button 104 shown in FIG. 1 and controls each part of the apparatus. The drive unit 206 includes a driver circuit that performs motor control in order to operate the scan unit 203. The PC interface 207 is an interface with a PC that operates as a host device (not shown), and the image reading apparatus performs data transfer with the PC via the PC interface 207. If a buffer is required for data transfer, the RAM 208 is used.

  The scan unit 203 includes a light source unit and a plurality of light receiving elements, and the arrangement direction of the plurality of light receiving elements is a main scanning direction. Further, the direction in which the scan unit 203 moves by driving the motor is defined as the sub-scanning direction. When reading an image of a document, the light source unit is driven to irradiate the image document with light, and reflected light is received by a plurality of light receiving elements. The CPU 201 controls to electrically scan the plurality of light receiving elements in the main scanning direction, and the plurality of light receiving elements photoelectrically convert the received light according to the scanning to generate an image signal. The scan unit 203 scans in the main scanning direction and moves in the sub scanning direction to read the entire image of the document.

  Since the image reading is performed as described above, the document end portion refers to an end portion on the side where the scan unit 203 moves in the sub-scanning direction and the movement ends. Therefore, the length of the edge of the document end portion is the length of the document along the main scanning direction.

<Outline of image processing (FIGS. 3 to 4)>
FIG. 3 is a flowchart showing an outline of image processing executed by the image processing unit 204 of the image reading apparatus.

  According to the flowchart, image data obtained by reading a document with the scan unit 203 is A / D converted. In step S301, the image data is subjected to shading correction. Thereby, the white reference and the black reference prepared in advance are read, and correction is performed from the luminance value read for each image pickup device (light receiving device), thereby correcting variations in the image pickup device (light receiving device).

  In step S302, gamma correction is executed. This correction may be performed by holding a table in the ROM 202 in advance and referring to it, or may be performed by sequential calculation. After the processing in step S302, the image data is transferred to the PC via the PC interface.

  Although not shown as processing steps in FIG. 3, data compression may be performed to reduce the transfer amount before transferring image data. Further, whether to perform data compression may be appropriately switched depending on the transfer rate of the interface with the PC.

  FIG. 4 is a flowchart showing image processing after image data is transferred to the PC.

  In step S401, the image data transferred to the PC is developed in a RAM (not shown) in the PC, and in step S402, an edge portion of the document is detected. The edge detection method does not matter. For example, a Laplacian filter or the like may be used, or determination may be made in combination with the brightness and saturation of the document image.

  In the next step S403, it is evaluated whether or not four sides of the document can be detected appropriately. The evaluation method does not matter. If there are four detected edges and a closed region is constituted by the four sides, it is evaluated that the edge detection is successful, and the others may be failed. As a result of the evaluation, if the four sides can be confirmed in step S403, the process proceeds to step S407, the trimming process is executed, and the process ends. The trimming process will be described later.

  On the other hand, if the four sides cannot be determined, the process proceeds to step S404. In step S404, the length of the edge of the document end is measured. This length can be calculated by counting the number of pixels that make up the edge and using the scan resolution. For example, assume that a document is scanned in the sub-scanning direction with a resolution of 600 dpi and the length of the edge of the document end is 4960 pixels. At this time, since the length of one pixel is 25.4 [mm] / 600 [dpi], it can be seen that the edge length is about 210 [mm]. The document end portion length obtained in this way is compared with the end length list stored in advance in the memory of the PC. The end length list stores the width and length of a standard-size document. For example, an A4 document has a width of 210 [mm] and a length of 297 [mm]. For an A5 document, the width is 148 [mm] and the length is 210 [mm].

  In this way, for a standard size paper, the next standard size is determined by halving a document that is one size larger. Documents having the same width but different lengths are not limited to this. The standard document called letter size and legal size also have the same width but different lengths. If a standard document having a width or length corresponding to the length of the edge of the document end portion obtained from the scanned image is in the list, the size of the document can be narrowed down by selecting a candidate standard document size.

  Next, in step S405, it is checked whether or not the original is a standard original from the length of the edge of the original end obtained in step S404. Here, when it is determined that the original is not a standard original, the manual input of the size is necessary, and the process ends. On the other hand, if it is determined that the document is a standard document, the process proceeds to step S406 after narrowing down the candidate standard document size. In step S406, a process for estimating the direction in which the document is placed is performed. Details of this will be described later in <Image Analysis>.

  In step S406, the final document size is determined. The size determination process will be described in detail in <Size determination process> below. In step S407, a trimming process is executed. The trimming method will be described later in <Trimming process>.

  Here, the image processing performed in the image reading apparatus main body and the image processing in the PC have been described separately. Accordingly, in this case, the PC operates as an image processing apparatus. However, all the processing steps may be provided in the image reading apparatus and executed. This also applies to the case where the image reading apparatus is integrated with another device such as a recording apparatus to constitute an MFP apparatus.

<Size confirmation process>
Here, the document size determination process in step S406 of FIG. 4 will be described in detail.

  FIG. 5 and FIG. 6 are schematic diagrams of images obtained by placing images having the same width and different lengths on a document table and performing image reading.

FIG. 5 shows a state in which the crimping plate 501, the standard document 502, and the document end portion 503 indicated by diagonal lines are imaged. On the other hand, FIG. 6 shows a state where the standard document 602 is captured. As can be seen by comparing FIG. 5 and FIG. 6, the standard document 502 is a document that is half the size of the standard document 602. In addition, the lengths of the edges of both document end portions 503 are the same. From the length of the acquired edge of the document termination 503 at step S404, if the document size to be candidates is a standard document, either size and long side length of the edge (standard document 502), or short It is estimated that the size is a side (standard document 602). Now, it is impossible to determine whether the captured standard document is the standard document 502 or the standard document 602 only by the length of the edge of the document end portion 503.

  Accordingly, in step S406, the image is traced back by a length sufficient to cause the difference in the length of the standard document along the direction in which the document whose size is estimated is placed, and the difference between the standard document 502 and the standard document 602 is determined. Image analysis is performed on the obvious document determination area 504. Details of this image analysis will be described later. If it is determined that the document determination area 504 is outside the image area as a result of tracing back the image, the size of the placed document automatically becomes a small size. If an image area is found, image analysis is performed for that area. As a result of the analysis, if the document determination area 504 is a pressure plate, the size of the placed document can be determined to be the standard document 502. If the document determination area 504 is paper, the size of the placed document can be determined to be the standard document 602.

<Image analysis>
Since the purpose of this image analysis is to specify the size of the placed document, the area subject to image analysis is the area corresponding to the plain paper that is the document or the background. It is necessary to identify whether the area corresponds to the crimping plate.

  Plain paper has the following characteristics. Generally, paper is made from wood chips, and fibers are taken out during pulping. After the washing and bleaching steps, the fibers are arranged with directionality by the wire part and the press part in the paper making process. As a result, it has the characteristics of flow lines such as horizontal eyes and vertical eyes. In addition, the size of the fiber depends on the type of wood used as the material, and often has a size of several tens of μ × 1 to 3 mm. That is, as a feature of the fiber, it can be said that fine fine lines are arranged with directivity. This means that when image data is subjected to frequency analysis, it has a high frequency component as a feature from the viewpoint of frequency. At the same time, it means that the fiber has directionality. On the other hand, since the crimping plate is made of a chemical material with homogeneity, it does not have a high frequency component as a feature in terms of frequency. At the same time, it has no directionality. Therefore, the presence / absence of paper is determined by scanning and analyzing the original determination area 504 in FIGS. 5 to 6 with a resolution that can detect the fiber of the pulp.

  FIG. 7 is a flowchart showing image analysis of the document determination area 504.

  FIG. 8 is a diagram showing an image obtained by image analysis. In FIG. 8, the center of the image is the origin of the two-dimensional coordinates.

  In step S701, two-dimensional DFT (discrete Fourier transform) processing is performed on the image data of the target section in order to convert the image data into a spatial frequency. By this processing, an image having frequency characteristics as shown in FIG. 8A is obtained. In FIG. 8A, the origin indicates a DC component, the vertical axis (y-axis) indicates the vertical spatial frequency, and the horizontal axis (x-axis) indicates the horizontal spatial frequency. And it shows that it is a high frequency component, so that it leaves | separates from an origin in horizontal and vertical directions. A portion expressed in white in the figure is a power value (power spectrum) at a predetermined frequency component, and is plotted as white as the power value increases. FIG. 8A shows an example of conversion of the paper white section of the offset document.

  In step S702, binarization processing is performed on the obtained DFT-transformed image. As the binarization threshold value, a preset value may be used, or it may be determined based on an average value of images. FIG. 8B shows an image obtained by binarizing the image after the DFT conversion shown in FIG.

  In step S703, bold processing is performed on the DFT image binarized in step S702. This is executed based on a preset bold amount. FIG. 8C shows an image obtained as a result of performing a five-pixel bold process on a white pixel having a power value (power spectrum).

  In step S704, the isolated point is further removed from the image obtained in step S703. This is a process of scanning an image, detecting the number of pixels corresponding to a white area, and painting an area with an area equal to or smaller than a preset threshold in black. The threshold is set in consideration of the bold amount. In this embodiment, it is assumed that 2 to 3 isolated points are removed, and 350 pixels are set as the threshold value. FIG. 8D shows an image obtained by isolated point removal.

  In step S705 to step S706, processing for detecting the feature amount of the fiber is executed from the image created up to step S704. In the image shown in FIG. 8D obtained in step S704, determination is performed in consideration of the above-described features. In step S705, it is checked whether or not a power value is detected for a frequency component above a certain level, and whether or not a feature value is detected. Here, when a power value is not detected for a high-frequency component of a certain level or more, the process proceeds to step S709, and the region is determined as a non-fiber section. On the other hand, if a power value is detected for a certain high-frequency component, the process proceeds to step S706.

  In step S706, the flatness ratio is acquired for the image shown in FIG. That is, the white pixel shown in FIG. 8D is regarded as an ellipse, and the major axis length a and the minor axis length b are obtained as shown in FIG. Both values are obtained by counting the number of white pixels. Then, the flatness is calculated using the obtained value. Further, in step S707, it is checked whether or not the calculated flatness ratio is a certain value or more. Here, when it is determined that the flatness is less than a certain value, the process proceeds to step S709, and the region is determined as a non-fiber section. On the other hand, when it is determined that the flatness ratio is equal to or greater than a certain value, the process proceeds to step S708, and the region is determined as a fiber section.

  That is, if it is a fiber section, it is determined that there is a sheet of an image original, and if it is a non-fiber section, it is determined that it is a pressure-bonding plate as a background.

  As described above, it can be determined whether or not there is a document in the document determination area 504. As can be seen from the examples of FIGS. 5 to 6, if there is a document in the document determination area 504, the placed document is the standard document 602, and if there is no document in the document determination area 504, the placed document is This is a standard document 502. At this time, it is determined that the placed document is the standard document 502 or the standard document 602, and therefore the direction in which the document is placed can be determined.

<Trimming>
FIG. 9 is a flowchart showing details of the trimming process shown in step S408.

  First, in step S901, the pixel value of the edge detected in step S402 is acquired. Next, in step S902, the result of determination as an edge pixel is combined in a predetermined area unit. An example of the combining process will be described with reference to the drawings.

  FIG. 10 is a diagram schematically showing the joining process.

  In the combining process, as shown in FIG. 10A, a 3 × 3 pixel matrix, a total of 9 pixels, is used to count how many of these pixels are determined to be originals. If the number of pixels determined to be a document is equal to or greater than a predetermined number, all the pixels in the matrix, that is, 9 pixels are determined to be pixels belonging to the document. On the other hand, if the number of pixels is less than the predetermined number, nothing is done.

  The example shown in FIG. 10B is a case where the predetermined number is set to “4”. That is, if there are 4 or more pixels belonging to the document in a 3 × 3 pixel matrix, all the matrices are determined to be pixels belonging to the document, and the pixel values are changed.

  Note that the size of the matrix is just an example, and it is possible to realize stronger combination processing by increasing the number of processing units or repeatedly executing such processing.

  This combination process is intended to summarize the determination results having variations in units of pixels in the previous processes to some extent, and the method is limited to this example as long as the process meets the purpose. It is not a thing.

  In step S903, the isolated points of the small area remaining after the execution of the combining process in step S902 are removed. There are various methods for this as well. Measure the area (number) of adjacent pixels as a unit, and if the area is below a certain level, leave all of them as non-documents, or leave only the largest area. The method of removing is mentioned.

  FIG. 11 is a diagram illustrating an example of the small area removal process.

  In FIG. 11, reference numeral 1101 denotes the entire image document to be read. The dark area where the shadow is applied is a block portion determined to be a document, and the white region is a block portion determined not to be a document. ing. FIG. 11A shows that a small chunk portion determined as a document exists also in a portion other than a large chunk portion determined as a document. FIG. 11B shows an image obtained by applying the small area removing process to such an image, assuming that the above-described area of a certain area is not a document. By such a small area removal process, a portion determined to be a small area original that was noisy is deleted, and only the original area can be extracted.

  In step S904, the coordinate position of the region extracted in the processing up to step S903 is specified. First, the positions of the four corners of the document are calculated. FIG. 11B shows four points A, B, C, and D. For the area obtained in step S903, the maximum value and the minimum value are calculated in the coordinate system shown in FIG. 8, and the coordinate values of four points are calculated from the values. Furthermore, if an inclined document is also taken into consideration, the coordinates may be obtained again by further searching near the four corners.

  By executing the above processing, it is determined whether there is a document such as paper in the document determination area 504 or an image of a pressure plate.

  Note that in steps S702 to 704 in FIG. 7, in order to perform efficient detection, a process that highlights the features of the image is performed. However, if another process can achieve the same purpose, Processing may be used.

  As described above, according to this embodiment, it is possible to place a document on a document table either vertically or horizontally, or on any position on the document table. In the case of a manuscript, the size can be determined as one.

  The present invention can also be realized by executing the following processing. That is, software (program) for realizing the functions of the above-described embodiments is supplied to a system or apparatus via a network or various storage media, and a computer (or CPU, MPU, etc.) of the system or apparatus reads the program. It is a process to be executed.

Claims (13)

An image reading apparatus for reading an image of a document placed on a document table,
Storage means for storing sizes of a plurality of standard documents;
Reading means for optically reading the image of the document in the main scanning direction ;
Moving means for moving the reading means in the sub-scanning direction ;
Based on the image data obtained by reading the original while moving the reading means by the moving means, the length along the main scanning direction of the edge that is the terminal portion of the original in the sub- scanning direction is acquired. Acquisition means to
Based on the length of the edge of the acquired document by the acquisition unit, of the size of a plurality of standard document stored in the storage means, a plurality of sizes of candidates to be estimated as the size of the document, extracted Extraction means to
A determination unit that analyzes the image data and determines whether or not a document exists in the region with respect to a region specified by a difference in size of the plurality of candidates of the document extracted by the extraction unit;
When the determination unit determines that a document is present in the area, a standard document size having a short side as the edge length is determined as the document size among the plurality of candidate sizes, and the determination is performed. Determining means for determining, as a size of the original, a size of a standard original having a long side of the edge length among the plurality of candidate sizes when the means determines that no original exists in the area ; An image reading apparatus comprising: The reading means is an image reading apparatus according to claim 1 in which a plurality of light receiving elements, characterized that you have been arranged in the main scanning direction. Wherein the storage unit, the image reading apparatus according to claim 1 or 2, characterized in that storing a size of a plurality of standard document each of the short sides and long sides. The extraction means includes
When the length of the edge of the document that is the short side or long side of the standard document, the size of the standard document that corresponds to the short side or long side of each of the plurality of standard document stored in the storage means, the candidate The image reading apparatus according to claim 3 , wherein the image reading device is extracted as a size of the image. 5. The image reading apparatus according to claim 4, wherein the determination unit determines whether or not a document exists in the area when the size of the standard document is extracted by the extraction unit.   The image reading apparatus according to claim 1, further comprising a pressure-bonding plate for pressure-bonding the document placed on the document table. The determination means performs image processing on the image data with respect to the region, performs frequency analysis, and from the result of the frequency analysis, whether the image data corresponding to the region corresponds to the crimping plate, or the image reading apparatus according to claim 6, characterized that you determining whether they correspond to the document. Said determining means further to the image data, the image reading apparatus according to any one of claims 1 to 7, characterized in that to facilities the trimming process of extracting an area of the document from the image. In the trimming process, in the image data, a pixel belonging to the document is determined, a combining process for combining the pixels belonging to the document is executed, and an area combined by the combining process is extracted as an area of the document. The image reading apparatus according to claim 8. A first region in which pixels are combined by the combining processing; and a second region in which pixels are combined by the combining processing and are smaller than the first region and not combined with the first region. 10. The image according to claim 9, wherein when there is an area, the trimming process extracts the first area as the area of the original and does not extract the second area as the area of the original. Reader. Image reading apparatus is obtained by reading the document placed on the document table by using a moving means for moving the reading means said a reading means for reading an image of an original optically in the main scanning direction in the sub-scanning direction An image processing apparatus that inputs image data and processes the image,
Storage means for storing sizes of a plurality of standard documents;
Acquiring means for acquiring a length along a main scanning direction of an edge which is a terminal portion in a sub-scanning direction of the document based on the image data;
Based on the length of the edge of the acquired document by the acquisition unit, of the size of a plurality of standard document stored in the storage means, a plurality of sizes of candidates to be estimated as the size of the document, extracted Extraction means to
A determination unit that analyzes the image data and determines whether or not a document exists in the region with respect to a region specified by a difference in size of the plurality of candidates of the document extracted by the extraction unit;
When the determination unit determines that a document is present in the area, a standard document size having a short side as the edge length is determined as the document size among the plurality of candidate sizes, and the determination is performed. Determining means for determining, as a size of the original, a size of a standard original having a long side of the edge length among the plurality of candidate sizes when the means determines that no original exists in the area ; An image processing apparatus comprising: The program for functioning a computer as each means of the image processing apparatus of Claim 11 . Image reading apparatus is obtained by reading the document placed on the document table by using a moving means for moving the reading means said a reading means for reading an image of an original optically in the main scanning direction in the sub-scanning direction It was to input image data a images processing method for image processing,
An acquisition step of acquiring a length along the main scanning direction of an edge which is a terminal portion of the original in the sub-scanning direction based on the image data;
Extraction that extracts a plurality of candidate sizes estimated as the document size out of a plurality of standard document sizes stored in a memory based on the edge length of the document acquired in the acquisition step Process,
A determination step of analyzing whether the document exists in the region by analyzing the image data with respect to a region specified by a difference in size of the plurality of candidates of the document extracted in the extraction step;
When it is determined in the determination step that a document is present in the area, a size of a standard document having a short side of the edge length is determined as the document size among the plurality of candidate sizes, and the determination A determination step of determining, as a size of the original, a size of a standard original having a long side of the edge length among the plurality of candidate sizes when it is determined that no original exists in the region in the process; An image processing method comprising: