JPH0814845B2 - Area detection method and area detection apparatus - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a detection method and apparatus for performing area detection (for example, original area detection or line area detection) in image processing.

(Prior Art) In a character recognition device, an electronic copying machine, a facsimile device, and other image processing devices, image processing is performed by paying attention to only image data in a specific region (for example, a document region or a line region) of image data in the field of view. Doing is very effective as a means for improving the processing speed and as a means for performing individual processing for each area. Examples of this type of area detection method include, for example, JP-A-60-194868 and JP-A-58-12.
The method disclosed in Japanese Patent No. 3169 has been conventionally proposed.

In the former method, the reading area (field of view) in which the document or the form is arranged is scanned by the image sensor, and the change in the output of the image sensor when scanning in the main scanning direction is detected. Then, the coordinates of the pixel where the output has changed (edge coordinates) are used as the coordinates for defining the area, and the area of the document or the form placed in the reading area is determined by the coordinate series created from the plurality of edge coordinates. Make regulations.

Further, in the latter method, first, the reading area in which forms and the like are arranged is scanned by the image sensor, and the quantized image data is stored in the image memory via the image sensor. Thereafter, the stored image data is divided and scanned in the horizontal direction or the vertical direction, and a peripheral distribution is created for each of the divided image data. Then, based on the peripheral distribution, the area of the character line to be read or recognized by the character reading device or the recognition device is detected.

(Problems to be Solved by the Invention) However, in the former conventional method described above, in addition to the register for detecting the edge coordinates, a memory for storing the coordinate series is required, and moreover, as a memory for the coordinate series. A large capacity was needed. Therefore, there is a problem that the scale of the hardware becomes large. Further, when detecting the size or inclination of a document or the like placed in the reading area, if the edge (edge) of the document or the like is uneven, there is a problem that these detections cannot be performed accurately.

Furthermore, in the latter conventional method, in addition to an image memory for storing image data, a large-capacity memory for creating and storing the peripheral distribution is required. Therefore, there is a problem that the hardware scale becomes large. Further, since the reading area is scanned and the image data is temporarily stored in the memory and then the peripheral distribution is created, there is a problem that the processing speed is slow.

An object of the present invention is to solve the conventional problems described above,
An object of the present invention is to provide an area detection method and apparatus capable of reducing the scale of hardware and at the same time performing the desired area detection with high accuracy.

(Means for Solving the Problem) In order to achieve this object, the method invention of the present application uses quantized image data for area detection in a two-dimensional coordinate system, The X coordinate and the Y coordinate are given to, and the calculated value αX + βY is calculated using the X coordinate and the Y coordinate and at least one set of specific α value and β value, and the calculated value of the pixel having the predetermined pixel value is calculated. It is characterized in that the maximum calculated value and the minimum calculated value are detected, and the X and Y coordinates of the pixel giving the maximum calculated value and the minimum calculated value are set as area coordinates, and the detection area is defined by these area coordinates.

In carrying out this method invention, as α and β (α = β
It is preferable to use at least one set of α and β of = 1) and (α = 1, β = −1).

Further, in carrying out the method invention, it is preferable that the detection area is a manuscript area, image data capable of distinguishing the manuscript area from the background area is obtained, and a predetermined pixel value is set to a pixel value which means a background of the manuscript. is there.

Further, in implementing the method invention, the detection area is a row area, image data capable of distinguishing the character line portion of the line area from the background area is obtained, and a predetermined pixel value is set as a pixel value meaning the character line portion. Is preferred.

Further, in carrying out the method invention, it is preferable that the pixel value of the image data is determined through a filter for removing noise pixels or detecting a specific figure.

Further, the apparatus invention of this application is such that a reading unit for outputting quantized image data for area detection, and an X coordinate for generating an X coordinate of a pixel of the image data for expressing in a two-dimensional coordinate system of the image data. Generating means, Y coordinate generating means for generating the Y coordinate of the pixel of the image data for expressing in the two-dimensional coordinate system of the image data, X coordinate and Y coordinate, and at least one set of specific α value and β value And a calculation detecting means for calculating the calculation value αX + βY by using and to detect the maximum calculation value and the minimum calculation value regarding the calculation value of the pixel having a predetermined pixel value, and the maximum calculation value as the area coordinates defining the detection area. The maximum value coordinate storage means for storing the pixel coordinates and the minimum value coordinate storage means for storing the pixel coordinates giving the minimum calculated value as the area coordinates are provided.

Further, in carrying out the present invention, the calculation detecting means includes a calculating means for calculating a calculated value, a maximum value detecting means for detecting a maximum calculated value of a pixel having a predetermined pixel value, and a pixel having a predetermined pixel value. It is preferable to have a configuration including a minimum value detecting means for detecting the minimum calculated value of.

Further, in carrying out the invention of this device, as α and β (α
= Β = 1) and (α = 1, β = −1), it is preferable to use at least one set of α and β.

Further, in carrying out the present invention, it is preferable that the detection area is a manuscript area, image data capable of distinguishing the manuscript area from the background area is obtained, and a predetermined pixel value is set to a pixel value which means a background of the manuscript. is there.

Further, in carrying out the invention of the apparatus, the detection area is set as a row area, image data capable of distinguishing the character line portion of the line area from the background area is obtained, and a predetermined pixel value is set as a pixel value meaning the character line portion. Is preferred.

Further, in carrying out the apparatus invention, it is preferable that the reading section is provided with a filter for removing noise pixels or for detecting a specific figure.

(Operation) According to the method and apparatus invention described above, the X and Y coordinates are given to the pixel of the two-dimensional image data, and the maximum value and the minimum value of the calculated value αX + βY regarding the pixel having the predetermined pixel value are detected. Then, the X and Y coordinates of the pixel that gives the maximum and minimum calculated values are set as area coordinates, and the detection area is defined by these area coordinates.

The above-described procedure for defining the detection area in this way is a very simple procedure, and therefore the apparatus configuration can be simplified.

Further, since the above-mentioned procedure can be performed in parallel with the scanning of the reading unit, the processing time for area detection can be shortened.

(Embodiment) An embodiment of a method invention and an apparatus invention of this application will be described below with reference to the drawings. It should be noted that the drawings are only schematically shown to the extent that these inventions can be understood, and therefore the constitution of each component, the flow of data and the numerical conditions are not necessarily limited to the illustrated examples.

First Embodiment <Explanation of Device Invention> (Outline of Device Configuration) FIG. 1 is a functional block diagram for explaining an embodiment of the device invention.

As shown in the figure, the area detection apparatus of this embodiment uses a reading unit 10 that outputs quantized image data for area detection and a calculated value αX + βY (α and β are arbitrary constants). And an arbitrary suitable number (for example, two) of area detecting means 12 and 14 for detecting area coordinates for defining the detection area. In this embodiment, for example, (α
= Β = 1) and (α = 1, β = −1) two sets of α and β
Since the value is used, the two area detecting means 12 and 14 are provided.

The number of area detecting means is α and β for area detection.
It can be arbitrarily changed depending on how many sets of values are used. For example, when n sets of α and β values are used, n area detection means is provided and α and / or β are set for each area detection means. It may be set to any suitable value that is different from the above. Also,
In the case of detecting a document area or other areas having a substantially rectangular shape, (α = β = 1) and / or (α = 1, β = −1)
Is preferably used.

(Reading Unit) The photoelectric conversion unit 10a of the reading unit 10 is obtained by the light applied to the reading target, for example, the reading area in which the document is arranged.
The density information L of the reading area is input, and this density information L is converted into quantized original image data of, for example, a binary digital signal. The original image data is obtained by scanning the reading area line-sequentially, for example.

FIG. 2 is a diagram showing an example of original image data relating to the entire surface of the reading area. In the figure, 20 is the original image data of the entire reading area in which the original is placed, 20a is the image data of the original area (original data), and 20b is the image data of the background area (background data) that is the background of the original, The original image data 20 is composed of original data 20a and background data 20b. Further, a, b, c and d respectively indicate corner points of a substantially rectangular original. The original data 20a is the original image data 2
The image data of the white area surrounded by the one-dot chain line of 0 and the background data 20b indicate the image data of the hatched area of the original image data 20 excluding the original data.

The original image data 20 is represented by a two-dimensional coordinate system (XY coordinate system) whose origin (0,0) is the upper left corner point (pixel at the upper left corner) of this data 20, and the main direction is the X axis direction. The scanning direction and the axial direction of the Y axis are the sub-scanning direction.

In this embodiment, the original image data 20 output from the photoelectric conversion unit 10a of the reading unit 10 is used as image data for area detection. Moreover, the detection area is set as the original area, and image data that can distinguish the original area and the background area is obtained for detecting the original area. In the image data in which the original area and the background area can be distinguished, for example, the reading area is set wider than the original, and the background of the original is white and the background data 20b is represented by black bits (pixel value "1"). By
You can get it. In this case, the white bit (pixel value "0") of the original data 20a becomes a ground pixel which means the background of the original, and the black bit of the original data 20a means an information pixel which means information such as characters and figures written on the original. Or, it becomes a noise pixel caused by stains, breaks, tears, etc. of the original.

(Area Detecting Means) The area detecting means 12 is an X-coordinate generating means 16 for generating X-coordinates of pixels of image data for expressing image data for area detection in a two-dimensional coordinate system, and an image for area detection. Y coordinate generating means 18 for generating the Y coordinate of the image data for expressing the data in the two-dimensional coordinate system, and the X, Y coordinates and a specific α.
And the β value to calculate the calculated value αX + βY to detect the maximum calculated value and the minimum calculated value regarding the calculated value of the pixel having the predetermined pixel value, and the maximum as the area coordinates defining the detection area. A maximum value coordinate storage means 122 for storing the coordinates of the pixel giving the calculated value and a minimum value coordinate storage means 123 for storing the coordinates of the pixel giving the minimum calculated value as the area coordinates.

Then, the calculation detecting means 121 of the area detecting means 12 calculates the calculated value αX + βY, and the maximum value detecting means 121b that detects the maximum calculated value of a pixel having a predetermined pixel value.
And a minimum value detecting means 121c for detecting the minimum calculated value of a pixel having a predetermined pixel value.

The area detecting means 14 is similar to the area detecting means 12 in that
Coordinate generating means 16, Y coordinate generating means 18, calculation detecting means
141, maximum value coordinate storage means 142, minimum value coordinate storage means
143, and the calculation detection means 141 is the calculation means 141a.
A maximum value detecting means 141b and a minimum value detecting means 141c.

The area detecting means 12 and 14 are provided with common coordinate generating means 16 and 18.

In this example, there are two sets of α and β values, ie (α = β =
1) and (α = 1, β = −1) are used, the maximum and minimum calculated values for the calculated value X + Y are calculated by the calculation detection means 121, and the maximum and minimum calculated values for the calculated value XY are calculated by the calculation detection means 141. To detect.

In the above-mentioned area detecting means 12 and 14, for example, the calculating means 121a is provided with an adding circuit, the calculating means 141a is provided with a subtracting circuit, and the detecting means 121b, 121c, 141b and 141c are provided as comparators and registers, respectively. In addition, the coordinate storage means 12
Each of 2, 123, 142, and 143 can be configured with a register circuit.

<Explanation of Method Invention and Operation of Device Invention> FIGS. 3A to 3B are diagrams for explaining the operation of the embodiment of the method invention and the embodiment of the device invention.

Hereinafter, an example in which the method invention of the present application is applied to the above-described apparatus invention will be described with reference to FIGS. 1 and 2 in addition to FIGS. 3A and 3B.

As a method for detecting the maximum calculated value and the minimum calculated value for a pixel having a predetermined pixel value, for example, the following methods and can be considered. The calculation means 121a, 141a calculates a calculation value for all pixels of the image data for area detection, and the calculation values input from the means 121a, 141a in the detection means 121b, 121c and 141b, 141c have a predetermined pixel value. A method of determining whether or not the pixel is a pixel and detecting the minimum and maximum calculated values only from the calculated values of pixels having a predetermined pixel value based on this determination result, X, Y of the pixels input in the calculating means 121a, 141a It is determined whether or not the coordinates are the X and Y coordinates of a pixel having a predetermined pixel value, and the calculation value is calculated only for the pixel having the predetermined pixel value based on the result of the determination.
In the detection means 121b, 121c and 141b, 141c means 121a,
A method of detecting the maximum and minimum calculated values of the calculated values input from 141a and using the maximum and minimum calculated values as the maximum and minimum calculated values of a pixel having a predetermined pixel value. Although any of the above methods may be used, detection will be performed by the above method in the following description.

I: Description Focusing on Area Detection Means 12 * Steps (1) to (3) When the reading unit 10 starts scanning the reading region where the document is arranged (step (1)), the reading unit 10 is quantized. ,
Since the original image data 20 is output as the image data for area detection in a scanning sequence for each pixel, the maximum value detecting means 121b and the minimum value detecting means 121c input the original image data 20 for each pixel. At the same time, the X coordinate (main scanning) generating means 16 and the Y coordinate (sub scanning) generating means 18 generate the X and Y coordinates of the pixels of the original image data 20 in synchronization with the output of the reading section 10. As a result, the original image data 20 to which the X and Y coordinates are given for each pixel, that is, the image data for area detection expressed in the two-dimensional coordinate system is obtained. Coordinate generation means 16,
The X and Y coordinates from 18 are input to the calculation means 121a, the maximum value coordinate storage means 122 and the minimum value coordinate storage means 123. (Step (2)) When the calculating means 121a inputs X and Y coordinates, these X and Y coordinates are input.
The calculated value X + Y is calculated from the coordinates, and the calculated value is output to the maximum value detecting means 121b and the minimum value detecting means 121c (step (3)).

The maximum value detecting means 12 is obtained by the steps (2) to (3).
1b is the original image data 20 and the calculated value X + Y, the minimum value detection means 121c is the original image data 20 and the calculated value X + Y, the maximum value coordinate storage means 122 is the X and Y coordinates, and the minimum value coordinate storage means 123 is the X value. , Y coordinates are input for each pixel.

Then, the area detecting means 12 performs the step (4) described later,
The determination of (5a), (5b) or (7) is repeated for each pixel, and the operation according to the determination result is performed.

* Step (4) The maximum value detecting means 121b and the minimum value detecting means 121c judge for each pixel whether or not the input original image data 20 is a pixel having a predetermined pixel value. (Step (4)) When detecting the original area, it is determined whether or not it is a pixel (ground pixel) that means the background of the original, and in this embodiment, it is a white bit pixel. Find the maximum and minimum calculated values for.

When the pixel value does not have a predetermined pixel value, the maximum value detection means 121b, the maximum value coordinate storage means 122, the minimum value detection means 121c, and the minimum value coordinate storage means 123 do not operate.

The detecting means 121b and 121c make a judgment in step (7) after step (4).

When the pixel value has a predetermined pixel value, the maximum value detection means 121b determines the comparison of the comparison value and the calculated value (step (5)) after step (4), and the minimum value detection means 121c performs the step (4). Then, the comparison value and the calculated value are compared (step (5b)).

* Step (5a) When the calculated value is larger than or equal to the comparison value, the maximum value detecting means 121b replaces the previously stored comparison value, and is larger than or equal to the comparison value. The same calculated value is stored as a new comparison value (rewriting the comparison value), and at the same time, the set pulse is output to the maximum value coordinate storage means 122. Maximum value coordinate storage means with set pulse input 12
2 stores the X and Y coordinates output from the X coordinate generating means 16 and the Y coordinate generating means 18 (rewriting of the X and Y coordinates).
(Step (6)) When the calculated value is smaller than the comparison value, the maximum value detecting means 121b and the maximum value coordinate storing means 122 do not operate.

In any case, the maximum value detecting means makes the judgment of step (7) after step (5a).

The maximum value detecting means 122 may use, as the initial value of the stored comparison value, for example, a value smaller than the value that can be taken as the calculated value αX + βY. For example, when α = β = 1 and the original image data 20 is divided into pixels of 2048 rows and 3072 columns (thus 0 ≦ X ≦ 3071 and 0 ≦ Y ≦ 2047),
For example, -1 can be set as the initial value of the comparison value. Further, the comparison value and the X and Y coordinates are rewritten only when the calculated value αX + βY is larger than the comparison value, and when the calculated value is smaller than or equal to the comparison value, the maximum value detecting means 121b. The X and Y coordinates of the comparison value and maximum value coordinate storage means 122 may not be rewritten. Any suitable numerical value may be used as the initial values of the X and Y coordinates of the maximum value coordinate storage means 122.

* Step (5b) When the calculated value is smaller than the comparison value: The minimum value detecting means 121c stores the calculated value smaller than the comparison value as the comparison value instead of the previously stored comparison value (rewriting the comparison value). At the same time, a set pulse is output to the minimum value coordinate storage means 123. Minimum value coordinate storage means 123
Receives a set pulse, the X coordinate generating means 16 and Y
The X and Y coordinates output from the coordinate generating means 18 are stored (rewriting of the X and Y coordinates). (Step (6)) When the calculated value is larger than or equal to the comparison value, the minimum value detecting means 121c and the minimum value coordinate storing means 123 do not operate.

In any case, the minimum value detecting means 121c makes the judgment of step (7) after step (5b).

As the initial value of the comparison value stored in the minimum value detecting means 121c, for example, a value larger than the value that can be taken as the calculated value αX + βY may be used. For example, when α = β = 1 and the original image data 20 is divided into pixels of 2048 rows and 3072 columns (thus 0 ≦ X ≦ 3071 and 0 ≦ Y ≦ 2047), for example, 5119 is set as the initial value of the comparison value. Can be When the calculated value αX + βY is smaller than or equal to the comparison value, the comparison value and the X and Y coordinates are rewritten, and when the calculated value αX + βY is larger than the comparison value, the comparison value and the X and Y coordinates are rewritten. It may not be performed.
Any suitable numerical value may be used as the initial values of the X and Y coordinates of the minimum value coordinate storage means 123.

* Step (7) When the scanning of the reading area is completed, the maximum value detecting means 121b and the minimum value detecting means 121c complete the scanning of the reading area and complete the processing for all the pixels, and output the X and Y coordinate output signals to the maximum. It outputs to the value coordinate storage means 122 and the minimum value coordinate storage means 123. The maximum value coordinate storage means 122 and the minimum value coordinate storage means 123, which have received this output signal, output the stored X and Y coordinates as area coordinates for defining the original area. At the same time, the maximum value detecting means 121b and the minimum value detecting means 121c initialize the comparison value. When the processing is completed for all the pixels, the X and Y coordinates stored (stored) in the maximum value coordinate storage means 122 and the minimum value coordinate storage means 123 give the maximum and minimum calculated values, ie, the X and Y coordinates of the pixel. Area coordinates. (Step (8)) In the area detecting means 12, since α = β = 1,
When the processing is completed for all pixels, the X and Y coordinates stored in the maximum value coordinate storage means 122 become the coordinates of the corner point c, and the X stored in the minimum value coordinate storage means 123,
The Y coordinate is the X, Y coordinate of the corner point a.

When the scanning of the reading area does not end, the maximum value detecting means 121b and the minimum value detecting means 121c do not end the scanning of the reading area. Therefore, if the processing has not ended for all the pixels of the original image data 20, the original image data 20 The remaining pixels are subjected to the judgment of step (4), (5a), (5b) or (7), and the operation is performed according to the judgment result.

Since the area detecting means 14 performs the same operation as the area detecting means 12 in parallel with the operation of the area detecting means 12 described above,
The description of the operation of the area detecting means 14 is omitted.

II: Principle explanation of area occupant detection FIG. 4 is a diagram for explaining the principle of area coordinate detection. Hereinafter, description will be given with reference to FIG.

Assuming a straight line αX + βY = C passing through the original data 20a on the original image data 20, consider a straight line group αX + βY = C in which the inclination is fixed to a constant value and the value of C has various values. Then, a straight line passing through the corner point can be found in the straight line group passing through the document data 20a. At this time, the value of C of the straight line passing through the corner point is the minimum value or the maximum value in the straight line group. Becomes Therefore, the value of C, that is, the calculated value αX + βY is calculated for each pixel of the document data 20a, and the calculated value αX + βY is calculated.
It can be understood that the X and Y coordinates of the pixel having the maximum value or the minimum value may be used as the area coordinates.

For example alpha = beta = 1 straight lines passing through the area of the document data 20a as L: X + Y = thinking C _1, smaller to Yuku the near linear cage to the corner point a value of C ₁ is C ₁ obtained It can be understood that a straight line L ₁ passing through the corner point a is obtained when the value becomes the minimum. When the value of C ₁ is further increased, a straight line closer to the corner point c is obtained, and when the value of C ₁ is maximum, a straight line L _n passing through the corner point c is obtained. Therefore, among the pixels of the document data 20a, the coordinate giving the minimum value of the calculated value X + Y is the corner point a and the coordinate giving the maximum value is the corner point c.
It can be understood that the coordinates become, that is, the area coordinates.

Similarly, considering a straight line group M: XY = C ₂ that passes through the area of the document data 20a with α = 1 and β = -1, for example, when the value of C ₂ is the minimum, the angle point d is passed. Straight line M _m
And a straight line M ₁ passing through the point b when the value of C ₂ is maximum
Is obtained. Therefore, it is understood that among the pixels of the document data 20a, the coordinates of the pixel giving the minimum value of the calculated values XY are the coordinates of the corner point d and the coordinates of the pixel giving the maximum value are the coordinates of the corner point b, that is, the area coordinates. I can.

The detection area is a rectangular area defined by the coordinates of the corner points a, b, c, and d, that is, a rectangular area having these corner points.

In this embodiment, the coordinate generating means 16 and 18 are counters, the calculating means 121a is an adder and the calculating means 141a is a subtracter, and the maximum value detecting means 121b and 141b are a comparator and a register. Therefore, the minimum value detecting means 121c, 141c are provided with a comparator and a register, and further the coordinate holding means 122, 123, 14 are provided.
Since 2 and 143 are configured with registers, the hardware scale can be made smaller than before.

<Modification> In the first embodiment, the case where the corner point is detected by using two groups of straight lines L and M having different inclinations by using two sets of α and β values has been described. It is also possible to detect a region having a substantially polygonal shape (almost n-gonal shape) by using α and β values and thus using three or more straight line groups having different inclinations. In this case, each set of α
The maximum calculated value and the minimum calculated value for the .beta. Value and the .beta. The detection area is an n-gonal area defined by the coordinates of the detected n corner points, that is, an n-gonal area having n corner points. In this case, the device configuration may be the same as that of the above-described first embodiment except that the number of area detecting means provided is the same as the number of pairs of α and β values used for detecting the corner points. Good.

When detecting an area having a substantially rectangular shape, such as detecting an original area, the coordinates of two pixels giving the maximum and minimum calculated values are detected using one set of α and β values. You can The coordinates of the corner points obtained in this case are the coordinates of corner points at diagonal positions such as corner points c and a. The detection area is an area defined by the coordinates of the corner points at these diagonal positions, that is, a rectangular area having a line segment connecting the detected corner points as a diagonal line.
In this case, the device configuration is such that the number of area detecting means provided is 1
The configuration may be the same as that of the device of the first embodiment described above except that the number is set to one.

Further, the detection area may be a row area. The device configuration in this case can be similar to that of the above-described first embodiment, for example.

FIG. 5 is a diagram showing an example of image data in the case of line area detection. In the figure, 22 is original image data from the photoelectric conversion unit obtained by scanning the form by the reading unit 10, and the original image data 22 is image data (character data) 26 of the character line portion of the line area 24, The image data (background data) 28 of the character background area is included so that the character line portion and the background area can be distinguished. In the illustrated example, the background of the form is represented by white bits (pixel value "0"), for example, the background of the form is a white background, and the character line written on the form is a black line, so that the character data 26 is black. Image data in which the character line portion and the background area can be distinguished is obtained by being represented by bits (pixel value “1”).

The line area 24 is detected by scanning the area detecting means 12 and 14 with respect to the characters forming the line and using the corner points A as the area coordinates.
The coordinates of B, C and D may be detected. The line area is a rectangular area defined by these corner points A, B, C, D, that is, a rectangular area having corner points A, B, C, D. The detection of the coordinates of the corner points A, B, C, D is to detect the maximum and minimum calculated values for the pixel having a predetermined pixel value (in this case, the black bit pixel meaning the character data 26). Can be done by

The coordinates of the detected corner points A, B, C, D can also be used to express the inclination of the row area 24. For example, the inclination of a straight line including the corner points A and C is the inclination amount of the row area 24, Alternatively, a value obtained by averaging the inclination of the straight line including the corner points A and C and the inclination of the straight line including the corner points B and D can be used as the inclination amount of the row region.

Second Embodiment <Explanation of Device Invention> FIG. 6 is a functional block diagram for explaining a second embodiment of the device invention. The same components as those of the first embodiment are designated by the same reference numerals, and detailed description thereof will be omitted.

The second embodiment of the device invention has the same configuration as the first embodiment of the device invention described above, except that the reading unit includes a filter.

In the figure, reference numeral 30 denotes a reading unit, which is a photoelectric conversion unit 10a and a filter 30a for removing noise pixels.
With. The filter 30a is provided between the photoelectric conversion unit 10a and the area detection means 12 and 14. Filter 30a, like the filter has been used, for example conventional, and a shift register and a ROM (R ead- O nly M emory ).

The filter 30a of this embodiment outputs, for example, corrected image data of the original image data 20 input from the photoelectric conversion unit 10a. The maximum value detecting means 121b, 141b and the minimum value detecting means 121c, 141c of the area detecting means 12, 14 input the corrected image data with less noise pixels, and therefore, in this embodiment, the corrected image data is used for area detection. Used as image data.

By detecting the area coordinates from the corrected image data, there is an advantage that the area can be detected more accurately and accurately than in the first embodiment described above.

<Explanation of Method Invention and Operation of Apparatus Invention> The filter 30a uses a pixel value of one pixel (center pixel) of the original image data 20 as an arbitrary suitable number of pixels (observation pixel) around the center pixel. The pixel value of is used as the observation information to determine. The corrected image data in which the pixel value of the pixel of the original image data 20 is changed to the pixel value of the central pixel becomes the image data for area detection. Therefore, the pixel value of the image data for area detection is determined via the filter 30a.

FIG. 7 is a diagram showing an example of a filter. Filter 30
a is, for example, as shown in Figure 7 has a 3 × 3 of the observation window, the central pixel a ₀ and the observation pixel a _n (n is 8 or less natural number 1 or more and disposed around the central pixel a ₀ ) Are arranged in 3 rows and 3 columns, and the pixel values of 8 neighboring pixels of the central pixel a ₀ are used as the observation information.

FIGS. 8A and 8B are diagrams showing an example of original image data having noise pixels, and FIG. 8A is an enlarged view of a region surrounded by a dotted circle. It is a figure (B).

As shown in these figures, for example, when the original image data 20 in which the noise pixel data (noise data) 20c exists in the background data 20b is processed to detect the area coordinates, the X and Y coordinates of the noise pixels are set as the area coordinates. In some cases, the target detection area cannot be detected accurately as a result. In such a case, if the area coordinates are detected from the corrected image data from which the noise pixels in the background data 20b are removed, the area can be detected with high accuracy.

In this case, the filter 30a, when the sum of the pixel values of the observation pixel and the center pixel or the sum of the pixel values of the observation pixel becomes a predetermined comparison value or more, the pixel value of the center pixel, regardless of the value of the center pixel value, The background data is corrected to a pixel value that means and output.

For example, the original data 20a is a white bit and the background data 20b.
Is represented by a black bit, and therefore the noise data 20c becomes a white bit, and assuming that the predetermined comparison value is 5, the filter 30a uses the central pixel value when the total sum of the central pixel value and the observed pixel value is 5 or more. Is output as a black bit. As a result, the filter 30a converts the noise data 20c into the background data 20b.
Is output as a black bit.

The corrected image data output from the filter 30a of the reading unit 30 is input to the area detecting means 12 and 14, and the area detecting means 1
Reference numerals 2 and 14 detect the area coordinates from the input corrected image data.

The method and operation for processing the corrected image data and detecting the area coordinates are the same as those in the above-described first embodiment, and a description thereof will be omitted.

Further, in the case of detecting the row area described in the modified example of the first embodiment (see FIG. 5), the background data 28 is represented by white bits and the character data 26 is represented by black bits. The black bits inside become noise. Therefore, when the sum of the pixel values of the observation pixel and the center pixel or the sum of the pixel values of the observation pixels is equal to or more than a predetermined comparison value, the filter 30a sets the pixel value of the center pixel to the pixel value meaning the background data 28, that is, white. Output as bits.

According to the second embodiment, even if the original image data 20 contains noise, since the corrected image data from which the noise is removed by the filter 30a is used as the image data for area detection, accuracy not affected by noise It is possible to detect a good area.

Further, since the filter 30a has a simple configuration including, for example, a shift register and a ROM, the hardware scale can be made smaller in the second embodiment as in the first embodiment, as in the first embodiment.

In the second embodiment, the same modification as the first embodiment can be performed.

The present invention is not limited to the above-described embodiments, but each constituent can be arbitrarily modified or modified.

For example, not only binary-processed data but also multi-valued image data may be used as the image data for area detection. In this case, in the multi-valued image data, paying attention to a pixel having a pixel value that means that the pixel has a specific color, detecting the region coordinates from the maximum and minimum calculated values for this pixel, and thus specifying The image area having the color of may be detected. Alternatively, focusing on a pixel having a pixel value that means that the pixel has a specific density, the area coordinates are detected from the maximum and minimum calculated values of this pixel, and thus the image area having the specific density is detected. You can

Further, the filter may be a filter for detecting a specific figure, and by using this filter, it is possible to detect an image area including a specific figure, for example, a horizontal line or an X mark. In this case, when the filter detects a specific figure by observing the center pixel and an arbitrary suitable number of observation pixels in the vicinity of the center pixel, the pixel value of the center pixel is set as a pixel value indicating that the particular figure is detected. Output. Therefore, when the filter is a filter for detecting a specific figure, the pixel having the pixel value that means that the specific figure is detected is focused and the area coordinates are detected from the maximum and minimum calculated values of this pixel. By doing so, it is possible to detect the image area including the specific figure.

For example, when the filter is a filter that has a 3 × 3 observation window and detects horizontal lines (see FIG. 7), for example, a
₈ , a ₀ and a ₄ become black bits, and a ₁ , a ₂ , a ₃ and
When a ₇ , a ₆ and a ₅ become white bits, the filter may output the pixel value of the central pixel a _{0 as} a pixel value indicating that a specific figure (for example, a horizontal line) is detected. Good. Alternatively, as an example, when the filter has a 3 × 3 observation window and detects a mark x, a ₁ , a ₃ ,
a ₀ , a ₇ and a ₅ become black bits and are a ₂ , a ₈ , a ₄ and a
_{When 6} becomes a white bit, the filter may output the pixel value of the central pixel a _{0 as} a pixel value indicating that a specific figure (for example, x mark) is detected.

Further, the procedure (algorithm) for obtaining the maximum and minimum calculated values for the pixel having the predetermined pixel value is not limited to the above-described embodiment, but any suitable conventional method conventionally used for obtaining the maximum and minimum values can be used. The procedure may be used.

Further, the present invention may be applied to cut out the image data of the area (defined area) defined by the area coordinates from the original image data. In this case, for example, an image memory and a cutout unit may be added to the apparatus of the above-described embodiment. Then, as is conventionally done, the original image data from the reading unit is input to the area detecting unit and the image memory, and the area coordinates detected by the area detecting unit are input to the cutout unit.
The cutout unit cuts out and outputs the image data of the specified area from the original image data of the image memory based on the input area coordinates.

(Effect of the invention) As is apparent from the above description, according to the area detection method and the area detection apparatus of this application, the pixels of the two-dimensional image data are given X and Y coordinates and have a predetermined pixel value. The maximum value and the minimum value of the calculated value αX + βY regarding the pixel are detected. Then, the X and Y coordinates of the pixel giving the maximum and minimum calculated values are set as area coordinates, and the detection area is defined by these area coordinates.

Since the above-described procedure for defining the detection area is a very simple procedure, there is an advantage that the apparatus configuration can be greatly simplified and therefore the hardware scale can be reduced.

The above procedure can be performed in parallel with the scanning of the reading unit,
Therefore, since the area can be detected by a simple one-time scan, the area can be detected at high speed.

When the invention of this application is applied to an image processing apparatus having an image memory, the area detection can be ended at the same time when the writing of the image data to the image memory is completed,
Therefore, it is possible to realize an image processing device having a very high processing speed.

[Brief description of drawings]

FIG. 1 is a functional block diagram of a first embodiment of the apparatus invention, FIG. 2 is a diagram for explaining image data for detecting an original area, and FIGS. 3A to 3B are method inventions. FIG. 4 is a diagram for explaining an embodiment and an operation of the first embodiment of the apparatus invention, FIG. 4 is a diagram for explaining the principle of region coordinate detection, and FIG. 5 is a diagram of image data for detecting a line region. FIG. 6 is a diagram for explanation, FIG. 6 is a functional block diagram of a second embodiment of the device invention, and FIG. 7 is a filter for explaining a second embodiment of the method invention and operation of the second embodiment of the device invention. FIGS. 8A to 8B are diagrams showing examples of image data having noise. 10, 30 ... Reading unit, 12, 14 ... Area detection means 121, 141 ... Calculation detection means 121a, 141a ... Calculation means 121b, 141b ... Maximum value detection means 121c, 141c ... Minimum value detection means 122 , 142 ...... maximum value coordinate storage means 123, 143 ...... minimum value coordinate storage means 16 ...... X coordinate generation means, 18 ...... Y coordinate generation means 30a ...... filter.

Claims (11)

[Claims] 1. To express quantized image data for area detection in a two-dimensional coordinate system, X and Y coordinates are given to pixels of the image data, and the X and Y coordinates are provided. Calculating a calculated value αX + βY using at least one set of specific α value and β value to detect a maximum calculated value and a minimum calculated value with respect to the calculated value of a pixel having a predetermined pixel value; X of the pixel giving the minimum calculated value,
An area detection method, wherein each Y coordinate is set as area coordinates, and the detection area is defined by the area coordinates. 2. The α and β of at least one of (α = β = 1) and (α = 1, β = −1) are used as the α and β. First
The area detection method described in the item. 3. The detection area is a manuscript area, image data capable of distinguishing the manuscript area and the background area is obtained, and the predetermined pixel value is set as a pixel value which means a ground of the manuscript. The area detection method according to claim 1 or 2. 4. The detection area is a line area, image data capable of distinguishing a character line portion of the line area from a background area is obtained, and the predetermined pixel value is set as a pixel value meaning the character line portion. The area detection method according to claim 1 or 2. 5. The pixel value of the image data is determined through a filter for removing noise pixels or for detecting a specific figure, according to any one of claims 1 to 4. The area detection method according to any one of items. 6. A reading unit for outputting quantized image data for area detection, and X-coordinate generation for generating X-coordinates of pixels of the image data for expressing the image data in a two-dimensional coordinate system. Means, Y coordinate generation means for generating the Y coordinate of the pixel of the image data for expressing the image data in a two-dimensional coordinate system, the X coordinate and the Y coordinate, and at least one set of specific α value and β Calculation calculation means for calculating a calculation value αX + βY using the values and the maximum calculation value and the minimum calculation value with respect to the calculation value of a pixel having a predetermined pixel value; and the maximum calculation as area coordinates defining a detection area. A maximum value coordinate storage means for storing the coordinates of the pixel giving a value, and a minimum value coordinate storage means for storing the coordinates of the pixel giving the minimum calculated value as the area coordinates. Area detection device. 7. The calculation detecting means includes a calculating means for calculating the calculated value, a maximum value detecting means for detecting a maximum calculated value of a pixel having the predetermined pixel value, and a pixel having the predetermined pixel value. 7. The area detecting device according to claim 6, further comprising: a minimum value detecting means for detecting the minimum calculated value of. 8. The α and β of at least one of (α = β = 1) and (α = 1, β = −1) are used as the α and β. Sixth
The area detection device according to item 7 or 8. 9. A patent document characterized in that the detection area is an original area, image data capable of distinguishing the original area from the background area is obtained, and the predetermined pixel value is set to a pixel value which means a background of the original. The area detection device according to any one of claims 6 to 8. 10. The detection area is used as a line area, image data capable of distinguishing a character line portion of the line area from a background area is obtained, and the predetermined pixel value is set as a pixel value meaning the character line portion. Any one of claim 6 to claim 8
The area detection device according to item. 11. The reading section comprises a filter for removing noise pixels or for detecting a specific figure, according to any one of claims 6 to 10. The area detection device according to item 1.