JPH0821062B2 - Feature extraction method and feature extraction apparatus - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a method and apparatus for performing feature extraction of character / graphics at high speed and stably by a simple procedure.

(Prior Art) Conventionally, in feature extraction that is generally performed in a character / graphics recognition device, strokes are extracted from a character / graphics pattern, and recognition is performed using the positions, lengths, mutual relationships between strokes, and the like of the extracted strokes. The method of doing is widely adopted. The method is as follows: (1) The curvature is calculated for the contour point series detected by tracing the contour of the character figure, the contour series is divided with the point having a large curvature value as a division point, and the divided series is The stroke (line element of the pattern) is extracted by combining them, or (2) the character / graphic pattern is thinned to be a skeleton, and the connectivity and the skeleton pattern of the skeletonized pattern are traced to obtain a sharp angle. The change point or the like is detected, and the geometrical feature or the like of the stroke is extracted for identification.

Further, as a method when the number of characters to be recognized is small (for example, when numbers are recognized), for example, a method described in Document I
Matrix matching method disclosed in "Automatic recognition of characters and figures for computer input (edited by The Institute of Electrical Engineers of Japan)",
There are stroke analysis methods and geometric feature extraction methods.

(Problems to be Solved by the Invention) However, in the method of (1), which is the conventional general method described above, when the character pattern becomes large and the character / graphic pattern becomes complicated, the processing amount increases and the processing speed increases. Was in decline. Further, the method (2) requires that the character / graphic pattern be thinned, and there are problems such as the distortion of the pattern due to the thinning and the occurrence of beards, and the subsequent processing such as correction of the distortions and removal of the beards is complicated. I was trying. As a result of complicated processing, there is a problem that the scale of the device configuration becomes large and high-speed processing cannot be performed.

Also, the method described in Reference I is 10
It has the advantage of simplifying the processing procedure when recognizing several types of characters, but is vulnerable to noise such as print spots and chips, and is therefore easily affected by print quality, and further, print misalignment and positioning accuracy during reading are poor. There is a problem in that it is easily affected by the inclination (inclination) of the character pattern and the positional deviation that occur in the above.

Further, the methods (1) and (2) require the strokes to be extracted from the character / graphic pattern, and therefore have the same problems as those of the methods (1) and (2) described above. Therefore, the device scale is large and high-speed processing cannot be performed. There was a problem.

The purpose of this application is to solve the above-mentioned conventional problems, omitting a complicated processing procedure such as extracting a stroke from an input character / graphic pattern and performing feature extraction at a high speed with a simple processing procedure, and is stable. It is an object of the present invention to provide a method and apparatus capable of performing feature extraction.

(Means for Solving the Problem) In order to achieve this object, the method invention of the present application is a process for giving an X coordinate and a Y coordinate to a pixel of quantized image data including an extracted pattern to be a feature extraction target. And detecting the maximum and minimum calculated values αX + βY regarding the pixels of the extracted pattern having a predetermined pixel value using the X and Y coordinates and at least two sets of specific α and β values, and calculating these maximum and minimum values. A process of detecting the X and Y coordinates of a pixel giving a value as feature point coordinates, a process of detecting the size of the extracted pattern, and a normalization of the extracted pattern using the feature point coordinates and the size of the extracted pattern. And a process of calculating the geometrical feature amount.

In carrying out this method invention, it is preferable that the size of the extracted pattern is the length of the side of the circumscribed rectangular frame of the extracted pattern.

Further, the apparatus invention of this application includes: an X coordinate generating means for giving an X coordinate to a pixel of quantized image data including an extracted pattern to be a feature extraction target; and a Y coordinate to a pixel of the quantized image data. For Y
Using the coordinate generating means and the X and Y coordinates and at least two sets of specific α and β values, the maximum and minimum calculated values αX + βY regarding the pixels of the extracted pattern having a predetermined pixel value are detected, and these maximum values are detected. And coordinate detection means for detecting the X and Y coordinates of the pixel giving the minimum calculated value as feature point coordinates, size detection means for detecting the size of the extracted pattern, and feature point coordinates and extracted And a feature quantity calculating means for calculating a normalized geometric feature quantity of the extracted pattern using the size of the pattern.

In carrying out this device invention, the coordinate detecting means stores the calculating means for calculating the calculated value αX + βY, the maximum value detecting means for detecting the maximum calculated value, and the X and Y coordinates of the pixel giving the maximum calculated value. And a minimum value detecting means for detecting the minimum calculated value, and a minimum value coordinate storing means for saving the X and Y coordinates of the pixel giving the minimum calculated value. Is preferred.

Further, in carrying out this device invention, the size detecting means is
As the X and Y coordinates for defining the circumscribed rectangular frame of the extracted pattern, the maximum and minimum X of the pixels of the extracted pattern are set.
The size of the extracted pattern is calculated from the circumscribing rectangular frame detection means for detecting the coordinates and the maximum and minimum Y coordinates of the pixels of the extracted pattern, and the maximum and minimum X coordinates and the maximum and minimum Y coordinates. It is preferable to employ a configuration including a size calculating unit that

(Operation) According to the above-described method and apparatus invention, X-coordinates and Y-coordinates are given to the pixels of the quantized image data to convert the image data to X-images.
-It is expressed in the Y coordinate system. Then, by using the image data represented by this XY coordinate system, the maximum and minimum calculated values αX + βY regarding the pixels of the extracted pattern having a predetermined pixel value are detected and the pixels giving the maximum and minimum calculated values are detected. The processing of detecting the X and Y coordinates of the above as the characteristic point coordinates and the processing of detecting the size of the extracted pattern are performed.
Then, the normalized geometric feature amount of the extracted pattern is obtained using the feature point coordinates and the size of the extracted pattern, which are obtained by performing these processes in parallel or in chronological order. calculate.

In the above-described processing procedure for calculating the feature amount in this way, the procedure for detecting the maximum and minimum calculated values is a very simple procedure, so that the device configuration can be simplified. Further, since the procedure for detecting the maximum and minimum calculated values is simple, the feature amount can be calculated at high speed.

In addition, since the geometrical characteristic amount of the extracted pattern is normalized, the variation of the characteristic amount (in particular, the variation of the characteristic amount caused by the variation of the size of the extracted pattern) can be reduced.

(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, data, data flow and numerical conditions are not necessarily limited to the illustrated examples.

Embodiment <Structure of Embodiment of Device Invention> FIG. 1 is a functional block diagram for explaining the structure of an embodiment of the device invention.

In the figure, M indicates the quantized image data including the extracted pattern to be the feature extraction target. As shown in the figure, the feature extraction apparatus of this embodiment includes a read processing unit 8 for outputting one unit of image data M suitable for feature extraction, and an image data M.
X coordinate generation means 10 for giving X coordinates to the pixels of
A Y-coordinate generating means 12 for giving a Y coordinate to a pixel of the image data M, and a characteristic point detecting means 13 for detecting a characteristic point coordinate by using the image data M to which the X and Y coordinates are given.
And a size detecting unit 15 for detecting the size of the extracted pattern, and for calculating the normalized geometric feature amount of the extracted pattern using the feature point coordinates and the size of the extracted pattern. And the feature amount calculating means 18 of FIG.

 Hereinafter, this embodiment will be described in more detail.

(Image Data M) FIG. 4 is a diagram showing an example of quantized image data including an extracted pattern.

In the figure, M1 indicates an extracted pattern, and M2 indicates a background pattern which is a background of the extracted pattern. Quantized image data M
Is data consisting of these patterns M1 and M2 and including the extracted pattern M2 of one character or one figure which is the feature extraction target.

In this embodiment, the patterns M1 and M2 are represented by binary digital signals, and the extracted pattern M1 is a black bit having a pixel value "1" and the background pattern M2 is a white bit having a pixel value "0". .

In addition, the pixels of the image data M have an X axis in the main scanning direction (horizontal rightward direction) and a Y axis in the sub scanning direction (vertical downward direction), and in the XY coordinate system with the pixel at the upper left corner as the origin. As described later, the X and Y coordinates are given to the pixels of the image data M, as will be described later.

The extracted pattern M1 shown in FIG. 4 is a character pattern (character line) indicating the number "4", and in this embodiment, several characters of the E13B font are used as an example of the feature extraction target. For reference, the E1B font "1" is shown in Fig. 5 ~.
The character shapes of "9" and "0" are shown.

(Reading Processing Unit) The reading processing unit 8 has a configuration that outputs image data M including one unit of the extracted pattern suitable for feature extraction.
It may have any configuration. In this embodiment, although the reading processing unit 8 is not shown, it is described in a reading unit including a photoelectric conversion unit and a scanning mechanism (scanner), an image memory for storing original image data from the reading unit, a form, an original, and the like. The image data M is cut out from the original image data as is normally done based on the described position information of the character / graphic pattern.

(Feature Point Detection Means) FIG. 2 is a functional block diagram showing an example of the configuration of the feature point detection means.

As shown in the figure, the feature point detecting means 13 uses the X and Y coordinates of the pixel of the image data M and at least two sets of specific α and β values to determine the maximum and minimum of the pixel of the extracted pattern. Coordinate detecting means 14 and 16 are provided for detecting the calculated value αX + βY and detecting the X and Y coordinates of the pixel that gives the maximum and minimum calculated values, respectively, as feature point coordinates.

The coordinate detecting means 14 of this embodiment has a calculating means 141 for calculating a calculated value αX + βY, a maximum value detecting means 142 for detecting a maximum calculated value, an X of a pixel giving a maximum calculated value,
Maximum value coordinate storage means 143 for storing the Y coordinate, minimum value detection means 144 for detecting the minimum calculated value, and minimum value coordinate storage for storing the X and Y coordinates of the pixel giving the minimum calculated value. And means 145. Like the coordinate detecting means 14, the coordinate detecting means 16 also has a calculating means 161, a maximum value detecting means 162, a maximum value coordinate storing means 163, and a minimum value detecting means.
164 and minimum value coordinate storage means 165.

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 coordinate detecting means 14 and 16 are provided. In the coordinate detecting means 14, the calculated value X +
Calculated value X− for Y and for the coordinate detection means 16.
Find the coordinates of the pixels that give the maximum and minimum calculated values for Y.

Regarding the configuration of the coordinate detecting means 14 and 16, for example, the calculating means 141 includes an adding circuit, the calculating means 161 includes a subtracting circuit, and the detecting means 142, 144, 162 and 164 include comparators and registers, respectively. Therefore, the coordinate storage means 143, 145,
Each of 163 and 165 can be configured with a register. With such a configuration, the configurations of the coordinate detection means 14 and 16 can be greatly simplified, and the configuration of the feature extraction device can be simplified, thereby reducing the scale of hardware.

(Size Detection Means) FIG. 3 is a functional block diagram showing an example of the configuration of the size detection means.

The size detecting means 15 of this embodiment uses the X and Y coordinates for defining the circumscribing rectangular frame of the extracted pattern as the maximum and minimum X coordinates of the pixels of the extracted pattern and the maximum and minimum X coordinates of the pixels of the extracted pattern. A circumscribing rectangular frame detecting means 151 for detecting the minimum Y coordinate and a size calculating means 152 for calculating the size of the extracted pattern from the maximum and minimum X coordinates and the maximum and minimum Y coordinates are provided. .

The configuration of the circumscribing rectangular frame detection means 151 can be any suitable configuration capable of detecting the maximum and minimum X and Y coordinates of the pixels of the extracted pattern. , The minimum X coordinate (hereinafter, X
And X _S detector 151a for detecting the referred to as _S coordinates), the minimum Y-coordinate (hereinafter referred to as Y _S coordinate) for detecting
A Y _S detecting unit 151b, an X _E detecting unit 151c for detecting the maximum X coordinate (hereinafter referred to as X _E coordinate), and a maximum Y coordinate (hereinafter referred to as Y _E coordinate). And a Y _E detecting means 151d for that purpose.

In this circumscribing rectangular frame detecting means 151, the detecting means 151
Each of a, 151b, 151c and 151d can be configured by a comparator and a register, and the configuration can be simplified.

Further, the size calculation means 152 detects the detected X _S , Y _S , X _E and Y.
_{From E} , the lengths (X _E −X _S ) and (Y _E −Y _S ) of the sides of the circumscribing rectangular frame are calculated as the size of the extracted pattern.

(Modification) The configuration of the reading processing unit is not limited to that of the above-described embodiment.

For example, in the case where only one unit of character extraction for feature extraction is described on a form (for example, when only one character is written), the image memory and the reading processing unit may be omitted in the above-described embodiment. . In this case, the image data output from the reading processing unit by scanning the entire reading area of the form or the like may be used as the image data including the extracted pattern of one unit of feature extraction. Alternatively, the reading processing unit may be configured only by a scanning mechanism that operates so as to scan only an area for one unit of feature extraction based on the description position information of the character graphic written on the form, and the photoelectric conversion unit. May be.
In this case, the image data output from the reading processing unit by partially scanning the reading area may be used as the image data including the extracted pattern of one unit of feature extraction.

In addition, the number of coordinate detecting means provided can be arbitrarily changed depending on how many sets of α and β values are used for calculating the feature amount. For example, when n sets of α and β values are used,
The configuration may be such that n coordinate detecting means are provided, and the α and / or β values may be set to arbitrary and different values that are different for each coordinate detecting means.

Further, in the above-mentioned embodiment, the X-coordinate generating means and the Y-coordinate generating means are commonly used for the feature point detecting means and the size detecting means. Can be arbitrarily changed, and for example, the feature point detecting means and the size detecting means may be individually provided. In order to simplify the device configuration, it is preferable to use the X and Y coordinate generation means in common and reduce the number of arrangements.

The structure of the circumscribing rectangular frame detection means is X _S , Y _S , X _E and
Any suitable configuration that can detect Y _E may be used.

Further, in the above-mentioned embodiment, the size detecting means determines the size of the extracted pattern i) the length of the diagonal line of the circumscribing rectangular frame ii) Average side length of circumscribing rectangle May be detected and output.

<Description of Embodiment of Method Invention and Description of Operation of Embodiment of Apparatus Invention> (Explanation Focusing on Feature Point Detecting Means) FIGS. 6A to 6B show description focusing on the feature point detecting means. It is a figure to offer. Hereinafter, referring to FIG. 1 and FIG. 2 in addition to FIG. 6 (A) and (B), the operation of the feature point detecting means in this embodiment, and the process of giving X and Y coordinates related to this operation Also, the process of detecting the characteristic point coordinates will be described.

As a method of detecting the maximum calculated value and the minimum calculated value regarding the extracted pattern, for example, the following method can be considered. The calculation means 141 and 161 calculate calculation values for all pixels of the image data M, and the detection means 142, 144 and 162, 164 calculate the calculation values input from the means 141, 161 of the pixels having the pixel values of the extracted pattern. Detecting the minimum and maximum calculated values for the pixels of the extracted pattern by determining whether or not the calculated values of the pixels of the extracted pattern, and by inputting in the calculating means 141 and 161. It is determined whether or not the X and Y coordinates of the pixel are those of the pixel having the pixel value of the extracted pattern, and the calculation values are calculated from the X and Y coordinates of only the pixels of the extracted pattern, and the detection means 142, 144 and 162, at 164 means
The minimum and maximum calculated values for the pixels of the extracted pattern are detected by performing comparison and determination of all calculated values input from 141 and 161. The method may be any of the methods described above, and in the embodiments described below, detection is performed by the method described above.

I: Description Focusing on Coordinate Detection Means 14 * Steps (1) to (3) When the reading section of the reading processing section 8 starts scanning the reading area of the form in which the character graphic is written (step (1)),
The optical signal G of the character / graphic pattern on the form is input to the photoelectric conversion unit of the reading processing unit 8. The photoelectric conversion unit converts the input optical signal G into a binary quantized digital signal and outputs original image data composed of this digital signal. Then, the cutout unit of the reading processing unit 8 cuts out and outputs the image data M including the character / graphic pattern of one unit of feature extraction from the original image data.

Along with this, X coordinate generating means 10 and Y coordinate generating means 12
Generates the X and Y coordinates associated with each pixel of the data M in synchronization with the output of the image data M. As a result, the generation means 10 and 12 carry out the processing of giving the X and Y coordinates to the image data M. The output X and Y coordinates are calculated by the calculation means 141 and the maximum value coordinate storage means 143.
And the minimum value coordinate storage means 145. (Step (2)) When the calculating means 141 inputs the 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 142 and the minimum value detecting means 144.
(Step (3)) By the steps (2) to (3), the maximum value detecting means 14
2 is the image data M and the calculated value, the minimum value detection means 144 is the image data M and the calculated value, the maximum value coordinate storage means 143 is the X and Y coordinates, and the minimum value coordinate storage means 145 is the X and Y coordinates. Input each pixel. Then, the coordinate detecting means 14 repeats the determination of step (4), (5a), (5b) or (7) described later for each pixel and performs an operation according to the determination result, and as a result, the coordinate The detection means 14 performs the process of detecting the coordinates of the characteristic points.

* Step (4) The maximum value detecting means 142 and the minimum value detecting means 144 determine whether the pixels of the input image data M are the pixels of the extracted pattern M2. This judgment is made by judging whether or not the pixel value of the input pixel is a predetermined pixel value (pixel value "1" in this embodiment) meaning the extracted pattern M2.

When the pixel value does not have a predetermined pixel value, the maximum value detecting means 142 and the minimum value detecting means 144 do not rewrite the stored comparison value, and together with this, the maximum value coordinate storing means.
143 and the minimum value coordinate storage means 145 are stored in X, Y
Do not rewrite the coordinates.

The detecting means 142, 144 make a judgment in step (7) after step (4).

When the pixel value has a predetermined pixel value, the maximum value detecting means 142 carries out the comparison of the comparison value and the calculated value (step (5)) after step (4), and the minimum value detecting means 144 carries out the comparison after step (4). The value and the calculated value are compared (step (5b)).

* Step (5a) When the calculated value is larger than the comparison value, the maximum value detecting means 142 stores the calculated value larger than the comparison value as a new comparison value in place of the previously stored comparison value. Rewriting), and together with this, a set pulse is output to the maximum value coordinate storage means 143. The maximum value coordinate storing means 143 which has input the set pulse newly stores the X and Y coordinates of the pixel which gives a calculated value larger than the comparison value, instead of the stored X and Y coordinates (of the X and Y coordinates). Rewriting). (Step (6)) When the calculated value is smaller than or equal to the comparison value, the maximum value detecting means 142 stores the previously stored comparison value as it is without rewriting, and the maximum value coordinate storing means.
122 does not rewrite the stored X and Y coordinates.

In any of the above cases, the maximum value detecting means 142 makes the judgment of step (7) after step (5a).

As the initial value of the comparison value stored in the maximum value detecting means 142, for example, a value smaller than the value that can be taken as the calculated value α may be used. For example, α = β = 1 and the image data M is divided into pixels of l rows and m columns (thus, 0 ≦ X ≦ m−
1, and 0 ≦ Y ≦ l−1), for example, −1
Can be set as the initial value of the comparison value. Alternatively, the calculated value αX + βY input first to the maximum value detecting means 142 may be used as the initial value of the comparison value.

The maximum value detecting means 142 rewrites the comparison value and the X and Y coordinates when the calculated value αX + βY is larger than the comparison value and when the calculated value is equal to the comparison value, and the calculated value is larger than the comparison value. Comparison value and X, Y when small
The coordinates may not be rewritten.

Any suitable numerical value may be used as the X and Y coordinates of the maximum value coordinate storage means 143.

* Step (5b) When the calculated value is smaller than the comparison value The minimum value detecting means 144 newly stores the calculated value smaller than the comparison value as the comparison value in place of the previously stored comparison value. Rewriting) and minimum value coordinate storage means
Output a set pulse to 145. When the set pulse is input, the minimum value coordinate storage unit 145 rewrites the stored X and Y coordinates to the X and Y coordinates of the pixel that gives a calculated value smaller than the comparison value (rewriting of the X and Y coordinates). (Step (6)) The minimum value detecting means 144 makes the judgment of the step (7) after the step (5b) in either case.

As the initial value of the comparison value stored in the minimum value detecting means 144, for example, a value larger than the possible values of the calculated value αX + βY may be used. For example, α = β = 1 and the image data M is divided into pixels of 1 row and m column (thus 0 ≦
(X ≦ m−1 and 0 ≦ Y ≦ l−1), for example, m + n−1 can be set as the initial value of the comparison value.
Alternatively, the calculation value αX + βY input first to the minimum value detecting means 144 may be used as the initial value of the comparison value.

Further, the minimum value detecting means 144, when the calculated value is smaller than the comparison value and when the calculated value is equal to the comparison value, the comparison value and X,
The rewriting of the Y coordinate may be performed, and when the calculated value is larger than the comparison value, the rewriting of the comparison value and the X and Y coordinates may not be performed.

Further, as the X and Y coordinates of the minimum value coordinate storage means 145, any suitable numerical value may be used.

* Step (7) When the scanning of the image data M is completed, the maximum value detecting means 142 and the minimum value coordinate storing means 144 are completed when the scanning of the image data M is completed and all the pixels of this data M are processed, X, The output signal of the Y coordinate is output to the maximum value coordinate storage means 143 and the minimum value coordinate storage means 145. Coordinate storage means 143, 145 to which this output signal is input
Outputs the stored X and Y coordinates as feature point coordinates. At the same time, the detecting means 142 and 144 initialize the comparison value. (Step (8)) When the processing is completed for all the pixels, the X and Y coordinates of the pixel which gives the maximum and minimum calculated values of the X and Y coordinates stored in the maximum value coordinate storage means 143 and the minimum value coordinate storage means 145. ,
It is the Y coordinate, that is, the feature point coordinate.

In the coordinate detecting means 14, since α = β = 1, the coordinates of the feature point BR of the extracted pattern M2 shown in FIG. 4, for example, are stored in the maximum value coordinate storing means 143 at the end of the processing. Further, the coordinates of the feature point TL are stored in the minimum value coordinate storage means 145.

When the scanning of the image data M is not completed The coordinate detection means 14 executes the step () for the remaining pixels of the image data M unless the scanning of the image data M is completed and the processing is not completed for all the pixels of the data M. 4), (5a), (5b) or (7) is determined, and the operation is performed according to the determination result.

II: Description Focusing on Coordinate Detection Means 16 Since the coordinate detection means 16 performs the same operation as the coordinate detection means 14 in parallel with the operation of the coordinate detection means 14 described above, the operation of the detection means 16 and the detection means 16 The description of the process of assigning the X and Y coordinates and the process of detecting the feature point coordinates related to the above is omitted.

However, in the coordinate detecting means 16, α = 1 and β = −
Since it is set to 1, the coordinates of the feature point TR of the extracted pattern M2 shown in FIG. 4 are stored in the maximum value coordinate storage means 163 when the processing is completed for all the pixels, and the feature point BL The coordinates are stored in the minimum value coordinate storage means 165.

III: Principle Description of Feature Point Coordinate Detection FIG. 7 is a diagram for explaining the principle of feature point coordinate detection. This will be described below with reference to FIG. 7, but in this description, the extracted pattern M2 that is the feature extraction target is a rectangular pattern. In this case, the feature points are the corner points TL and T of the extracted pattern M2.
R, BL, BR.

In order to detect the feature point coordinates, first, a straight line αX + βY passing through the extracted pattern M2 on the image data M is assumed,
Consider a straight line group αX + βY = C in which the slope is fixed to a constant value and the value of C has various values. Then, a straight line passing through the extracted pattern M2 can be found, and at this time, the value of C of the straight line passing through the corner point which is the characteristic point becomes the minimum value or the maximum value in the straight line group.

Therefore, it can be understood that the value of C, that is, the calculated value αX + βY, is calculated for each pixel of the extracted pattern M2, and the X, Y coordinates of the pixel having the maximum or minimum calculated value αX + βY and the coordinates of the characteristic points are used. . In case of rectangular pattern,
Two corner points (feature points) can be detected by using a set of specific α and β values.

For example, considering α = β = 1 and considering a group of straight lines L: X + Y = C ₁ passing through the region of the extracted pattern M2, and making the value of C ₁ smaller, a straight line L ₁ passing through the feature point TL is obtained. I can understand Further, when the value of C ₁ is minimum, a straight line L ₁ passing through the feature point TL can be obtained. Further, when the value of C ₁ is increased, a straight line closer to the feature point BR is obtained, and when the value of C ₁ is maximum, a straight line L _n passing through the feature point BR is obtained. Therefore,
It can be understood that among the pixels of the extracted pattern M2, the coordinates of the pixel giving the minimum value of the calculated values X + Y are the coordinates of the feature point TL and the coordinates of the pixel giving the maximum value are the coordinates of the feature point BR.

Similarly, considering a group of straight lines K: X−Y = C ₂ that passes through the region of the extracted pattern M2 with α = 1 and β = −1, for example, when the value of C ₂ is the minimum, the feature point BL is Passing straight line K _m
A straight line passing through the feature point TR when the values of and C ₂ are maximum
You get K ₁ . Therefore, among the pixels of the extracted pattern M2, the coordinates of the pixel giving the minimum value of the calculated values XY become the coordinates of the feature point BL and the coordinates of the pixel giving the maximum value become the coordinates of the feature point TR.

(Explanation Focusing on Size Detecting Means) Next, with reference to FIGS. 1 and 3, the operation of the size detecting means in this embodiment, the X related to this operation,
The process of assigning the Y coordinate and the process of detecting the size of the extracted pattern will be described.

First, when the reading processing unit 8 starts scanning the reading area of the form in which the character graphic is written, the reading processing unit 8 outputs one unit of image data M suitable for feature extraction for each pixel,
At the same time, the coordinate generating means 10 and 12 generate X and Y coordinates which are synchronized with the output of the image data M and are associated with each pixel of the image data M (the coordinate generating means 10 and 12 give the X and Y coordinates. Processing is done).

As a result, the X _S detection means 151 of the circumscribing rectangular frame detection means 151
a, Y _S detecting means 151b, X _E detecting means 151c and Y _E detecting means 151
In d, the image data M from the reading processing unit 8 and the X and Y coordinates from the coordinate generating means 10 and 12 are input.

(1) X _S detected X _S detection means 151a of operation and X _S of the detection means determines whether the pixel of the image data M which has been input is a pixel having a predetermined pixel value which means the extraction pattern M2 (For example, it is determined whether or not the pixel value is "1"), and if the pixel has a predetermined pixel value, the X coordinate of this pixel is compared with the X _S coordinate stored in the detection means 151a. .

If the stored X _S coordinate is smaller than the input X coordinate, the X _S detection means 151a stores this X coordinate as a new X _S coordinate (rewriting of the X _S coordinate). Further, when the stored X _S coordinate is larger than the input X coordinate and equal to the X coordinate, the stored X _S coordinate is not rewritten.

When the comparison and the rewriting of the X _S coordinates according to the comparison result are completed for all the pixels of the image data M,
X _S coordinates stored in the X _S coordinate detection means 151a is used as the minimum X coordinates for the pixels of the extracted pattern M2.

(2) Detection Y _S detection means 151b of Y _S detector and Y _S, as in the X _S detection means 151a, Y
The Y coordinate input from the coordinate generating means 12 and this detecting means 15
The Y _S coordinate stored in 1b is compared, and the Y _S coordinate is rewritten according to the comparison result.

When the processing is completed for all the pixels of the image data M, the Y _S coordinate stored in the Y _S detecting means 151b is used as the minimum Y coordinate for the pixel of the extracted pattern M2.

(3) X _E Detection Unit and X _E Detection The X _E detection unit 151c determines whether or not the pixel of the input image data M is a pixel having a predetermined pixel value that means the extracted pattern M2, If the pixel has a pixel value, the X coordinate of this pixel is compared with the X _E coordinate stored in the detecting means 151c.

If the stored X _E coordinate is smaller than the input X coordinate, the X _E detecting unit 151c stores this X coordinate as a new X _E coordinate (rewriting of the X _E coordinate). When the stored X _E coordinate is larger than the input X coordinate or equal to the X coordinate, the stored X _E coordinate is not rewritten.

Such processing, upon completion for all the pixels of the image data M X _E-coordinate detection unit 151c to the stored and are X _E
The coordinate is used as the maximum X coordinate for the pixel of the extracted pattern M2.

(4) Y _E Detection Means and Y _E Detection The Y _E detection means 151d is the same as the X _E detection means 151c.
The Y coordinate input from the coordinate generating means 12 and this detecting means 15
The Y _E coordinate stored in 1d is compared, and the Y _E coordinate is rewritten according to the comparison result.

The Y _E coordinate stored in the Y _E detecting means 151d at the time when the processing is completed for all the pixels of the image data M is used as the maximum Y coordinate for the pixel of the extracted pattern M2.

Then, the size detecting means 152 of the size detecting means 15
The maximum and minimum X and Y coordinates of the pixels of the extracted pattern M2 from the circumscribing rectangular frame detection means 151, that is, X _S , Y _S , X _E, and
Enter the Y _E coordinates, for example, as the size of the extracted patterns M2 from these coordinates, and outputs the calculated (X _E -X _S) and (Y _E -Y _S).

The circumscribed rectangular frame is (X _S , Y _S ), as shown in FIG.
It is a rectangular frame defined by four points (X _S , Y _E ), (X _E , Y _S ) and (X _E , Y _E ), that is, a rectangular frame having these four points as corner points.

(Explanation Focusing on Feature Quantity Calculating Means) Next, with reference to FIG. 1, the operation of the feature quantity calculating means in this embodiment and the operation related thereto will be performed.
The process of calculating the normalized geometric feature amount of the extracted pattern will be described.

The feature amount calculating means 18 is the feature point from the feature point detecting means 13.
When the coordinates of BR, TL, TR, BL and the size of the extracted pattern M2 from the size detecting means 15 are input, the geometrical pattern normalized using these feature point coordinates and the size of the extracted pattern M2. The characteristic features DT, DB, DR, DL are calculated according to the following equations (1) to (4), and the calculated characteristic amounts DT, DB, DR, DL are output. In addition, in the equations (1) to (4), the feature points BR, TL, TR,
BL X coordinate is BRX, TLX, TRX, BLX and Y coordinate is BR
Represented as Y, TLY, TRY, BLY.

The method invention and apparatus invention of this application are not limited to the above-described embodiments, but the configurations and operations of the respective constituents of the apparatus invention, the processing procedure in the method invention, and the like can be arbitrarily changed.

For example, not only the binary processed data but also the multivalued data may be used as the image data M for the feature extraction. It is possible to perform feature extraction of a character / graphic pattern of a specific color or feature extraction of a character / graphic pattern having a specific range of density from multi-valued image data. In the case of a character / graphic pattern of a specific color, the feature point coordinates can be detected by focusing on a pixel having a pixel value that means that the pixel has a specific color and calculating the maximum and minimum calculated values for this pixel. . In the case of a character / graphic pattern having a density in a specific range, the feature point coordinates can be detected by focusing on a pixel having a pixel value in the specific density range and obtaining the maximum and minimum calculated values for this pixel. I can.

In the above-described embodiment, the case where the characteristic point coordinates are detected by using two groups of straight lines L and K having different inclinations by using two sets of α and β has been described. In the case of detecting the characteristic point coordinates related to the (shape) figure pattern, the characteristic point coordinates may be detected by using three or more sets of α and β and thus using three or more straight line groups having different inclinations. In this case, regarding the α and β values of each set, the maximum and minimum calculated values are obtained according to the same principle as described above, and the geometric feature amount of the character / graphic pattern is obtained from the coordinates of the pixels giving these maximum and minimum calculated values. Can be done. In this case, the device configuration is such that the number of coordinate detecting means provided is set to α and β used for detecting the feature point coordinates.
The configuration may be the same as that of the above-described embodiment except that the number of sets of values is the same.

In addition, various geometric feature amounts obtained based on the coordinates of the detected feature points can be used as the geometric feature amount before normalization. In the above-described embodiment, the example in which the distance between the feature points is normalized has been described. However, in addition to this, for example, geometric features obtained by normalizing the area of a polygonal region such as a triangle or a quadrangle obtained by connecting feature points. The amount and the like may be used. Further, as the size of the extracted pattern, in addition to the above-described embodiment, for example, the length of the side of the circumscribed polygon circumscribing the extracted pattern M2 or the length of the diagonal line, the radius of the circumscribed circle, and the center of gravity of the extracted pattern. It is possible to use the second moment around and the like. The normalized geometric feature amount of the extracted pattern is obtained by dividing the geometric feature amount before being normalized by the size of the extracted pattern. Then, various normalized feature quantities can be used alone or in various combinations to recognize a character pattern or a graphic pattern.

Further, the procedure (algorithm) for calculating the maximum and minimum calculation values for the pixel having the predetermined pixel value is not limited to the above-described embodiment, and any procedure conventionally used for obtaining the maximum and minimum values. Any suitable procedure may be used.

Further, the method and apparatus invention of this application is particularly suitable for application to a character recognition device that recognizes simple characters, such as numbers, and in this case, realizes a small character recognition device with high processing speed. Can be done. Further, these method and device inventions may be widely applied to a character recognition device, a pattern recognition device and other image processing devices.

(Effect of the Invention) As is clear from the above description, according to the feature extraction method and feature extraction apparatus of this application, the X-coordinate and the Y-coordinate are given to the pixels of the quantized image data to convert the image data to the X-
Expressed in the Y coordinate system. Then, by using the image data represented by this XY coordinate system, the maximum and minimum calculated values αX + βY regarding the pixels of the extracted pattern having a predetermined pixel value are detected and the pixels giving the maximum and minimum calculated values are detected. A process of detecting the X and Y coordinates of the above as feature point coordinates;
Processing for detecting the size of the extracted pattern is performed. Then, the normalized geometric feature amount of the extracted pattern is obtained using the feature point coordinates and the size of the extracted pattern, which are obtained by performing these processes in parallel or in chronological order. calculate.

In the above-described processing procedure for calculating the feature amount in this way, the procedure for detecting the maximum and minimum calculated values is a very simple procedure, so that the device configuration can be simplified. Further, since the procedure for detecting the maximum and minimum calculated values is simple, the feature amount can be calculated at high speed. In particular, when the reading processing unit has an image memory, the characteristic point coordinates can be detected by one scan of the image memory, so that the speedup can be effectively achieved.

Further, since the geometric feature amount of the extracted pattern is normalized, it is possible to reduce the variation of the characteristic amount even if the size of the extracted pattern changes or the printing accuracy is insufficient.
Therefore, it is possible to obtain a stable feature amount that is not easily affected by variations in the size of the extracted pattern and insufficient printing accuracy.
The variation in the size of the extracted pattern occurs when, for example, photoelectric conversion units having different resolutions are used or character figures having different sizes such as uppercase letters and lowercase letters are used.

In addition, since the coordinates of the pixels that give the maximum and minimum calculated values are the feature point coordinates, compared to the conventional matrix matching method, it is stable and less susceptible to the inclination of the form, original, etc., and the lack of the extracted pattern. There is an advantage that the specified feature amount can be obtained.

[Brief description of drawings]

FIG. 1 is a functional block diagram used to explain an embodiment of the apparatus invention, FIG. 2 is a functional block diagram showing an example of the configuration of a feature point detecting means composed of coordinate detecting means, and FIG. FIG. 4 is a functional block diagram showing one configuration example, FIG. 4 is a diagram showing an example of quantized image data used for feature extraction, FIG. 5 is a diagram showing an example of a character pattern suitable as a feature extraction target, and FIG. (A) and (B) are operation flow charts for explaining the embodiment of the method invention and the apparatus invention, focusing on the feature point detecting means, and FIG. 7 is a diagram for explaining the principle of feature point coordinate detection. 8 ... Read processing section, 10 ... X coordinate generating means 12 ... Y coordinate generating means, 13 ... Feature point detecting means 14, 16 ... Coordinate detecting means 141, 161, ... Calculating means 142, 162 ... Maximum value detecting means 143, 163 ... Maximum value coordinate storage means 144, 164 ... Minimum value detection means 145, 165 ... Minimum value coordinate storage means 15 ... Size detection means 151 ... Enclosing rectangular frame detection means 152 ... Size calculation means 18 ... Feature amount calculation means.

Claims (5)

[Claims] 1. A process of assigning an X coordinate and a Y coordinate to a pixel of quantized image data including an extracted pattern to be a feature extraction target, and the X and Y coordinates and at least two sets of specific α and β values. Detect the maximum and minimum calculated values αX + βY relating to the pixels of the extracted pattern having a predetermined pixel value using
A process of detecting the X and Y coordinates of the pixel giving the maximum and minimum calculated values as feature point coordinates, a process of detecting the size of the extracted pattern, and the feature point coordinates and the size of the extracted pattern. And a process for calculating a normalized geometric feature amount of the extracted pattern using the above. 2. The feature extraction method according to claim 1, wherein the size of the extracted pattern is the length of a side of a circumscribing rectangular frame of the extracted pattern. 3. X-coordinate generating means for giving an X coordinate to a pixel of quantized image data including an extracted pattern to be a feature extraction target, and Y coordinate for giving a pixel to the quantized image data. Using Y coordinate generation means and the X and Y coordinates and at least two sets of specific α and β values, the maximum and minimum calculated values αX + βY regarding the pixels of the extracted pattern having a predetermined pixel value are detected. ,
Coordinate detecting means for detecting the X and Y coordinates of the pixel giving the maximum and minimum calculated values as feature point coordinates, size detecting means for detecting the size of the extracted pattern, and the feature A feature extraction device, comprising: a feature amount calculation means for calculating a normalized geometric feature amount of the extracted pattern using the point coordinates and the size of the extracted pattern. 4. The coordinate detecting means calculates the calculated value αX + βY, maximum value detecting means for detecting the maximum calculated value, and X and Y coordinates of a pixel giving the maximum calculated value. Maximum value coordinate storage means for storing, minimum value detection means for detecting the minimum calculated value, and minimum value coordinate storage means for storing the X and Y coordinates of the pixel giving the minimum calculated value. The feature extraction device according to claim 3, wherein the feature extraction device is provided. 5. The size detecting means includes X and Y for defining a circumscribing rectangular frame of the extracted pattern.
As the coordinates, a circumscribing rectangular frame detecting means for detecting maximum and minimum X coordinates regarding pixels of the extracted pattern and maximum and minimum Y coordinates regarding pixels of the extracted pattern, and the maximum and minimum X coordinates and the above The feature extraction apparatus according to claim 3, further comprising a size calculation unit that calculates the size of the extracted pattern from the maximum and minimum Y coordinates.