JPH05324917A - Fuzzy pattern recognizing device - Google Patents

Abstract

PURPOSE:To provide a fuzzy pattern recognizing device which can accelerate an operation processing for calculating the degree of recognition by comparing the pattern of a recognizing object with a pattern for collation. CONSTITUTION:A first processor 3 calculates the degree of recognition with an arithmetic operation using the adaptability of the pattern for collation concerning the pattern of the inputted recognizing object. A second processor 8 previously calculates the adaptability for the parameter of the pattern for collation and for the respective dots of the pattern for collation calculated by a three-dimensional membership function concerning all the patterns for collation and stores the results in a fuzzy template adaptability storage part 6 and an average adaptability storage part 7 for the total adaptability of ideal input patterns. In the case of pattern recognition, the pattern of the recognizing object and the pattern for collation are comparatively operated by a first processor 3 while referring to the values of the adaptability stored in the storage parts 6 and 7.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fuzzy pattern recognition apparatus having a function of accurately recognizing a pattern transformed from a basic pattern, such as a handwritten character for a typeface, by using a fuzzy theory.

[0002]

2. Description of the Related Art A conventional pattern recognition device generally recognizes a pattern whose shape is strictly defined.
A method using a neural network is known as a means for recognizing an ambiguous pattern that is not limited to a certain shape such as handwritten characters. In this, a neural network is formed by learning a pattern to be recognized in advance, and the input pattern is recognized thereby.

However, the method using this neural network requires a work of learning, and requires re-learning each time the number of pattern candidates to be identified increases. In addition, if the neural network is a black box and changes in the characteristics of the candidate of the pattern to be identified due to secular change of the input device that takes in the pattern signal of the recognition target, it is reflected in the operation of the device. It is difficult to do so, and there is a problem that erroneous recognition may occur.

On the other hand, in recent years, a method of recognizing an ambiguous pattern using a fuzzy theory has been proposed. This method is based on having a plurality of knowledge about the characteristics of the pattern and checking them one by one, such as "there is a vertical line near this portion". However, in the conventional pattern recognition method based on the fuzzy theory, it is necessary to acquire knowledge about pattern features (rule creation work) when developing a system including software, and pattern feature extraction is performed when actual pattern recognition is performed. An algorithm is needed. for that reason,
There is a problem that processing is complicated because many rules and membership functions of fuzzy sets are required.

[0005]

Therefore, as a solution to the problem of pattern recognition based on the conventional fuzzy theory, the inventor combines a preset fuzzy set membership function to collate it with a recognition target. We have proposed a pattern recognition method that prepares a basic pattern that should be used and compares it with the pattern to be recognized to calculate the certainty factor, which is an index for evaluating the pattern ( Japanese Patent Application No. 2-274355
issue). According to this method, a pattern to be collated with a recognition target is formed by combining membership functions indicating a fuzzy region having a boundary of an arbitrary shape (for example, an ellipse). It is possible to efficiently recognize patterns such as characters and figures without needing to do so. The matching pattern formed by combining the membership functions by this method is called a "fuzzy template", and this "fuzzy template" is also used in this specification.

Here, taking the number "8" as an example, the fuzzy template is expressed as follows.

An extension of the range of goodness of fit of a three-dimensional elliptic membership function with a goodness of fit of a circle having a center point of (6, 15) and a radius of 4.5 to [-1, 1] (the upper part of "8") Circle)

[0008]

[Equation 1]

When the range of goodness of fit of a three-dimensional elliptic membership function having a goodness of fit of a circle having a center point of (6, 6) and a radius of 4.5 is [-1, 1] (of "8", Bottom circle)

[0010]

[Equation 2]

The reason for expanding the range of the fitness to [-1, 1] is to give an evaluation value of- (minus) to an inappropriate input coordinate value. For example, a black image signal This is because, when is input, the total sum of the degree of conformity described later becomes a small value.

The fuzzy template of "8" is expressed by the following function created by the combination with the above.

F = max (f ₁ , f ₂ ) (3) With respect to the above equation, as shown in FIG. 2, 12 horizontal dots × 21 vertical dots are set on the xy coordinate plane. That is,
An area including the character "8" is divided into 12 areas in the x-axis direction and 21 areas in the y-axis direction, and each area (dot) is converted into a binary signal corresponding to the character shape. Here, the blank area is converted into a binary signal “0”, and the black area is converted into a binary signal “1”. As the size correction, the second black areas from the upper, lower, left, and right ends are placed in the second column from the upper, lower, left, and right ends, respectively.

At this time, assuming that the pattern shown in FIG. 7 is input, n = 42 out of 12 × 21 = 252 dots.
The binary signal “1” (black) is given to each of the two.

Therefore, by using the membership function f of the above equation (3), the sum T ₁ of the fitness of these n black dots and the average fitness S ₁ = T ₁ / n for each dot are obtained. .. In the example shown in FIG. 7, the total fitness T ₁ = 36.05 calculated by the coordinate values of n = 42 black dots, and the average fitness S ₁
＝ 36.05 / 42 ＝ 0.858.

Similarly, the sum T ₂ of the goodness of fit and the average goodness of fit S ₂ when the most ideal “8” pattern is input are obtained. In this case, the sum T of the suitability of 44 black dots
₂ = 43.59 and the average goodness of fit S ₂ = 43.59 / 44 = 0.991.

From the above results, the confidence is calculated by taking the weighted average of the ratio T ₁ / T ₂ of the total sum of the goodness of fit (total goodness of fit) and the ratio S ₁ / S ₂ of the average goodness of fit. in this case,
The confidence is as follows.

[0018]

[Equation 3]

The ratio T ₁ / T ₂ of the total goodness of fit exceeds 1 in some cases, but in that case, it is unconditionally set to 1. That is, min (T ₁ / T ₂ , 1).

Here, the fuzzy template is displayed as follows. That is, as shown in FIG. 2, a three-dimensional membership function base set plane (a coordinate plane formed by two coordinate axes other than the coordinate axis indicating the goodness of fit) is divided into a finite number of dots, and the center point of each dot is When the goodness of fit of the three-dimensional membership function obtained by the coordinate values is obtained,
A black circle with a radius proportional to the size of the goodness of fit for each dot is displayed in each dot. As a result, like the fuzzy template shown in FIG. 7, a place with a high degree of conformity is displayed in black, and a place with a low degree of conformity is displayed in white.

However, in the above fuzzy pattern recognition method, the degree of conformity of the input pattern with respect to the fuzzy template, which is necessary for matching the input pattern with the fuzzy template, must be calculated by the three-dimensional membership function formula. , It takes a long calculation time. In addition, the ideal input pattern for forming a fuzzy template is manually set in advance, and the matching degree for the ideal input pattern must be calculated when matching, which is troublesome for the operation. At the same time, there is a problem that the calculation time becomes longer.

Therefore, an object of the present invention is to perform a calculation process for calculating a certainty factor of recognition by comparing a pattern to be recognized with a pattern for collation in a device which performs pattern recognition based on a membership function representing a fuzzy set. An object of the present invention is to provide a fuzzy pattern recognition device capable of speeding up.

[0023]

According to the present invention, in a device for pattern recognition based on a membership function representing a fuzzy set, a pattern input section for inputting a pattern to be recognized and one or more membership functions are generated. Then, by combining the membership functions, a matching pattern forming means for forming a matching pattern for recognizing a recognition target pattern input from the pattern input section, and a recognition target input from the pattern input section. And a calculation processing means for calculating the certainty factor of recognition by comparing the pattern with the matching pattern formed by the matching pattern forming means, the calculation processing means including the parameters of the matching pattern and the three-dimensional membership. The matching degree of each dot of the matching pattern calculated using the function is calculated for all matching patterns. It is characterized in that the pattern to be recognized is compared with the collation pattern by referring to the stored value of the goodness of fit in the case of pattern recognition. ..

In another aspect of the present invention, the arithmetic processing means comprises:
An ideal input pattern corresponding to the matching pattern is a pattern obtained by considering a dot whose pre-calculated fitness value is equal to or greater than a predetermined reference value as one of two values of the ideal input pattern. Is characterized in that the values of the total goodness of fit and the average goodness of fit are obtained and stored.

[0025]

In the present invention, the arithmetic processing means preliminarily calculates the suitability of each dot calculated using the parameters of the matching pattern (fuzzy template) and the three-dimensional membership function formula for all the matching patterns. Then, the value (fitness) of the calculation result is stored. Then, at the time of pattern recognition, the fitness calculation by the three-dimensional membership function expression is not performed, and the comparison calculation between the input pattern and the matching pattern is performed with reference to the stored value. Therefore, high-speed matching is possible, and the calculation time for pattern recognition is significantly shortened.

Further, a dot whose pre-calculated fitness value is equal to or larger than a predetermined reference value is defined as an ideal input pattern 2
A pattern obtained by regarding one of the values (for example, “1” that becomes “ON” of binary {0,1}) is an ideal input pattern corresponding to the matching pattern.
The values of the total goodness of fit and the average goodness of fit are obtained and stored. At the time of pattern recognition, by referring to these stored values, the input pattern and the matching pattern can be matched at high speed.

[0027]

1 is a block diagram showing an embodiment of a fuzzy pattern recognition apparatus according to the present invention. This device
It is configured to recognize characters and consists of the following components.

The image reading device 1 includes a light source for illuminating a recognition target in order to read an image of a typeface or handwritten characters,
A photoelectric converter that reads an image by reflected light of the image and converts the image into an electric signal is included. Alternatively, a pressure-sensitive handwritten character input device may be used.

The signal conversion section 2 converts the image signal read by the image reading apparatus 1 into a binary signal of, for example, "0" (= "white") or "1" (= "black"), and , Is configured to correct the size of the read image. This correction (scaling) extends the shape up and down, for example, when the character has a shape crushed up and down.

The image reading device 1 and the signal conversion unit 2 constitute a pattern input unit for inputting a pattern to be recognized.

The first processor 3 recognizes the read image (input pattern) which has been binarized and whose shape has been corrected by the signal conversion unit 2. That is, when a pattern to be recognized is input, a matching calculation is performed on the matching pattern (fuzzy template) of the input pattern with reference to the information stored in the storage units 6 and 7 described later, and the recognition certainty factor is calculated. To calculate. Further, the pattern is classified into the pattern with the highest certainty as compared with other patterns with ambiguous boundaries, and the recognition result is output.

The pattern indicating the recognition result by the processor 3 is converted from a binarized signal to an analog signal by the signal conversion unit 4 and output by the output device 5 such as a CRT.

The fuzzy template suitability storage unit 6
A second processor 8, which will be described later, stores the values of the goodness of fit of each dot of all fuzzy templates calculated using the parameters of the matching pattern (fuzzy template) and the three-dimensional membership function formula.

Similarly, the total goodness-of-fit and average goodness-of-fit storage unit 7 of the ideal input pattern is determined by the second processor 8 so that the dots having the value of the goodness of fit of a predetermined reference value or more are converted into the ideal input pattern. One of the two values (for example, the binary value {0,
The calculated values of the total goodness-of-fit and the average goodness-of-fit are stored as the ideal input pattern corresponding to the pattern for collation, which is obtained by regarding 1} which is “ON” of 1}.

Fuzzy template parameter storage unit 9
Stores the parameters of a fuzzy template which serves as a reference for recognizing the input pattern. The fuzzy template parameters stored in the storage unit 9 can be manually adjusted and changed via a CRT used as a terminal device of a computer and a man-machine interface 10 such as a keyboard.

The three-dimensional membership function generator 11 is
Generate the membership functions needed to form a fuzzy template.

The fuzzy template parameter storage unit 9, the man-machine interface 10 and the three-dimensional membership function generator 11 constitute a collation pattern forming means.

The second processor 8 calculates the degree of conformity of the fuzzy template with respect to the dots of the input image as shown in FIG. 2 based on the parameters stored in the fuzzy template parameter storage unit 9, and the degree of fuzzy template conformance. It is sent to the storage unit 6. Further, the total goodness of fit and the average goodness of fit are calculated only for the dots having the goodness of fit of the reference value α (for example, 0.97) or more, and these values are stored in the total goodness of fit of the ideal input pattern and the average goodness of fit storage unit. Send to 7. The operation of the second processor 8 is performed when the apparatus is started up, and is not performed during pattern recognition.

The contents stored in the fuzzy template goodness-of-fit storage unit 6 and the total goodness-of-fit and average goodness-of-fit storage unit 7 of the ideal input pattern are stored in the first processor 3 when pattern recognition is actually performed. Used in processing by.
That is, the first processor 3 refers to the information in the storage units 6 and 7 as described above, performs the input pattern matching calculation, and sends the result to the signal conversion unit 4.

In the above embodiment, the two processors 3 and 8 are used as the arithmetic processing means, but in the actual device (hardware), one CP is used.
U (main processor) can realize the functions of these two processors.

Next, an example of pattern recognition according to the above embodiment will be described.

Taking the number "8" as an example, the fuzzy template is expressed by the above-mentioned equation (3).

Then, as shown in FIG. 2, horizontal 12 × vertical 2
When one dot is set on the xy coordinate plane and the fitness is calculated from the coordinates of the center point of each dot, the result shown in FIG. 3 is obtained. Then, these values are stored in the fuzzy template conformance storage unit 6 as elements of a 12 × 21 matrix. This is shown in FIG.

Next, when the reference value α is set to 0.97 and the conformity of 0.97 or more is extracted from the value shown in FIG. 3, the one enclosed by the frame in FIG. 5 is applicable. When this dot is regarded as the binary ON value (1) of the ideal input pattern, the ideal input pattern of "8" as shown in FIG. 6 is determined. For the dots at this time, the total goodness-of-fit and the average goodness-of-fit are calculated by the above-described matching calculation, and those values are stored in the total goodness-of-fit and average goodness-of-fit storage unit 7 of the ideal input pattern. This is shown in the figure as shown in FIG.

The calculation processing for pattern recognition is performed at high speed by referring to the values in the storage units 6 and 7 in the first processor 3.

[0046]

As described above, according to the present invention, the information necessary for pattern recognition is obtained in advance, and at the time of pattern recognition, the input pattern and the matching pattern are referred to by referring to these information. Since the comparison calculation is performed, the calculation processing time for pattern recognition can be significantly shortened, and the field of use of the pattern recognition apparatus to which the present invention is applied is expanded. Further, since an ideal input pattern can be automatically set in correspondence with the fuzzy template, the labor of the conventional manual operation is omitted and the operability is improved.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a fuzzy pattern recognition device according to the present invention.

FIG. 2 is a diagram showing coordinates of an input image.

FIG. 3 is a diagram showing a value of a goodness of fit of each dot representing a fuzzy template of a numeral “8”.

FIG. 4 is a diagram showing an example of a fuzzy template of “8”.

FIG. 5 is a diagram showing a degree of conformance that is equal to or greater than a reference value in FIG.

FIG. 6 is a diagram showing an example of an ideal input pattern of “8”.

FIG. 7 is a diagram showing a collation example using a fuzzy template corresponding to an input pattern.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Image reading device, 2 ... Signal conversion part, 3 ... 1st processor, 4 ... Signal conversion part, 5 ... Output device, 6 ... Fuzzy template matching degree storage part, 7 ... Total matching degree of an ideal input pattern, and Average goodness-of-fit storage unit, 8 ... second processor,
9 ... Fuzzy template parameter storage unit, 10 ... Man-machine interface, 11 ... Three-dimensional membership function generator.

Claims (2)

[Claims] 1. An apparatus for pattern recognition based on a membership function representing a fuzzy set, a pattern input section for inputting a pattern to be recognized, and one or more of the membership functions are generated and the membership functions are combined. As a result, a matching pattern forming unit that forms a matching pattern for recognizing a recognition target pattern input from the pattern input unit, and a recognition target pattern input from the pattern input unit by the matching pattern formation unit And a calculation processing means for calculating the certainty factor of the recognition by comparing with the matching pattern formed in 1., the calculation processing means is calculated by using a parameter of the matching pattern and a three-dimensional membership function. The suitability of each dot in the matching pattern is calculated in advance for all matching patterns and recorded. Keeping in mind, at the time of pattern recognition, by referring to the stored goodness of fit value,
A fuzzy pattern recognition device, which compares the pattern to be recognized with the matching pattern. 2. A pattern obtained by the calculation processing means by regarding a dot having a preliminarily obtained fitness value equal to or larger than a predetermined reference value as one of binary values of an ideal input pattern,
2. The fuzzy pattern recognition device according to claim 1, wherein a total goodness of fit and an average goodness of fit are obtained and stored as ideal input patterns corresponding to the matching pattern.