JP2010086324A - Binarization processing apparatus, information processing apparatus, binarization processing method, and binarization processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the binarization accuracy, by taking into account the degradation of the density distribution characteristics that depend on the location on the image of a barcode. <P>SOLUTION: An image density distribution detecting part 11 detects the density distribution of a character area separated from a barcode. A basic threshold determining part 13 and an auxiliary threshold determining part 14 obtain different thresholds from the density distribution obtained. A distribution characteristic determining part 12 determines the characteristics of the density distribution; determines which is to be applied as a target character area, the basic threshold or the auxiliary threshold; and makes a binarization threshold selecting part 15 select it. A binarization part 16 uses the threshold selected by the binarized threshold selecting part 15 and performs binarization. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a binarization processing apparatus, an information processing apparatus, a binarization processing method, and a binarization processing program for binarizing a barcode image, and in particular, creates a binarization pattern from an image captured by a general-purpose camera. The present invention relates to a binarization processing device, an information processing device, a binarization processing method, and a binarization processing program.

  Conventionally used barcodes are composed of a plurality of black bars and white bars arranged side by side. Among the plurality of bars, a combination of a specific number of black bars and white bars represents specific data. When the bar code indicates character information, this specific number is an area indicating one character (character), that is, a character area, and the entire bar code takes a form in which a plurality of character areas are arranged.

  When a barcode is recognized from a code image acquired by a digital camera or a scanner, data for one line along the long side direction of the barcode is cut out and recognition processing is performed. In this recognition, when a general-purpose optical device such as a digital camera is used, distortion may occur in the barcode image due to optical distortion. If the reading angle of the optical device is inclined with respect to the plane of the barcode, the read barcode image is distorted.

  If the length (width) of the bar on the barcode image varies depending on the location due to these distortions, accurate recognition processing cannot be performed in the processing for recognizing the entire barcode based on a single width. Therefore, conventionally, the barcode image is divided into small areas for each character, and the recognition process is performed by setting the bar width standard for each character, thereby suppressing the influence of the distortion of the bar width due to the position on the captured image. The recognition accuracy was improved.

JP 2004-185058 A

  By the way, when a barcode is read using a general-purpose optical device such as a digital camera, in addition to the above-described change in the bar width, there may be a case where degradation in resolution or luminance occurs due to the position on the captured image. is there.

  The degradation of resolution and luminance depending on the position on the image generally has an effect as it goes from the center of the captured image to the periphery. Therefore, particularly in a code image in which the entire length of the code extends to the peripheral part of the photographed image, the image characteristics of the photographed image central part and the peripheral part image are greatly different. The same phenomenon occurs when the reading angle of the optical device is inclined with respect to the plane of the barcode. Further, when a phenomenon occurs in which a part of the code is shaded and darkens, the resolution and luminance characteristics deteriorate in the same manner.

  When such a change in resolution or luminance occurs, an appropriate value of the binarization threshold, that is, a threshold for separating which range of the density distribution of the image is a black bar and which is a white bar also changes. However, in the conventional technique, the threshold value used for the process of binarizing the code image is obtained from the image characteristics of the entire code or determined according to the image characteristics for each character.

  In the former case, if it is determined according to the characteristics of the center of the captured image, which has relatively good characteristics, it does not conform to the deteriorated characteristics of the peripheral part, the binarization accuracy deteriorates, and as a result, the recognition rate decreases. It has occurred. Even in the latter case, since the threshold determination method is the same regardless of the position on the code image, there is a problem in that the consistency between the image characteristics and the threshold determination method is poor depending on the location, and the recognition rate decreases. It was.

  That is, in the conventional technique, the deterioration of the bar width depending on the position of the barcode on the image is considered, but the change in the density distribution depending on the position of the barcode on the image is not considered. There was a problem that binarization could not be performed with accuracy, and as a result, the code recognition rate might decrease.

  In recent years, the use of barcode recognition by mobile phone terminals and PDAs (Personal Digital Assistants) has been increasing in recent years. However, mobile phone terminals and PDAs do not have camera units specialized for barcode recognition, Therefore, it is required to take a barcode with a camera and perform recognition. Therefore, the deterioration of the density distribution characteristic depending on the position of the code image is remarkable, and it is an important problem to cope with this problem.

  The present invention has been made in order to solve the above-described problems in the prior art and to solve the problems. In consideration of deterioration of density distribution characteristics depending on the position of the barcode on the image, the binarization accuracy is improved. An object of the present invention is to provide a binarization processing device, an information processing device, a binarization processing method, and a binarization processing program.

  In order to solve the above-described problems and achieve the object, the present apparatus, method, and program provide the density of an image included in a divided area for a divided area created by dividing image data for one line of a barcode image. A feature of the distribution is detected, and a basic threshold value that is a first ratio defined in advance between a maximum value and a minimum value of density in the divided region, and an auxiliary threshold value that is a second ratio defined in advance. Then, based on the distribution characteristics in which the density distribution of the divided area is defined in advance, one of the basic threshold value and the auxiliary threshold value is selected as a threshold value used for binarization of the divided area.

  According to the present apparatus, method, and program, a binarization processing apparatus, an information processing apparatus, and a binarization process with improved binarization accuracy in consideration of deterioration of density distribution characteristics depending on the position of the barcode on the image The method and the binarization processing program can be obtained.

  Hereinafter, embodiments of a binarization processing apparatus, an information processing apparatus, a binarization processing method, and a binarization processing program according to the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a schematic configuration diagram illustrating a schematic configuration of an information processing apparatus having a binarization processing apparatus according to the present embodiment. The information processing apparatus illustrated in FIG. 1 includes a camera unit 30 and a code image processing apparatus 20. The code image processing device 20 further includes a line image extraction unit 21, a character region division unit 22, a processing target character selection unit 23, a binarization processing device 10, a code recognition processing unit 24, and a memory 25.

  The information processing apparatus is, for example, a mobile phone terminal or a PDA, and the camera unit 30 is a general-purpose optical device that may be shared with photography other than barcodes, for example, photography or video photography.

  The code image processing apparatus 20 performs processing for recognizing a barcode taken by the camera unit 30. FIG. 2 shows a specific example of a barcode processed by the code image processing apparatus 20. A barcode is an image that expresses information with a combination of two parallel lines of light and shade with different thicknesses and intervals. As shown in FIG. 2, there are various types of bar codes such as JAN code, NW7 code, code 39, and code 128. The ratio of white bar to black bar, the correspondence between code and character, etc. are determined for each type of code. It has been.

  The line image extraction unit 21 performs processing for scanning the barcode image data, which is an image obtained by photographing the barcode, and extracting image data for one line.

  The character area dividing unit 22 performs a process of dividing the area for each pattern constituting the character from the extracted image of one line. By this division processing, an area indicating one character (character), that is, a character area can be obtained as a divided area.

  The processing target area selection unit 23 performs a process of selecting a target area for recognition processing from a plurality of character areas (divided areas) obtained by division.

  The binarization processing device 10 binarizes the line image data of the region selected by the processing target region selection unit 23 to create a binarization pattern.

  Then, the code recognition processing unit 24 performs code recognition on the binarized pattern and converts the binarized pattern into character information. The memory 25 holds various data used by the code recognition processing unit 24, specifically, a monochrome / character pattern correspondence table 25a, a character / character information correspondence table 25b, and recognition result information 25c.

  Next, the configuration and operation of the binarization processing apparatus 10 will be described. When a barcode is read using a general-purpose optical device, there are inherent problems such as degradation in resolution and luminance due to position on the captured image. In general, these influences increase from the center of the captured image to the periphery.

  FIG. 3 is an explanatory diagram illustrating image quality deterioration due to camera characteristics. When the barcode image B10 is photographed, an amplitude sufficient to saturate both the white side and the black side can be obtained as the density gradation value B11 at the center of the image with good characteristics. On the other hand, the density gradation value B12 in the peripheral portion of the image has deteriorated characteristics and the amplitude becomes small.

  Such deterioration of characteristics is particularly noticeable when a barcode is photographed, which has a long overall code length and extends to the periphery of the photographed image. Similar deterioration of characteristics also occurs when the reading angle of the optical device is inclined with respect to the plane of the barcode.

  Further, as shown in FIG. 4, when a phenomenon occurs in which a part of the barcode is shaded and darkened, the same characteristic deterioration occurs. FIG. 4 shows a specific example when a barcode image B20 with a shadow on the right side is photographed. The tone gradation value B21 at the position where the shadow is not obtained can obtain an amplitude sufficient to saturate both the white side and the black side, but the tone gradation value B22 at the shadowed position becomes dark overall. The gradation value is generally a low value.

  In this way, considering that the gradation value of the density changes depending on the position on the barcode image, the binarization processing apparatus 10 determines the density gradation distribution for each character area, that is, for each divided area. The characteristic is discriminated, and a basic threshold value or an auxiliary threshold value is selected for binarization.

  As a specific configuration, as shown in FIG. 1, the binarization processing apparatus 10 includes an image density distribution detection unit 11, a distribution characteristic determination unit 12, a basic threshold determination unit 13, an auxiliary threshold determination unit 14, and the like. A binarization threshold selection unit 15 and a binarization unit 16 are included.

  The image density distribution detection unit 11 detects the characteristics of the density distribution of the image included in the divided area selected by the processing target area selection unit 23. Then, the distribution characteristic determination unit 12 performs a process of determining the characteristics of the detected light and shade distribution.

  The basic threshold value determination unit 13 performs a process of determining a binarization basic threshold value based on the detection result of the light and shade distribution. In addition, the auxiliary threshold value determination unit 14 performs processing for determining the auxiliary threshold value using a standard and method different from the basic threshold value.

  The binarization threshold selection unit 15 selectively adopts either the basic threshold or the auxiliary threshold as a threshold used for binarization based on the determination result by the distribution characteristic determination unit 12. Then, the binarization unit 16 uses the threshold value selected by the binarization threshold value selection unit 15 and uses the threshold value selected from the divided regions as black and the white portion as the portion where the gradation value is lower than the threshold value. create.

  FIG. 5 is an explanatory diagram for explaining an application example of the basic threshold value and the auxiliary threshold value. For the basic threshold value, for example, the midpoint of the maximum value (= white side) / minimum value (= black side) of the gradation value of the image data in the character is used as the threshold value. This is an appropriate threshold value for the gradation distribution B31 when the barcode is photographed with sufficient resolution and brightness as in the vicinity of the center of the photographed image.

  On the other hand, in the peripheral portion and shadow portion of the captured image where the barcode cannot be captured with sufficient brightness, the image data is captured closer to the dark side. In particular, in the case of a digital camera or the like, a characteristic is given so that the gradation value of the output image is originally non-linear with respect to the density of the input data. For example, when the output gradation value range is 0 to 255, the vicinity of the center of the gradation value In many cases, the gradation value of the output image with respect to the same density change is designed to be smaller in the vicinity of the maximum (= 255) / minimum (= 0) than (= 127). It is difficult to make it. In addition, when the density distribution is saturated on the dark side, it is difficult to perform binarization with the basic threshold value accurately. In FIG. 5, a basic threshold value, that is, a threshold value that divides the maximum value and the minimum value of the gradation on a one-to-one basis is used for the gradation distribution B32 in which the gradation value approaches the black side and is saturated on the black side. The threshold value is too large and the black pattern tends to become thick.

  For example, when the threshold value for dividing the maximum value and the minimum value of the gradation into 3 to 7 is used as the auxiliary threshold value and binarization is performed using the auxiliary threshold value, the gradation value is similar to the gradation distribution B32. The gradation distribution B33 whose value approaches the black side and is saturated on the black side can be appropriately binarized.

  The distribution characteristic discriminating unit 12 determines whether to use the basic threshold value or the auxiliary threshold value for the division area to be processed based on the distribution characteristic. Specifically, as shown in the gradation distribution B41 of FIG. 6, the distribution of gradation values of the image data is on either the maximum (= 255) side or the minimum (= 0) side of the gradation values. If it is saturated, an auxiliary threshold is used.

  Also, as shown in the gradation distribution B42 in FIG. 7, the maximum value of the gradation value is smaller than the median value (127 if 256 gradations) or the minimum value is larger than the median value. Use auxiliary threshold.

  As shown in the gradation distribution B43 in FIG. 8, the auxiliary threshold is also used when the difference between the maximum value and the minimum value of the gradation value is smaller than a predetermined specific level a.

  As the auxiliary threshold value, for example, a fixed threshold value that divides the maximum and minimum gradation values into 3 to 7 may be used. However, the auxiliary threshold value is set by using the characteristics determined for each barcode type. You may do it. In the case of a commonly used barcode, the ratio of the total length of the widths of black and white bars in a character is limited to a specific number of patterns. For example, in the case of a JAN code, any one of four patterns of black 2: white 5, black 3: white 4, black 4: white 3, and black 5: white 2 is used. Similarly, in the case of code 128, one of the three patterns of black 4: white 7, black 6: white 5, black 8: white 3 is used.

  The auxiliary threshold value determination unit 14 counts the number of pixels in order from the smallest pixel value in the gradation value distribution in the character area, and the number of pixels counted relative to the total number of pixels is When the ratio specified by the barcode type is reached, the gradation value of the pixel is set as an auxiliary threshold candidate.

  Therefore, for example, when it is assumed that the barcode is code 128, the auxiliary threshold value determination unit 14, as shown in FIG. 9, the auxiliary threshold value candidate THa that divides the distribution of gradation values into the ratio of black 4: white 7. Three auxiliary threshold candidates THb divided into the ratio of black 6: white 5 and THc divided into the ratio of black 8: white 3 are obtained.

  Then, the auxiliary threshold value determination unit 14 sets a value closest to the basic threshold value among the three auxiliary threshold value candidates, in FIG. 9, the auxiliary threshold value candidate THa as the auxiliary threshold value. Alternatively, the black / white bar width ratio closest to the basic threshold value may be selected.

  FIG. 10 is an explanatory diagram illustrating a case where the basic threshold is more effective than the auxiliary threshold. When the gradation value of the image data in the character area is uniform and the difference between the maximum value and the minimum value is sufficiently large as in the gradation distribution B61, an appropriate value cannot be determined using the auxiliary threshold value. There is a possibility that an accurate black-and-white pattern cannot be obtained. That is, when the pixel group having a high gradation value and the pixel group having a low gradation value are sufficiently separated from each other in the distribution, an appropriate threshold value is a gradation value existing between the two pixel groups. There is no pixel having a gradation value. If pixels are counted as described above for such a distribution and the value of a predetermined number of pixels is adopted as a threshold value, the threshold value is too close to either the upper or lower pixel group, which is inappropriate than the basic threshold value. It becomes.

  In particular, such a phenomenon is remarkable when the number of image data in the character area is small and the distribution is discrete. For this reason, when the gradation value is uniform and the difference between the maximum value and the minimum value is sufficiently large, an accurate binarization pattern can be obtained by selecting a basic threshold and using it for binarization. is there.

  On the other hand, as shown in the gradation distribution B62, when the difference between the maximum and minimum gradation values is small and there are intermediate gradation pixels, the auxiliary threshold value is more appropriate than the basic threshold value.

  Next, the processing operation of the code image processing apparatus 20 will be described with reference to FIG. FIG. 11 is a flowchart for explaining the processing operation of the code image processing apparatus 20. When a barcode image is input from the camera unit 30, first, the line image extraction unit 21 scans the barcode image and extracts image data for one line (S101).

  Next, the character area dividing unit 22 divides the image data for one line for each character area, and outputs the divided area (S102). The processing target area selection unit 23 sequentially selects a plurality of divided areas obtained by the division from the top as processing targets (S103).

  For the character region selected as the processing target, the image density distribution detection unit 11 creates a frequency histogram of the tone values of the image data (S104). Then, the distribution characteristic determination unit 12 confirms the distribution characteristic of the gradation value based on the information of the frequency value histogram (S105).

  When the distribution characteristic of the gradation value matches the assumption of the basic threshold value (S106, Yes), the basic threshold value determination unit 13 calculates the basic threshold value from the frequency value histogram (S107) and binarizes it. The threshold selection unit 15 selects a basic threshold (S108).

  On the other hand, if the distribution characteristic of the gradation value does not match the assumption of the basic threshold value (S106, No), the auxiliary threshold value determining unit 14 calculates the auxiliary threshold value from the frequency value histogram (S109). The value threshold selection unit 15 selects an auxiliary threshold (S110).

  The binarization unit 16 extracts the black and white pattern of the image data in each character area based on the threshold selected by the binarization threshold selection unit 15 and outputs it as a binarization pattern (S111).

  The code recognition processing unit 24 refers to the black-and-white / character pattern correspondence table 25a in the memory 25 to confirm the presence / absence of a character pattern corresponding to the binarized pattern, and displays the binarized pattern / character pattern correspondence table. If there is a corresponding character pattern, the character pattern is converted into character information with reference to the character-character information correspondence table 25b (S112).

  If a character pattern corresponding to the binarized pattern exists and can be converted into character information (S113, Yes), the character information is output as a successful process (S114). If the character pattern corresponding to the binarized pattern does not exist and cannot be converted into character information (No at S113), an error is output as a process failure (S115).

  Thereafter, the processing target area selection unit 23 checks whether the processing has been completed for all the character areas of the barcode (S116), and if there is an unprocessed character area remaining (S116, No), the processing target area (S103). Then, when the processing is completed for all the regions divided from the barcode image (S116, Yes), the processing ends.

  Next, a division method by the character area dividing unit 22 will be described. In the case of a commonly used barcode, the number of black: white bars in the character area is uniquely determined. For example, the JAN code is composed of two black bars and two white bars. Similarly, the NW7 code is composed of 4 black bars + 3 white bars, code 30 is composed of 5 black bars + 4 white bars, and code 128 is composed of 3 black bars + 3 white bars.

  In the conventional technique, the boundary position of the character is obtained by measuring the number of inflection points of the change rate of the gradation value of the image data at the boundary of the black and white bar. On the other hand, the character area dividing unit 22 obtains the black peak positions of the gradation values of the image data, measures the number thereof, and determines the character area based on the number of black bars defined in the barcode.

  Next, processing for a plurality of line images acquired from the barcode image will be described. Conventionally, when a code image obtained by photographing a barcode is recognized, data for one line along the long side direction of the barcode is cut out and a recognition process is performed. Conventionally, the recognition is successful only when all characters included in the recognition line can be correctly recognized. However, in this method, if a part of the code image has a defect or dust and it becomes impossible to recognize even one of the characters, the recognition has failed. Therefore, it is necessary to cut out a plurality of lines at different locations and repeat the recognition. For this reason, the problem is that the recognition process takes time.

  For example, when the recognition process for the barcode B70 shown in FIG. 12 is executed, if dirt or a defect occurs as in the code image B71, the first line becomes a line due to a recognition failure in the sixth character region caused by the defect. The entire recognition fails, and the recognition of the entire line fails due to the recognition failure in the fourth character area caused by the dirt on the second line. Therefore, the recognition of the entire line is successful only after the recognition is successful in all the character areas as in the third line.

  Further, as shown in FIG. 13, there is a problem that recognition cannot be performed when a defect or dust is generated across a plurality of lines so as to cross the short side direction of the barcode.

  In contrast to the conventional line processing, the code recognition processing unit 24 manages recognition success / failure for each character. Specifically, when there is a recognition failure character on one line, data for one line is cut out from another place and the recognition process is performed. At this time, the process for the character that has been recognized successfully in the previous process is not performed, and only the character that has failed to be recognized is recognized. The above processing is repeated, and as shown in FIG. 8, when all characters have been successfully recognized, the processing is terminated assuming that the barcode has been successfully recognized. As a result, the time required for the recognition process can be shortened. In addition, when a defect or dust is generated, the recognition succeeds, and the recognition performance is improved.

  In the conventional technique, it is generally assumed that the barcode is successfully recognized when the recognition success in units of lines is obtained a predetermined number of times or more. Correspondingly, in the recognition process in the code recognition processing unit 24, the number of successes is accumulated for each character area, the authentication is successful for the character area where the success has reached the predetermined number of times, and all the character areas are successfully authenticated. In this case, the authentication for the entire barcode may be successful.

  FIG. 15 is a flowchart for explaining the multiple line processing by the code image processing apparatus 20. First, image data for one line is extracted (S201), and code pattern recognition is performed (S202). This code pattern recognition is the operation described with reference to FIG.

  Thereafter, the code recognition processing unit 202 selects a processing target from the code pattern recognition result (S203), and reads past recognition result information if it exists in the memory 25 (S204). If the number of successful recognitions in the character area to be processed recorded in the recognition result information is greater than the predetermined number (S205, Yes), the process returns to the process target character selection (S203) and the next character area is selected.

  On the other hand, if the number of successful recognitions in the processing target character area recorded in the recognition result information is equal to or smaller than the predetermined number (S205, No), it is determined whether or not the current pattern conversion is successful (S206). If yes (S206, Yes), the recognition result information 25c in the memory 25 is updated (S207). If unsuccessful (S206, No), the process proceeds to step S208.

  In step S208, it is determined whether or not the processing for all the character areas has been completed (S208). If there is an unprocessed character area remaining (S208, No), the process returns to the process target character selection (S203) to return to the next character. Select an area.

  On the other hand, if the processing for all the character areas is completed (S208, Yes), the code recognition processing unit 24 refers to the recognition result information 25c, and the authentication success is obtained a predetermined number of times or more for all the character areas. If (S209, Yes), the code recognition is successful (S212), and the process is terminated.

  If there is still a character area that has not been successfully authenticated more than a predetermined number of times, the code recognition processing unit 24 determines whether the processing has been completed for all lines (S210), and unprocessed lines remain. If it is (S210, No), the process returns to step S201 to acquire the next line. On the other hand, when all the lines have been processed (S210, Yes), the code recognition processing unit 24 outputs a code recognition failure (S211) and ends the process.

  As described above, the binarization processing apparatus 10 uses two of the basic threshold value and the auxiliary threshold value that uses a calculation method suitable when the image characteristics are different. In addition, by selecting and using the appropriate one of the two thresholds according to the distribution characteristics of the gradation values of the image data for each character, even when the characteristics of the image differ depending on the position and shadow of the captured image The binarization pattern can be obtained accurately, and the recognition success rate is improved.

  Therefore, the reading performance is limited to the inherent problems such as the distortion of the captured image that occurs when reading a barcode using a general-purpose optical device such as a digital camera, the deterioration of the resolution due to the position on the captured image, and the decrease in luminance. Bar code reading without dropping can be realized.

  In addition, when the code image processing apparatus 20 repeats the recognition process by cutting out data for one line, the code image processing apparatus 20 stores the information of the character that has been successfully recognized in the previous process and uses it for the subsequent processes. Thus, the recognition time can be shortened, and the recognition performance when a defect or dust occurs is improved.

  In addition, a present Example is an example to the last, It can change and implement suitably, without being limited to description of a present Example. For example, after calculating both the basic threshold value and the auxiliary threshold value, the threshold value may be selected according to the distribution characteristics, or only the necessary threshold value may be calculated after selecting the threshold value to be used first. Further, the basic threshold value can always be calculated, and the auxiliary threshold value can be determined using the basic threshold only when necessary.

  Also, some or all of the various components shown in the present embodiment can be realized by software. Therefore, by realizing all the components of the binarization processing device 10 by software, a binarization processing program that causes the computer to function as the binarization processing device can be obtained, and all the components of the code image processing device 20 can be obtained. A code image processing program that causes a computer to function as a code image processing apparatus can be obtained by being realized by software.

  The following additional notes are further disclosed with respect to the embodiment including the above examples.

(Supplementary Note 1) For a divided area created by dividing image data for one line extracted from barcode image data obtained by shooting a barcode image into patterns constituting a character, an image included in the divided area Image density distribution detecting means for detecting the characteristics of the density distribution of
A basic threshold value determining means for determining, as a basic threshold value, a binarization threshold value between a maximum value and a minimum value of the density in the divided region, which is a first ratio defined in advance;
An auxiliary threshold value determining means for determining, as an auxiliary threshold value, a binarization threshold value at which the density distribution in the divided region is a second ratio defined in advance;
Distribution characteristic determining means for determining whether or not the density distribution of the divided region has a predetermined distribution characteristic;
A binarization threshold selection unit that selectively employs any one of the basic threshold and the auxiliary threshold as a threshold used for binarization of the divided region to be discriminated based on the distribution identification determination result; ,
Binarization means for binarizing the divided region to be discriminated using the adopted threshold;
A binarization processing apparatus comprising:

(Supplementary Note 2) The basic threshold value determining means determines, as the basic threshold value, a binarization threshold value where the density between the maximum value and the minimum value of the divided region is 1: 1.
The supplementary threshold value 1 is characterized in that the auxiliary threshold value determining means determines, as the auxiliary threshold value, a binarization threshold value at which the ratio of the density distribution in the divided region becomes a predetermined ratio for each type of the barcode. Binarization processing device.

(Supplementary Note 3) The auxiliary threshold value determining means obtains a plurality of auxiliary threshold value candidates whose density distribution of the divided regions has a plurality of predetermined ratios, and among the plurality of auxiliary threshold value candidates, The binarization processing apparatus according to appendix 1 or 2, wherein a near one is determined as the auxiliary threshold value.

(Supplementary Note 4) When one of the maximum value and the minimum value of the density distribution is saturated, the binarization threshold value selection unit deviates from an appropriate range in which at least one of the maximum value and the minimum value is set for each. Or the binarization according to any one of appendices 1 to 3, wherein the auxiliary threshold is employed when the difference between the maximum value and the minimum value is smaller than a predetermined specific value. Processing equipment.

(Appendix 5) Camera and
Image extracting means for scanning a barcode image photographed by the camera and extracting image data for one line;
Character area dividing means for dividing the image data for one line into patterns constituting a character to create a divided area;
Image gray level distribution detecting means for detecting the characteristics of the gray level distribution of the image included in each divided area;
A basic threshold value determining means for determining, as a basic threshold value, a binarization threshold value between a maximum value and a minimum value of the density in the divided region, which is a first ratio defined in advance;
An auxiliary threshold value determining means for determining, as an auxiliary threshold value, a binarization threshold value at which the density distribution in the divided region is a second ratio defined in advance;
Distribution characteristic determining means for determining whether or not the density distribution of the divided region has a predetermined distribution characteristic;
A binarization threshold selection unit that selectively employs any one of the basic threshold and the auxiliary threshold as a threshold used for binarization of the divided region to be determined based on the determination result of the distribution characteristics; ,
Binarization means for binarizing the division area to be discriminated using the adopted threshold and generating a binarization pattern;
An information processing apparatus comprising:

(Additional remark 6) The said character area division | segmentation means detects the line | wire with the higher density value formed in the bar of the said barcode image from the image data for said 1 line, and the number of the lines contained in each division | segmentation area | region is detected. The information processing apparatus according to appendix 5, wherein the information processing apparatus is divided so as to have a predetermined number.

(Supplementary note 7) For a plurality of lines of image data with different scanning positions extracted from the barcode image, a conversion result from the binary pattern to character information is obtained for divided areas corresponding to the same character area on different lines. A code recognition processing unit for accumulating and determining that the barcode has been successfully recognized when the number of successful conversions exceeds a predetermined reference number in all the character areas of the barcode image; The information processing apparatus according to appendix 5 or 6, characterized by:

(Supplementary Note 8) Characteristics of the density distribution of an image included in a divided region created by dividing image data for one line obtained by scanning a barcode image into patterns constituting a character An image density distribution detection step for detecting
A basic threshold value determining step for determining, as a basic threshold value, a binarization threshold value between a maximum value and a minimum value of the density in the divided region, which is a first ratio defined in advance;
An auxiliary threshold value determining step of determining, as an auxiliary threshold value, a binarization threshold value at which the density distribution in the divided region is a second ratio defined in advance;
A distribution characteristic determination step for determining whether or not the density distribution of the divided region has a predetermined distribution characteristic;
A binarization threshold selection step that selectively employs any one of the basic threshold and the auxiliary threshold as a threshold used for binarization of the divided region to be discriminated based on the distribution identification determination result; ,
A binarization step for binarizing the division area to be discriminated using the adopted threshold;
A binarization processing method characterized by comprising:

(Supplementary Note 9) Regarding the divided area created by dividing the image data for one line obtained by scanning the barcode image for each pattern constituting the character, the characteristics of the density distribution of the image included in the divided area Image density distribution detection procedure for detecting,
A basic threshold value determination procedure for determining, as a basic threshold value, a binarization threshold value between a maximum value and a minimum value of the density in the divided region, which is a first ratio defined in advance;
An auxiliary threshold determination procedure for determining, as an auxiliary threshold, a binarization threshold at which the density distribution in the divided region is a second ratio defined in advance;
A distribution characteristic determination procedure for determining whether or not the density distribution of the divided region has a predetermined distribution characteristic;
A binarization threshold selection procedure that selectively employs any one of the basic threshold and the auxiliary threshold as a threshold to be used for binarization of the divided region to be determined based on the determination result of the distribution specification; ,
A binarization procedure for binarizing the divided region to be discriminated using the adopted threshold;
A binarization processing program characterized by causing a computer to execute.

FIG. 1 is a schematic configuration diagram illustrating a schematic configuration of an information processing apparatus having a binarization processing apparatus according to the present embodiment. FIG. 2 is an explanatory diagram illustrating a specific example of a barcode. FIG. 3 is an explanatory diagram illustrating image quality deterioration due to camera characteristics. FIG. 4 is an explanatory diagram for explaining quality degradation when a part of the barcode is shaded and darkened. FIG. 5 is an explanatory diagram for explaining an application example of the basic threshold value and the auxiliary threshold value. FIG. 6 is an explanatory diagram for explaining a case where the gradation value is saturated to the minimum value side. FIG. 7 is an explanatory diagram for explaining a case where the maximum tone value is smaller than the median value. FIG. 8 is an explanatory diagram for explaining a case where the difference between the maximum value and the minimum value of the gradation values is smaller than a predetermined range. FIG. 9 is an explanatory diagram for explaining the setting of the auxiliary threshold value. FIG. 10 is an explanatory diagram for explaining a case where the basic threshold is valid. FIG. 11 is a flowchart for explaining the processing operation of the code image processing apparatus 20. FIG. 12 is an explanatory diagram for explaining the conventional line processing. FIG. 13 is an explanatory diagram for explaining a case where recognition cannot be performed by conventional line processing. FIG. 14 is an explanatory diagram for explaining line processing for managing recognition in character units. FIG. 15 is a flowchart for describing line processing for managing recognition in character units.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Binary processing apparatus 11 Image density distribution detection part 12 Distribution characteristic discrimination | determination part 13 Basic threshold value determination part 14 Auxiliary threshold value determination part 15 Binary value threshold selection part 16 Binary value part 20 Code image processing apparatus 21 Line image extraction part 22 Character region dividing unit 23 Processing target region selecting unit 24 Code recognition processing unit 25 Memory 25a Monochrome-character pattern correspondence table 25b Character-character information correspondence table 25c Recognition result information

Claims (7)

For the divided area created by dividing the image data for one line extracted from the barcode image data obtained by photographing the barcode image for each pattern constituting the character, the density distribution of the image included in the divided area Image density distribution detecting means for detecting features;
A basic threshold value determining means for determining, as a basic threshold value, a binarization threshold value between a maximum value and a minimum value of the density in the divided region, which is a first ratio defined in advance;
An auxiliary threshold value determining means for determining, as an auxiliary threshold value, a binarization threshold value at which the density distribution in the divided region is a second ratio defined in advance;
Distribution characteristic determining means for determining whether or not the density distribution of the divided region has a predetermined distribution characteristic;
A binarization threshold selection unit that selectively employs any one of the basic threshold and the auxiliary threshold as a threshold used for binarization of the divided region to be discriminated based on the distribution identification determination result; ,
Binarization means for binarizing the divided region to be discriminated using the adopted threshold;
A binarization processing apparatus comprising: The basic threshold value determining means determines, as the basic threshold value, a binarization threshold value that has a one-to-one relationship between the maximum value and the minimum value of the density of the divided region,
The auxiliary threshold value determining means determines, as the auxiliary threshold value, a binarization threshold value at which the ratio of the density distribution in the divided region is a predetermined ratio for each type of the barcode. The binarization processing apparatus described.   The auxiliary threshold value determining means obtains a plurality of auxiliary threshold value candidates in which the density distribution of the divided regions has a plurality of predetermined ratios, and among the plurality of auxiliary threshold value candidates, the one closest to the basic threshold value is determined. The binarization processing apparatus according to claim 1, wherein the binarization processing apparatus is determined as an auxiliary threshold value.   The binarization threshold selection means, when one of the maximum value and the minimum value of the concentration distribution is saturated, or when at least one of the maximum value and the minimum value deviates from the appropriate range determined for each Alternatively, when the difference between the maximum value and the minimum value is smaller than a predetermined specific value, the auxiliary threshold value is adopted. A camera,
Image extracting means for scanning a barcode image photographed by the camera and extracting image data for one line;
Character area dividing means for dividing the image data for one line into patterns constituting a character to create a divided area;
Image gray level distribution detecting means for detecting the characteristics of the gray level distribution of the image included in each divided area;
A basic threshold value determining means for determining, as a basic threshold value, a binarization threshold value between a maximum value and a minimum value of the density in the divided region, which is a first ratio defined in advance;
An auxiliary threshold value determining means for determining, as an auxiliary threshold value, a binarization threshold value at which the density distribution in the divided region is a second ratio defined in advance;
Distribution characteristic determining means for determining whether or not the density distribution of the divided region has a predetermined distribution characteristic;
A binarization threshold selection unit that selectively employs any one of the basic threshold and the auxiliary threshold as a threshold used for binarization of the divided region to be determined based on the determination result of the distribution characteristics; ,
Binarization means for binarizing the division area to be discriminated using the adopted threshold and generating a binarization pattern;
An information processing apparatus comprising: An image for detecting characteristics of density distribution of an image included in a divided area created by dividing image data for one line obtained by scanning a barcode image for each pattern constituting a character. A concentration distribution detection step;
A basic threshold value determining step for determining, as a basic threshold value, a binarization threshold value between a maximum value and a minimum value of the density in the divided region, which is a first ratio defined in advance;
An auxiliary threshold value determining step of determining, as an auxiliary threshold value, a binarization threshold value at which the density distribution in the divided region is a second ratio defined in advance;
A distribution characteristic determination step for determining whether or not the density distribution of the divided region has a predetermined distribution characteristic;
A binarization threshold selection step that selectively employs any one of the basic threshold and the auxiliary threshold as a threshold used for binarization of the divided region to be discriminated based on the distribution identification determination result; ,
A binarization step for binarizing the division area to be discriminated using the adopted threshold;
A binarization processing method characterized by comprising: An image for detecting characteristics of density distribution of an image included in a divided area created by dividing image data for one line obtained by scanning a barcode image for each pattern constituting a character. Concentration distribution detection procedure;
A basic threshold value determination procedure for determining, as a basic threshold value, a binarization threshold value between a maximum value and a minimum value of the density in the divided region, which is a first ratio defined in advance;
An auxiliary threshold determination procedure for determining, as an auxiliary threshold, a binarization threshold at which the density distribution in the divided region is a second ratio defined in advance;
A distribution characteristic determination procedure for determining whether or not the density distribution of the divided region has a predetermined distribution characteristic;
A binarization threshold selection procedure that selectively employs any one of the basic threshold and the auxiliary threshold as a threshold to be used for binarization of the divided region to be determined based on the determination result of the distribution specification; ,
A binarization procedure for binarizing the divided region to be discriminated using the adopted threshold;
A binarization processing program characterized by causing a computer to execute.

Images