JPH06348895A - Character segmenting method on x-ray film - Google Patents

Abstract

PURPOSE:To reduce the noise and thinning of a background generated in binary images and to easily segment characters by binarizing multi value image data after executing a first derivative processing. CONSTITUTION:The image information of an X-ray film as converted to the multi value image data 10 (data representing the density values of 256 gradations for instance) by using an image scanner or the like similarly to a conventional case. Then, first derivative is applied on the multi value image data 10 and first derivative image data 50 are obtained. The first derivative image data 50 obtained in such a manner are normalized and first derivative images for which the intensity of the first derivative is spuriously expressed by four gradations by the size of dots and a histogram for which the first derivative value is projected respectively in an (x) direction and in a (y) direction are obtained. Thus, at the time of obtaining binary image data 40 for segmenting the characters from the multi value image data 10 of a patient character information part printed on an X-ray filter, a binarization processing is performed after executing the pre-processing of the first derivative for the multi value image data 10.

Description

Detailed Description of the Invention

[0001]

FIELD OF THE INVENTION The present invention relates to an X-ray film patient I.
The present invention relates to a method for cutting out characters on an X-ray film in an automatic D recognition device.

[0002]

2. Description of the Related Art Conventionally, when reading an X-ray film image, the patient and other information have been manually input, but this is a very complicated work in a large hospital where 1000 or more X-ray images are generated per day. And was inefficient. One of the devices devised for this is a device for automatically recognizing and reading patient character information printed on an X-ray film, for example, the X-ray film disclosed in Japanese Utility Model Publication No. 63-61073. There is an automatic ID recognition device.

In the case of the apparatus disclosed in the above publication, the X-ray film ID is placed in a specific position on the X-ray film, for example, most of them are composed of handwritten characters, kanji of printed characters and kana mixed characters. Patient character information, which is image-related information such as the date of imaging, patient name, imaging part code, imaging order, etc., and this patient character information is burned in between the patient ID information consisting of a bar code and the character string end code. is there. In the case of the above-mentioned publication, the contents of the patient ID information and the position information are obtained by scanning this with a bar code, and the patient character information on the X-ray film is extracted from this position information. Patient character information is held as image information of the area. Further, the image data of the patient character information extracted from the X-ray film ID printed on the X-ray film is recognized as a character by the character recognition function and is stored in the database.

Conventionally, in this type of apparatus, as a method for cutting out a character portion of patient character information for recognizing a character from image data of patient character information (multivalued image data), as shown in a processing flow of FIG. Further, a method is used in which the multi-valued image data 10 captured by a scanner or the like is used as it is for binarization, and the binary image data 42 is used to cut out a character.

[0005]

However, according to the above-described character cutting method, the character portion on the X-ray film is binarized by using the multi-valued image data captured by the scanner or the like as it is. . For this reason, there is a problem in that an optimum threshold value for binarization cannot be selected due to the influence of unevenness of the X-ray film, and noise or blurring occurs.

Hereinafter, this problem will be specifically described with reference to FIGS. 3 to 5. FIG. 3 shows multi-valued image data 10 captured by a scanner or the like for character portions on X-ray film.
(For example, data representing density values of 256 gradations) is normalized, and the multivalued image 10 'in which the density values are pseudo-expressed in 4 gradations according to the dot size, and the density values in the x direction and y It is a figure which shows the histograms 20 and 30 projected in the direction. It can be seen that there is a bias in the density value in the background depending on the location. FIG. 4 is a diagram showing a result of binarizing the multi-valued image 10 'shown in FIG. 3 as it is with a threshold value 128, and a binary image 42 obtained by binarization, It is a figure which shows the histograms 22 and 32 which projected the density value in the x direction and the y direction, respectively.
In FIG. 4, some of the characters on the right side of the character string arranged in the x direction are blurred. FIG. 5 shows a binary image 44 obtained by binarizing the multi-valued image 10 shown in FIG. 3 as it is by further lowering the threshold value to 96 in order to eliminate the blurring of the characters generated in FIG. It is a figure which shows the histograms 24 and 34 which projected the density value in the x direction and the y direction, respectively. In FIG. 5, although the blur as shown in FIG. 4 has disappeared, the background appears as noise from the left side in the x direction. As described above, it is difficult to select the optimum threshold value for binarization by the conventional method of binarizing the multi-valued image data as it is, resulting in noise and blurring.

Therefore, an object of the present invention is to provide a binary image when the image data of the character portion is cut out in an apparatus for automatically recognizing the character portion by cutting out the character portion from the patient character information printed on the X-ray film. Another object of the present invention is to provide a method for extracting characters on an X-ray film, which can reduce noise and blurring caused by the unevenness of the X-ray film that has occurred.

[0008]

According to the method for cutting out characters on an X-ray film according to the present invention, a patient character information portion printed on an X-ray film is digitized and the character information portion is cut out to recognize a character. X-ray film patient ID
In the automatic recognition apparatus, when the multi-valued image data of the character information part is binarized and the character information part is cut out using this binary image data, the multi-valued image data including the character information part is pre-processed for the binarization. It is characterized in that the value image data is subjected to first-order differentiation.

In the character cutting method on the X-ray film, the preprocessing of the first-order differentiation is performed when the density value of the point (x, y) of the multivalued image data is f (x, y). It is preferable to apply the first-order differential defined by the following equation 4 to the multi-valued image data.

[0010]

[Equation 4]

In the first-order differential preprocessing, the density value at the point (x, y) of the multivalued image data is set to f (x, y), and x
When the function g (x) depending on is expressed by the following formula 5, the multi-valued image data can be subjected to the first-order differentiation defined by the formula 6.

[0012]

[Equation 5]

[0013]

[Equation 6]

[0014]

According to the method for cutting out characters on an X-ray film according to the present invention, it is possible to cut out a character from multi-valued image data of a patient character information portion printed on an X-ray film.
When the value image data is obtained, the binarization process is performed after the pre-processing of the first-order differential of the equation 4 is performed on the multi-value image data, so that it is possible to reduce background noise and character blurring. it can.

Further, by performing the first-order differential pre-processing using the equation (6) above, binary image data with reduced background noise and character blurring can be obtained even when the character spacing of the characters to be cut is narrow. be able to.

[0016]

EXAMPLE An example of a method for cutting out characters on an X-ray film according to the present invention will be described below in detail with reference to the accompanying drawings.

FIG. 1 is a process flow chart showing an outline of a method for cutting out characters on an X-ray film according to the present invention. In the following, one embodiment of a method of cutting out characters on an X-ray film will be briefly described according to the processing flow of FIG.

First, the image information of the X-ray film is converted into multi-valued image data (for example, data representing the density value of 256 gradations) 10 by using an image scanner or the like as in the conventional method. In FIG. 3, the multi-valued image data 10 is normalized, and the multi-valued image 10 'in which the density values thereof are pseudo-expressed in four gradations according to the dot size, and the density values are projected in the x direction and the y direction, respectively. Histograms 20 and 30.

Next, for this multi-valued image data 10, f
When (x, y) is the concentration value of the point (x, y), the first-order differentiation is performed using the equation defined by the following equation 7 and 1
Next differential image data 50 is obtained.

[0020]

[Equation 7]

In FIG. 6, the first-order differential image data 50 thus obtained is normalized in the same manner as in FIG. 3, and the intensity of the first-order differential is pseudo-expressed in four gradations according to the dot size. A first-order differential image 50 'and histograms 60 and 70 obtained by projecting the first-order differential values in the x-direction and the y-direction, respectively.

FIG. 6 shows that by performing the first-order differentiation process on the multi-valued image data 10, the background density unevenness which is a low frequency component is cut and only the outline of the character which is a high frequency component remains. There is. The histograms 20 and 30 shown in FIG. 3 and the histogram 6 shown in FIG.
Comparing with 0 and 70, it can be seen that the background density unevenness is significantly reduced. FIG. 7 shows the above 1
It is a figure which shows the result which binarized the secondary differential image data 50 by the threshold value 48, and is the binary image 52 obtained by binarization.
6A and 6B are diagrams showing histograms 62 and 72 obtained by projecting the density values in the x direction and the y direction, respectively. The binary image 52 does not have the blurring that appears in the binary image 42 shown in FIG. 4 and the background noise that appears in the binary image 44 shown in FIG. Therefore,
By using the projection result of the binarized image 52 shown in FIG. 7, it is possible to easily cut out characters on the X-ray film for character recognition.

As described above, according to the character segmentation method on the X-ray film according to the present invention, the primary differential processing is performed before the binarization of the multi-valued image data 10 to extract the contour of the character. Therefore, the image data 50 which is first-order differentiated by the equation (7)
The character of the binary image 52 obtained from the above is slightly larger than the character binarized by the conventional method. Therefore, when the character spacing is narrow and it is considered that the character may come into contact with an adjacent character, the following equation 8 may be used in advance.

[0024]

[Equation 8]

However, in the above equation, f (x, y) is the density value at the point (x, y), and g (X) is the equation defined by the following equation (9).

[0026]

[Equation 9]

[0027]

As is apparent from the above-described embodiments, according to the method of extracting characters on the X-ray film according to the present invention, the multi-valued image data is binarized after being subjected to the primary differential processing. Since the background density unevenness is removed, it is possible to reduce background noise and blurring that have occurred in the binary image by the conventional method, and it is possible to easily cut out characters.

Although the preferred embodiments of the present invention have been described above, the present invention is not limited to the above embodiments, and various design changes can be made without departing from the spirit of the present invention. Is.

[Brief description of drawings]

FIG. 1 is a process flow chart showing an embodiment of a method for cutting out characters on an X-ray film according to the present invention.

FIG. 2 is a processing flow chart showing a conventional method for cutting out characters on an X-ray film.

FIG. 3 is a multi-valued image in which multi-valued image data (for example, data representing a density value of 256 gradations) is normalized, and the density value is pseudo-expressed in 4 gradations according to the dot size, and the density thereof. It is a figure which shows the histogram which projected the value in the x direction and the y direction.

4 is a diagram showing a result of binarizing the multi-valued image shown in FIG. 3 as it is with a threshold value 128. The binarized image obtained by binarization and its density value are respectively x. It is a figure which shows the histogram projected in the direction and y direction.

5 is a diagram showing a result of binarizing the multi-valued image shown in FIG. 3 as it is with a threshold value 96. FIG. 5 shows a binarized image obtained by binarization and its density values in the x direction. It is a figure which shows and the histogram projected in the y direction.

FIG. 6 is an embodiment showing a method of cutting out characters on an X-ray film according to the present invention, in which the primary differential image data is normalized and the intensity of the primary differential is pseudo-fourth order according to the dot size. It is a figure which shows the 1st-order differential image represented by the key, and the histogram which projected the 1st-order differential value in the x direction and the y direction, respectively.

7 is a diagram showing a result of binarizing the first-order differential image of FIG. 6 with a threshold value 48, showing a binary image obtained by binarization and its density values in the x direction and the y direction, respectively. It is a figure which shows the histogram projected on.

[Explanation of symbols]

 10 Multi-valued image data 10 'Multi-valued image 20 Histogram 22 Histogram 24 Histogram 30 Histogram 32 Histogram 34 Histogram 42 Binary image data 44 Binary image 50 First derivative image data 50' First derivative image 52 Binary image 60 Histogram 62 Histogram 70 Histogram 72 Histogram

Claims (3)

[Claims] 1. An X-ray film patient ID automatic recognition apparatus which digitizes a patient character information portion printed on an X-ray film and cuts out the character information portion to recognize a character. When the data is binarized and the character information portion is cut out using the binary image data, the multi-valued image data including the character information portion is subjected to first-order differentiation as a pre-processing of the binarization. The method of extracting characters on X-ray film. 2. The pre-processing of the first-order differential is the number of the following equations for the multi-valued image data when the density value of a point (x, y) of the multi-valued image data is f (x, y). The method for extracting a character on an X-ray film according to claim 1, which comprises applying a first derivative defined by 1. [Equation 1] 3. In the preprocessing of the first-order differential, the density value of a point (x, y) of multivalued image data is set to f (x, y), and a function g (x) depending on x is expressed by The method of cutting out characters on an X-ray film according to claim 1, wherein the multi-valued image data is subjected to a first-order differential defined by the equation (3) when the equation (2) is used. [Equation 2] [Equation 3]