JP7283312B2 - Information reader and program - Google Patents

Description

The present invention relates to an information reading device and program for reading imaged character information and information codes.

In recent years, in addition to an information code reading function for reading information codes such as bar codes and QR codes (registered trademark), many information reading apparatuses have been used that have a character recognition function for recognizing imaged character information. Such an information reading device can also be realized by installing a predetermined application program in a personal mobile terminal such as a smart phone.

As an information reading device that reads both information codes and character information, for example, an optical character reading device disclosed in Patent Document 1 below is known. This optical character reader reads a barcode and the characters corresponding to the barcode located directly below the barcode. The position of the corresponding character is specified and processing for character recognition is performed.

JP-A-09-237312

By the way, when recognizing captured character information, if the captured image also contains a barcode, the bar portion of the barcode may be misidentified as an uppercase letter "I", a lowercase letter "l", or a number "1". There is a problem of storage. This problem occurs not only with bar codes but also with other types of information codes such as QR codes, and there is a possibility that character misrecognition may occur due to the information code imaged together with character information.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and its object is to provide a configuration capable of accurately recognizing character information from a captured image in which character information and an information code are captured. That's what it is.

In order to achieve the above object, the invention described in claim 1 of the scope of claims includes:
an imaging unit (23);
an information code reading unit (21) that performs a decoding process for reading the information code from the captured image obtained by capturing the information code by the imaging unit;
a character recognition unit (21) that performs character recognition processing for reading the character information from the captured image in which the character information is captured by the imaging unit;
An information reader (10) comprising
an image correction unit (21) for correcting the captured image so that, when the information code is read by the information code reading unit, a range corresponding to the information code in the captured image becomes a monochrome image range ;
a light/dark reversal determination unit that determines whether or not the information code read by the information code reading unit is a reversed code obtained by reversing the light/darkness of a light-colored cell and a dark-colored cell;
with
When the captured image is corrected by the image correction unit, the character recognition unit performs the character recognition process on the corrected captured image,
The information code reading unit classifies the cells into light-colored cells and dark-colored cells based on a predetermined brightness determination threshold value, and performs the decoding process,
The image correction unit
When the information code read by the information code reading unit is determined to be the inverted code by the light/dark reversal determining unit, the range corresponding to the information code in the captured image and the predetermined light/dark determination threshold value correcting the captured image so that the range where the brightness is low and the range of dark monochromatic images,
When the light/dark reversal determination unit determines that the information code read by the information code reading unit is not the reversed code, the range corresponding to the information code in the captured image is larger than the predetermined light/dark determination threshold value. Correcting the captured image so that the range where the brightness is high and the image range of a single light color system,
When the information code read by the information code reading unit is determined by the light/dark reversal determination unit to be the reverse code, the imaged image partially corrected to the dark monochromatic image range is When the performed character recognition processing fails, the range corresponding to the information code and the range in which the brightness is higher than the predetermined light/dark determination threshold in the captured image before correction become a single-color image in a bright color system. Correct the captured image so that it is within the range,
By determining that the information code read by the information code reading unit is not the reversed code by the light/darkness reversal determining unit, the imaged image partially corrected to the single color image range of the bright color system. If the performed character recognition processing fails, the range corresponding to the information code and the range in which the brightness is lower than the predetermined light/dark judgment threshold in the picked-up image before correction are dark monochromatic image ranges. It is characterized by correcting the captured image so that
It should be noted that the symbols in parentheses above indicate the corresponding relationship with specific means described in the embodiments to be described later.

In the invention of claim 1, when the information code is read by the information code reading unit, the range corresponding to the information code in the captured image captured by the imaging unit is a monochrome (for example, white) image range. The captured image is corrected by the image correction unit, and the character recognition unit performs character recognition processing on the corrected captured image.

As a result, even if an information code is included in the captured image captured for recognizing character information, the range corresponding to the information code is corrected to a monochrome image range, and the character recognition target is substantially the same. outside. As a result, misrecognition of characters due to the information code is suppressed, so that the character information can be accurately recognized from the captured image in which the character information and the information code are captured.
In particular, in the image correcting unit, when the information code read by the information code reading unit is determined by the light/dark reversal determination unit to be a reversed code, the range corresponding to the information code and the predetermined light/dark determination threshold value in the captured image are corrected. The captured image is corrected so that the range in which the lightness is lower than the range becomes a dark monochromatic (for example, black) image range. Further, in the image correcting unit, when the light/dark reversal determining unit determines that the information code read by the information code reading unit is not a reversed code, the range corresponding to the information code in the captured image and the predetermined light/dark determination threshold value The captured image is corrected so that the range in which the lightness is high is an image range of bright single color (for example, white).
The character information displayed together with the reverse code on the display medium is displayed in white characters, and there is a high possibility that the surroundings of the white characters will be a dark background. For this reason, in addition to the range corresponding to the reverse code, the range in which the brightness is lower than the predetermined brightness determination threshold in the captured image, that is, the dark background around the white character is converted into a dark monochrome image. By correcting to the range and substantially excluding the character recognition object, character misrecognition caused by not only the information code (inverted code) but also the pattern of the background can be easily suppressed. On the other hand, the character information displayed together with the normal information code in which the contrast between the light-colored cells and the dark-colored cells is not reversed on the display medium is displayed in black characters, and the black characters are surrounded by light-colored cells. likely to be the background of For this reason, in addition to the range corresponding to the information code, the range in which the brightness is higher than the predetermined brightness determination threshold value in the captured image, that is, the bright background around the black character is converted into a bright monochromatic image. By correcting to the range and substantially excluding the character recognition object, character misrecognition caused by not only the information code but also the pattern of the background can be easily suppressed. Therefore, it is possible to further improve the accuracy of character recognition by the character recognition unit.
Further, in the image correcting unit, the light/dark reversal determining unit determines that the information code read by the information code reading unit is a reversed code, so that a part of the captured image is corrected to a dark monochromatic image range. If the character recognition processing performed on the image fails, the range corresponding to the information code and the range where the brightness is higher than the predetermined brightness determination threshold in the captured image before correction are regarded as a bright monochrome image range. The captured image is corrected so that Also, in the image correcting unit, the light/dark reversal determining unit determines that the information code read by the information code reading unit is not a reversed code, so that a part of the picked-up image is corrected to a monochromatic image range of a bright color system. If the character recognition processing performed on the image fails, the range corresponding to the information code and the range in which the brightness is lower than the predetermined brightness determination threshold become a dark monochromatic image range in the captured image before correction. The captured image is corrected as follows.
As a result, the character information displayed in black characters along with the reverse code is imaged, and even if the character recognition fails because the background around the black characters is corrected to a dark monochromatic image range, the background remains bright. Since the character recognition process is performed again after being corrected to a monochromatic image range of the color system, misrecognition of characters caused by background patterns or the like can be further suppressed. On the other hand, the image of the character information displayed with white characters is imaged together with the normal information code in which the light-colored cells and dark-colored cells are not reversed, and the background around the white characters is light-colored. Even if character recognition fails because the image is corrected to a single-color image range, the background is corrected to a single-color image range and character recognition processing is performed again. can be further suppressed.

In the third aspect of the invention, when the code determination unit determines that a part of the information code is captured, the range corresponding to the part of the information code in the captured image is a monochrome image range. A captured image is corrected by the image correction unit.

As a result, not only the decoded part of the information code, but also the part of the information code that cannot be decoded but is recognized as part of the information code is virtually excluded from character recognition. can be further suppressed.

In the fourth aspect of the present invention, the code determination unit determines that part of the information code is captured by capturing an image of at least one of the three finder patterns and part of the timing pattern. Since the finder pattern and the timing pattern are formed so as to be easily recognizable from their purpose, even when a part of the QR code is imaged, the range corresponding to the part of the QR code can be easily and accurately grasped. be able to.

In the fifth aspect of the invention, character recognition processing is performed on the captured image based on the cell arrangement direction of the information code detected by the detection unit. As a result, when the character arrangement direction of the character information and the cell arrangement direction of the information code are the same, the character arrangement direction of the character information can be identified, so the character information can be recognized based on the detected cell arrangement direction. By performing character recognition processing on the captured image that has been rotation-corrected to facilitate recognition, the accuracy of character recognition by the character recognition unit can be further improved. Further, even when the character arrangement direction of the character information is deviated from the cell arrangement direction of the information code by a predetermined angle, the character information can be easily recognized on the basis of the direction deviated by the predetermined angle from the detected cell arrangement direction. By performing character recognition processing on the captured image that has undergone rotation correction, the character recognition accuracy of the character recognition unit can be further improved.

In the invention according to claim 2 , when the information code is read by the information code reading unit in the image correcting unit, the brightness difference in the picked-up image between the light-colored cells and the dark-colored cells constituting the information code is a predetermined brightness difference. When it is equal to or greater than the threshold, the captured image is corrected so that the range corresponding to the information code in the captured image becomes a monochrome image range.

For example, character information is printed on the front side of a print medium and an information code is printed on the back side. The information code may appear faintly blurred. In such a captured image, when at least part of the information code overlaps the character information and the decoding process is successful, the range corresponding to the information code is corrected to a monochrome image range. As a result, there is a problem that part of the character information is lost.

Therefore, when the brightness difference in the captured image between the light-colored cells and the dark-colored cells that constitute the read information code is equal to or greater than the predetermined brightness difference threshold value, the range corresponding to the information code in the captured image is The captured image is corrected so that it becomes a monochromatic image range. That is, in an imaging state in which the information code on the back side is faintly blurred as described above, the range corresponding to the information code is reduced by the brightness difference being less than the predetermined brightness difference threshold. It will not be subject to correction. Therefore, even when an unintended information code is imaged so as to be superimposed on character information in a faintly blurred state, loss of character information due to image correction can be prevented.

In the invention according to claim 6 , when the information code is read by the information code reading unit in the image correction unit, if the brightness difference is less than the predetermined brightness difference threshold value, the range corresponding to the information code in the captured image Among them, the captured image is corrected so that the range having the same brightness as the light-colored cells and the range having the same brightness as the dark-colored cells forming the information code are monochromatic image ranges.

As a result, in an imaging state in which the information code on the back side is faintly blurred as described above, the brightness difference becomes less than the predetermined brightness difference threshold value, so that the brightness constituting the information code is The captured image is corrected so that the range having the same brightness as the color cell and the range having the same brightness as the dark color cell are monochrome image ranges. For this reason, even if the unintended information code is captured in a faint and blurred state so that it overlaps with the character information, the clear character information is not corrected, and only the unclear and blurry information code portion is subject to correction. Therefore, the accuracy of character recognition by the character recognition unit can be further improved.

In the seventh aspect of the invention, the information code has an empty area different from the bright-color cells and the dark-color cells in the code area in which the plurality of bright-color cells and dark-color cells are arranged. Then, in the image correction section, when the information code reading section reads an information code having a blank area, the range corresponding to the code area in the captured image is excluded from the correction target.

As a result, when character information to be character-recognized is arranged in the empty area, it is possible to prevent the character information from being corrected so as to become a monochromatic image range together with the information code.

In the eighth aspect of the present invention, the information code has an empty area different from the bright-colored cells and the dark-colored cells in the code area in which the plurality of bright-colored cells and dark-colored cells are arranged. Then, when the information code reading unit reads an information code having a blank area, the image correcting unit adjusts the range of the captured image, excluding the blank area from the code area, to be a monochrome image range. The image is corrected.

Even in this way, when character information to be character-recognized is arranged in the empty area, it is possible to prevent the character information from being corrected so as to become a monochromatic image range. In particular, since the area excluding the empty area from the code area is corrected to be a monochrome image area, the area around the character information to be recognized becomes a monochrome image area. It can be improved further.

According to the ninth aspect of the invention, it is possible to realize a program having the same effect as the first aspect.
According to the tenth aspect of the invention, it is possible to realize a program having the same effect as the second aspect.

1 is an explanatory diagram schematically showing an information reading device according to a first embodiment, character information to be read, and an information code; FIG. 2 is a block diagram schematically showing an electrical configuration of an information reader; FIG. 4 is a flowchart illustrating the flow of reading processing performed by a control unit in the first embodiment; FIG. 4A is an explanatory diagram showing a captured image in which a bar code is captured in a decodable manner together with character information, and FIG. 4B is a corrected range corresponding to the bar code in FIG. FIG. 10 is an explanatory diagram showing a captured image; FIG. 5A is an explanatory diagram showing a captured image in which a part of a barcode is captured together with character information, and FIG. 5B is a range corresponding to a part of the barcode in FIG. 5A. is an explanatory diagram showing a captured image in which is corrected. FIG. 6(A) is an explanatory diagram showing a captured image of a barcode captured in a state where the margin is not recognized together with character information, and FIG. 6(B) is a range corresponding to the barcode in FIG. 6(A). is an explanatory diagram showing a captured image in which is corrected. FIG. 7A is an explanatory diagram showing a captured image in which a part of the QR code is captured together with character information, and FIG. 7B is a range corresponding to a part of the QR code in FIG. is an explanatory diagram showing a captured image in which is corrected. 9 is a flowchart illustrating the flow of reading processing performed by a control unit in the second embodiment; FIG. 9(A) is an explanatory diagram showing a captured image in which a reverse code is captured together with character information, and FIG. 9(B) is a corrected image range in which the reverse code, background, etc. are black in FIG. 9(A). FIG. 10 is an explanatory diagram showing a captured image that has been captured; FIG. 10A is an explanatory diagram showing a captured image in which a non-inverted code is captured together with character information, and FIG. 10B is an image range in which the non-inverted code and the background are white in FIG. FIG. 10 is an explanatory diagram showing a captured image corrected to . FIG. 4 is an explanatory diagram illustrating a state in which the cell arrangement direction of a barcode or QR code and the character arrangement information of character information are tilted with respect to the reading reference direction; FIG. 11 is a flowchart illustrating the flow of reading processing performed by a control unit in the third embodiment; FIG. FIG. 13A is an explanatory diagram showing a captured image in which the information code and the character information are captured so that the cell arrangement direction and the character arrangement direction are inclined at the same angle with respect to the reading reference direction, and FIG. 13B is an explanatory diagram showing a captured image in which the captured image of FIG. 13A is subjected to rotational correction or the like according to the cell arrangement direction; FIG. FIG. 14A is an explanatory diagram showing a captured image in which an information code and character information are captured such that the character arrangement direction is shifted by a predetermined angle from the cell arrangement direction, and FIG. 14B is a cell arrangement. FIG. 15 is an explanatory diagram showing a captured image in a state in which the captured image of FIG. 14A is rotationally corrected or the like according to the direction; FIG. 14 is a flowchart illustrating the flow of reading processing performed by a control unit in the fourth embodiment; FIG. FIG. 16A is an explanatory diagram showing a captured image in which a faintly blurred bar code is superimposed on character information, and FIG. 16B is a range corresponding to the bar code in FIG. 16A. FIG. 16C is an explanatory diagram showing a captured image in which is corrected, and FIG. is an explanatory diagram showing a captured image in which is corrected. FIGS. 17A and 17B are explanatory diagrams for explaining the main part of the fifth embodiment, FIG. 17A shows a captured image of an information code in which character information is arranged in an empty area, and FIG. 17(A) shows a state in which the range obtained by removing the blank area from the code area is corrected in the captured image of 17(A).

Hereinafter, a first embodiment embodying an information reading device and a program according to the present invention will be described with reference to the drawings.
As shown in FIG. 1, an information reading device 10 according to this embodiment is configured as a portable information processing terminal such as a smart phone that is carried by a user and used in various places. As shown in FIG. 2, the information reading device 10 includes a control unit 21 including a CPU, a memory 22, an imaging unit 23 configured as a camera, a touch panel 24 whose display contents are controlled by the control unit 21, and It is configured to include an operation unit 25 for outputting a signal to the control unit 21 in response to a touch operation or key operation, a communication unit 26 for communicating with an external device or the like.

The information reading device 10 according to the present embodiment is installed with a predetermined application program (hereinafter also referred to as a reading application), and reads the information code from the captured image in which the information code is captured by the imaging unit 23. and a function (OCR) for performing character recognition processing for reading the character information from the captured image in which the character information is captured by the imaging unit 23 . In particular, in the present embodiment, even if the information code is imaged when reading the character information by the reading process performed by the control unit 21 by activating the reading application, character misrecognition caused by the information code is suppressed. character information is recognized.

Hereinafter, the reading process performed by the control unit 21 in this embodiment will be described in detail with reference to the flowchart shown in FIG. In addition, as illustrated in FIG. 1, a case where "March 06, 2019" on which the barcode B is printed nearby on the display surface of the display medium R is recognized as the character information W will be described as an example.

When the reading application is activated in response to a predetermined operation on the operation unit 25 and the reading process is started by the control unit 21, the imaging process of step S101 in FIG. 3 is performed. In this process, the character information W is imaged by the image capturing unit 23 by performing a reading operation or the like on the operation unit 25 while the image capturing direction of the image capturing unit 23 is directed toward the character information W.

Subsequently, a light/dark determination threshold value setting process shown in step S103 is performed, and based on the light/dark distribution of the captured image captured in step S101, the information code is set to a white cell (light-colored cell) and a black cell (dark-colored cell). ), and a predetermined brightness determination threshold value is set as a reference for classifying the image. Then, in the determination processing in step S105, it is determined whether or not the information code is captured in the captured image captured as described above. Here, if the bar code B is captured decodably together with the character information W as in the captured image P1a shown in FIG. 4A, it is determined as Yes in step S105.

In this case, decoding processing shown in step S107 is performed, and processing for decoding the imaged information code is performed based on the brightness determination threshold value set as described above. Note that step S107 corresponds to an example of an "information code reading step", and the control unit 21 that performs the decoding process in step S107 can correspond to an example of an "information code reading unit".

If this decoding process succeeds (Yes in S109), the first image correction process shown in step S111 is performed. In this processing, the range corresponding to the successfully decoded information code in the imaged image captured as described above is corrected (replaced) with a monochrome image range so as to eliminate the color boundary with the surroundings. If the background of the barcode B is white, as in the captured image P1a shown in FIG. 4A, the image P1b shown in FIG. 4B is corrected by the first image correction process. In the previous captured image P1a, the range corresponding to the barcode B is corrected to a white image range. That is, the captured image is corrected so that the successfully decoded information code is removed. Note that step S111 and step S113, which will be described later, correspond to an example of the "image correction step". It can correspond to an example of an “image correction unit”.

Subsequently, character recognition processing shown in step S115 is performed. When a part of the image is corrected to a monochromatic image range as described above, character recognition processing using a known character recognition technology (OCR technology) is performed on the corrected captured image. In this way, since the character information W is recognized without the barcode B, misrecognition of characters caused by the barcode B can be eliminated. Note that step S115 described above corresponds to an example of a "character recognition step", and the control unit 21 that performs the character recognition process in step S115 can correspond to an example of a "character recognition unit".

On the other hand, as in the captured image P2a shown in FIG. 5A, when only a part of the barcode B is captured, the barcode can be obtained from the arrangement of bars (black cells) and spaces (white cells). Even if it is determined that the image has been captured (Yes in S105), the decoding process in step S107 fails (No in S109). Note that the control unit 21 that performs the determination process of step S105, which determines whether or not the information code is captured for the captured image for which the decoding process fails, does not include a part of the information code for the captured image for which the decoding process fails. It can correspond to an example of a “code determination unit” that determines whether or not is imaged.

In this case, the second image correction process shown in step S113 is performed. In this processing, the range corresponding to part of the barcode B (part of the information code) in the captured image captured as described above is corrected to a monochrome image range ( replaced). As in the imaged image P2a shown in FIG. 5(A), if a white background surrounds a part of the barcode B that has been recognized as an information code imaged and has failed to be decoded, As in the captured image P2b shown in (B), in the captured image P2a before correction, the range corresponding to part of the barcode B is corrected to a white image range (S113). That is, even if the decoding fails, if a part of the information code is imaged, the imaged image is corrected so that the entire imaged range of the information code is removed. Then, the character recognition process in step S115 is performed on the captured image corrected as described above. In addition, even if the barcode can be recognized as a barcode but decoding fails because some of the cells that make up the barcode B are missing, the range corresponding to the barcode B is corrected to a monochrome image range. (S113).

Also, as in the captured image P3a shown in FIG. 6A, even if the entire barcode B is captured, if the image is captured such that the necessary margin cannot be recognized, the decoding process fails (S109). No). Even in such a case, as in the captured image P3b shown in FIG. Character recognition processing in step S115 is performed on the picked-up image.

Further, when the information code is not imaged in the imaging process of step S101, the determination process of step S105 is determined as No, and the character recognition process of step S115 is performed without correcting the captured image. is done.

As described above, in the information reading device 10 according to the present embodiment, when the information code is read in the reading process performed by the control unit 21 (Yes in S109), the captured image captured by the imaging unit 23 is reproduced. The captured image is corrected so that the range corresponding to the information code is a monochrome image range (S111), and character recognition processing is performed on the corrected captured image (S115).

Then, the program according to the present embodiment causes the information reading device 10 having the imaging unit 23 to perform information code reading for performing decoding processing for reading the information code from the captured image in which the information code is captured by the imaging unit 23. When the information code is read by the step (S107) and the information code reading step, the image correction step (S111, S113) corrects the captured image so that the range corresponding to the information code in the captured image becomes a monochrome image range. ), and a character recognition step (S115) for performing character recognition processing for reading character information on the corrected captured image when the captured image is corrected by the image correction step (S115).

As a result, even if an information code is included in the captured image captured for recognizing character information, the range corresponding to the information code is corrected to a monochrome image range, and the character recognition target is substantially the same. outside. As a result, misrecognition of characters due to the information code is suppressed, so that the character information can be accurately recognized from the captured image in which the character information and the information code are captured.

Furthermore, when the code determination unit determines that a part of the information code is captured, the captured image is corrected so that the range corresponding to the part of the information code in the captured image becomes a monochrome image range. be done.

As a result, not only the decoded part of the information code, but also the part of the information code that cannot be decoded but is recognized as part of the information code is virtually excluded from character recognition. can be further suppressed.

In addition, the range corresponding to the information code or the range corresponding to part of the information code is not limited to being corrected to a white image range as a monochrome image range, but depending on the background color etc. around the information code Therefore, the correction may be made to a bright monochromatic image range or to a dark monochromatic image range.

Also, the information code imaged together with the character information W is not limited to the barcode B, and may be an information code of another code type such as a QR code. Even with such an information code, if the decoding is successful (Yes in S109), the range corresponding to the successfully decoded information code in the captured image is corrected to a monochrome image range, and decoded as described above. If decoding fails because the image is not properly captured (No in S109), the range corresponding to the information code for which decoding failed is corrected to a monochrome image range, thereby preventing character misrecognition caused by the information code. can be suppressed.

In particular, the QR code has a configuration in which finder patterns FP are arranged at the three corners of a rectangular code area and two timing patterns TP are arranged between the finder patterns FP and parallel to the outer edge of the code area. Therefore, when at least one of the three finder patterns FP and part of the timing pattern TP are captured as in the captured image P4a shown in FIG. 7A, part of the QR code is captured. As a result of the second image correction processing in step S113, the range corresponding to a part of the QR code in the captured image P4a before correction is changed to a single color (Fig. 7 ( In B), the image range is corrected to white). Here, the range corresponding to part of the QR code in the captured image can be obtained based on the positions and arrangement directions of the finder pattern FP and the timing pattern TP. In this way, the finder pattern FP and the timing pattern TP are formed so as to be easily recognizable for their purpose. It can be grasped easily and accurately.

[Second embodiment]
Next, an information reading device and program according to the second embodiment will be described with reference to the drawings.
The second embodiment is mainly different from the first embodiment in that the entire captured image including the information code is subject to correction. Therefore, the same reference numerals are assigned to substantially the same components as in the first embodiment, and the description thereof will be omitted.

A pattern or the like may be displayed as a background on the display surface of the display medium on which the character information and the information code are displayed. For this reason, for example, by correcting a range in which the lightness is higher than a predetermined value in the captured image so as to uniformly become a white image range, it is possible to remove bright-colored patterns and the like from the captured image. On the other hand, when the character information to be read is displayed in white characters, if the surroundings of the white characters are also corrected to be white, the characters cannot be recognized. Similarly, in order to remove dark patterns and the like from the captured image, when correcting the range where the lightness is lower than a predetermined value so that it becomes a uniformly black image range, the character information to be read is displayed in black characters. In some cases, if the periphery of the black character is also corrected to be black, the character cannot be recognized.

Therefore, in the reading process of the present embodiment, it is determined whether or not the read information code is a reversed code obtained by reversing the brightness of the light-colored cells and the dark-colored cells, and is determined to be the reversed code. In this case, the range in which the brightness is lower than the brightness determination threshold value in the captured image is corrected so as to become a dark monochromatic image range. This is because the character information displayed together with the reverse code on the display medium is displayed in white characters, and there is a high possibility that the surroundings of the white characters will be a dark background. Further, when it is determined that the read information code is not a reversed code, the range in which the brightness is higher than the brightness determination threshold value in the captured image is corrected so as to become a single-color image range of a bright color system. Character information displayed together with a normal information code (hereinafter also referred to as a non-inverted code) in which the contrast between light-colored cells and dark-colored cells is not reversed with respect to the display medium is displayed in black characters. This is because the surrounding area is likely to be a light-colored background.

Hereinafter, the reading process performed by the control unit 21 in this embodiment will be described in detail with reference to the flowchart shown in FIG.
As in the first embodiment, the brightness determination threshold value is set based on the brightness distribution of the captured image (S103 in FIG. 8), and if the decoding process is successful (Yes in S109), the determination process shown in step S117 , whether or not the successfully decoded information code is an inverted code. Note that the control unit 21 that performs the determination process of step S117 can correspond to an example of a "brightness/darkness reversal determination unit".

Here, since the successfully decoded information code is an inverted code, if it is determined as Yes in step S117, dark color image correction processing is performed in step S119. In this process, on the premise that the character information to be recognized is displayed in white characters, the range corresponding to the reverse code and the range in which the brightness is lower than the brightness judgment threshold in the captured image are converted into a single dark color. (for example, black). For example, as in the imaged image P5a shown in FIG. 9A, when the barcode B in which the brightness is reversed is imaged as a reversed code, the character information W to be recognized is assumed to be white characters. As a result of the dark color image correction processing in step S119, in the captured image P5a before correction, as in the captured image P5b shown in FIG. is corrected so that it becomes a black image range. This increases the contrast between the character information W and the background portion. Character recognition processing shown in step S121 is performed on the captured image corrected to increase the contrast between the character information W and the background portion.

If the character recognition process in step S121 is successful (Yes in S123), the reading process ends. On the other hand, if the character recognition process in step S121 fails (No in S123), the character information to be recognized is assumed to be a black character instead of a white character, and the bright color image correction process in step S125 is performed. In the captured image captured in S101, the range corresponding to the reversed code and the range in which the brightness is higher than the brightness determination threshold are corrected so as to become a single bright color (for example, white) image range. Then, character recognition processing shown in step S127 is performed on the captured image corrected in this way.

Further, since the successfully decoded information code is a non-inverted code, if the determination in step S117 is No, bright color image correction processing is performed in step S129. In this processing, on the premise that the character information to be recognized is displayed in black characters, the range corresponding to the non-inverted code and the range in which the brightness is higher than the brightness judgment threshold in the captured image are divided into bright single-color images. (e.g. white) image range. For example, as in the captured image P6a shown in FIG. 10A, when the barcode B is captured as a non-inverted code, the character information W to be recognized is assumed to be black characters, and the light color By the system image correction processing, in the captured image P6a before correction, as in the captured image P6b shown in FIG. It is corrected so that it becomes a white image range. This increases the contrast between the character information W and the background portion. Character recognition processing shown in step S131 is performed on the captured image corrected to increase the contrast between the character information W and the background portion.

If the character recognition process in step S131 is successful (Yes in S133), the reading process ends. On the other hand, if the character recognition process in step S131 fails (No in S133), the character information to be recognized is assumed to be white characters instead of black characters, and the dark color image correction process in step S135 is performed to correct the character information in step S101. In the captured image captured in , the range corresponding to the non-inverted code and the range in which the brightness is lower than the brightness determination threshold are corrected so as to become a dark monochromatic (for example, black) image range. Then, character recognition processing shown in step S137 is performed on the captured image corrected in this way.

As described above, in the information reading apparatus 10 according to the present embodiment, when the read information code is determined to be a reversed code in the reading process performed by the control unit 21, the reversed code is displayed in the captured image. The captured image is corrected so that the range corresponding to the code and the range in which the lightness is lower than a predetermined brightness determination threshold are dark monochromatic image ranges. Further, when it is determined that the read information code is not the inverted code, the area corresponding to the non-inverted code and the area having the brightness higher than the predetermined brightness determination threshold value in the imaged image are bright and monochromatic. The captured image is corrected so as to be within the image range.

In this way, the character information displayed together with the reverse code on the display medium is displayed in white characters, and the background of the white characters is likely to be a dark background, so it corresponds to the reverse code. In addition to the range, the range in which the brightness is lower than the predetermined brightness determination threshold in the captured image, that is, the dark background around the white character is corrected to a dark single color image range and substantially By not recognizing characters, misrecognition of characters due to not only reversed codes but also background patterns can be suppressed. On the other hand, the character information displayed together with the non-inverted code on the display medium is displayed in black characters, and the area around the black characters is likely to be a bright background, so the range corresponding to the non-inverted code In addition, in the captured image, the range in which the brightness is higher than a predetermined brightness determination threshold, that is, the bright background around the black character is corrected to a bright monochromatic image range, and the character is substantially reproduced. By excluding characters from the recognition target, misrecognition of characters due to not only non-inverted codes but also background patterns and the like can be suppressed. Therefore, it is possible to further improve the accuracy of character recognition by character recognition processing.

In particular, when it is determined that the read information code is a reversed code, the character recognition processing performed on the captured image partially corrected to a dark monochromatic image range fails (No in S123). ), in the captured image before correction, the captured image is corrected so that the range corresponding to the information code and the range in which the brightness is higher than a predetermined brightness determination threshold become a single-color image range of a bright color system. . Further, when it is determined that the read information code is not the reversed code, and the character recognition processing performed on the picked-up image partially corrected to a bright monochromatic image range fails (No in S133). ), the captured image before correction is corrected so that the range corresponding to the information code and the range in which the brightness is lower than a predetermined threshold value for determining lightness are dark monochromatic image ranges.

As a result, the character information displayed in black characters along with the reverse code is imaged, and even if the character recognition fails because the background around the black characters is corrected to a dark monochromatic image range, the background remains bright. Since the character recognition process is performed again after being corrected to a monochromatic image range of the color system, misrecognition of characters caused by background patterns or the like can be further suppressed. On the other hand, the character information displayed in white characters along with the non-reversed code was imaged, and character recognition failed because the background around the white characters was corrected to a bright single color image range. However, since the background is corrected to a single dark color image range and the character recognition process is performed again, character misrecognition caused by the pattern of the background can be further suppressed.

[Third Embodiment]
Next, an information reading device and program according to the third embodiment will be described with reference to the drawings.
The third embodiment is mainly different from the first embodiment in that the character recognition process is performed on the captured image based on the cell arrangement direction of the information code. Therefore, the same reference numerals are assigned to substantially the same components as in the first embodiment, and the description thereof will be omitted.

Information codes such as barcodes and QR codes can be decoded without any problem regardless of the orientation of the cell arrangement direction F1 in the captured image. On the other hand, character information may be erroneously recognized depending on the character arrangement direction F2 in the captured image. For example, like the captured image P7 illustrated in FIG. 11, decoding can be performed without problems regardless of the orientation of the cell array direction F1 of the barcode B or the cell array direction F1 of the QR code Q in the captured image. On the other hand, as illustrated in FIG. 11, the character arrangement direction F2 of "177", which is the character information W, is rotated 180° with respect to the reading reference direction (see the X direction in FIG. 11) in the captured image. , there is a possibility that a character misrecognition such as "LLl" will occur.

Therefore, when character information is imaged and read, it is necessary to limit the imaging direction so that the character arrangement direction F2 of the character information is close to 0° with respect to the reading reference direction in the captured image. be. Then, even if an information reader reads an information code, when character information is also read, it is necessary to take an image of the read target while being aware of the orientation of the read target. The problem arises that the

Therefore, in the reading process according to the present embodiment, on the premise that the character arrangement direction F2 of the character information and the cell arrangement direction F1 of the information code are the same, the picked-up image in which the information code and the character information are picked up is read as the information. The picked-up image is rotationally corrected so that the code cell arrangement direction F1 coincides with the reading reference direction, and character recognition processing is performed on this rotationally-corrected picked-up image.

Hereinafter, the reading process performed by the control unit 21 in this embodiment will be described in detail with reference to the flowchart shown in FIG.
As in the first embodiment, the brightness determination threshold value is set based on the brightness distribution of the captured image (S103 in FIG. 12), and if the decoding process is successful (Yes in S109), the cell arrangement direction is detected in step S141. processed. In this process, the cell arrangement direction F1 of the successfully decoded information code is detected. Note that the control unit 21 that performs the cell arrangement direction detection process in step S141 can correspond to an example of a "detection unit".

Subsequently, a third image correction process shown in step S143 is performed. In this process, the picked-up image is rotationally corrected so that the cell arrangement direction F1 detected as described above coincides with the reading reference direction, and the range corresponding to the information code in the picked-up image corrected for rotation is out of the surroundings. It is corrected (replaced) to a monochrome image area to eliminate color boundaries. Then, in the character recognition processing shown in step S115, character recognition processing is performed on the picked-up image that has undergone rotation correction or the like as described above.

For this reason, for example, as in a captured image P8a shown in FIG. 13A, the barcode B and the character information W are arranged such that the cell arrangement direction F1 and the character arrangement direction F2 are inclined at the same angle with respect to the reading reference direction. When the image is captured, the captured image P8a is rotationally corrected such that the cell arrangement direction F1 coincides with the reading reference direction, as in the captured image P8b shown in FIG. 13B. A range corresponding to the information code in the captured image is corrected to a monochrome image range (white in FIG. 13B) (S143). Character recognition processing is then performed on the picked-up image that has undergone rotation correction or the like in this way.

As described above, in the information reader 10 according to the present embodiment, the cell arrangement direction F1 of the information code detected by the cell arrangement direction detection process is used as a reference, and the cell arrangement direction F1 coincides with the reading reference direction. Character recognition processing is performed on the picked-up image that has undergone rotation correction or the like. As a result, when the character arrangement direction F2 of the character information and the cell arrangement direction F1 of the information code are the same, the character arrangement direction F2 of the character information can be specified. Character recognition accuracy can be further improved by performing character recognition processing on a captured image that has undergone rotation correction or the like so that character information can be easily recognized.

As a first modification of the present embodiment, when it can be grasped in advance that the character arrangement direction F2 of the character information is deviated from the cell arrangement direction F1 of the information code by a predetermined angle θ, the predetermined angle θ The captured image may be rotated and corrected by using this. That is, the character recognition accuracy can be further improved by performing character recognition processing on the picked-up image that has undergone rotation correction or the like so that the direction shifted by the predetermined angle θ from the detected cell arrangement direction F1 coincides with the reading reference direction. It can be improved further.

Specifically, for example, like a captured image P9a shown in FIG. 14A, the barcode B and the character information W are imaged such that the character arrangement direction F2 is deviated from the cell arrangement direction F1 by a predetermined angle θ. 14B, the captured image P8a is changed so that the direction deviated from the detected cell arrangement direction F1 by the predetermined angle θ coincides with the reading reference direction, as in the captured image P8b shown in FIG. Rotationally corrected. Then, the range corresponding to the information code in this rotation-corrected captured image is corrected to a monochrome image range (S143), and character recognition processing is performed on the thus corrected captured image. In this manner, the character recognition accuracy can be further improved by performing the character recognition processing on the picked-up image that has undergone rotation correction or the like so that the character information can be easily recognized in consideration of the predetermined angle θ.

Further, as a second modification of the present embodiment, even if decoding fails in the reading process (No in S109), if the cell arrangement direction of the failed decoding can be detected, as described above, the detected Character recognition accuracy can be further improved by performing character recognition processing on captured images that have undergone rotation correction or the like so that character information can be easily recognized based on the cell arrangement direction.

The characteristic configuration of the present embodiment and modifications, etc., in which character recognition processing is performed on a captured image that has undergone rotation correction or the like with reference to the cell arrangement direction F1 of the information code, can also be applied to other embodiments.

[Fourth embodiment]
Next, an information reading device and program according to a fourth embodiment will be described with reference to the drawings.
The fourth embodiment is mainly different from the above-described first embodiment in that an information code with a small lightness difference is excluded from correction targets. Therefore, the same reference numerals are assigned to substantially the same components as in the first embodiment, and the description thereof will be omitted.

For example, character information is printed on the front side of a print medium and an information code is printed on the back side. The information code may appear faintly blurred. Further, when character information displayed on one surface of a transparent member such as a PET bottle is imaged and character recognition is performed, the information code on the opposite surface of the transparent member may appear in the captured image in a faintly blurred state. may be lost.

In such a captured image, when at least part of the information code is superimposed on the character information and the decoding process is successful in the reading process performed by the control unit 21, it corresponds to the information code. Since the range is corrected to a monochromatic image range, there is a problem that part of the character information is lost. For example, as shown in FIG. 16A, the captured image P10a is captured such that the faintly blurred barcode B overlaps the character information W. If the barcode B is successfully decoded, the image shown in FIG. , part of the character information W is lost because the range corresponding to the barcode B is corrected to a white image range in the first image correction process in step S111.

Therefore, in the present embodiment, when the difference in brightness (difference in brightness of the information code) in the captured image between the light-colored cells and the dark-colored cells that constitute the read information code is equal to or greater than a predetermined brightness difference threshold, the image is captured. The captured image is corrected so that the range corresponding to the information code in the image becomes a monochrome image range. That is, in an imaging state in which the information code on the back side is faintly blurred as described above, the difference in brightness of the information code is less than the predetermined light-dark difference threshold, so that it corresponds to the information code. Exclude the range from correction.

Hereinafter, the reading process performed by the control unit 21 in this embodiment will be described in detail with reference to the flowchart shown in FIG.
As in the first embodiment, the brightness determination threshold value is set based on the brightness distribution of the captured image (S103 in FIG. 15), and if the decoding process is successful (Yes in S109), the determination process shown in step S151 is performed. be. In this determination process, the difference in brightness (e.g., PCS (Print Contrast Signal) value) in the captured image between the bright-colored cells and the dark-colored cells that constitute the successfully decoded information code is obtained. A determination is made as to whether the difference is greater than or equal to a predetermined contrast threshold.

Here, for example, as shown in FIG. 4A, since a clear barcode B is imaged, if the difference in brightness of the information code increases and exceeds the predetermined brightness difference threshold value, the process proceeds to step S151. is determined to be Yes, and the processing from step S111 onward is performed.

On the other hand, for example, as shown in FIG. 16(A), since the barcode B is imaged in a dimly blurred state, if the brightness difference of the information code becomes lower than the predetermined brightness difference threshold, , is determined as No in step S151. In this case, the character recognition process of step S115 is performed without performing the first image correction process of step S111. In other words, the barcode B that is imaged in a dimly blurred state is excluded from the correction target.

As described above, in the present embodiment, when the brightness difference of the information code is equal to or greater than the predetermined brightness difference threshold, the range corresponding to the information code is subject to correction, and the brightness difference of the information code reaches the predetermined brightness difference threshold. If it is less than that, the range corresponding to that information code will not be corrected. Therefore, even when an unintended information code is imaged so as to be superimposed on character information in a faintly blurred state, loss of character information due to image correction can be prevented.

As a modification of the present embodiment, when the brightness difference of the information code is less than the predetermined brightness difference threshold, the bright color system constituting the information code is selected from the range corresponding to the information code in the captured image. The captured image may be corrected so that the range having the same brightness as the cell and the range having the same brightness as the dark-colored cell are monochrome image ranges. In this case, if the captured image P10a illustrated in FIG. 16A is captured, the light color system of the barcode B in the captured image P10a is displayed as in the captured image P10c illustrated in FIG. 16C. The captured image is corrected so that the range having the same brightness as the cell and the range having the same brightness as the dark-colored cell become a monochromatic image range.

As a result, even if the unintended information code is imaged in a faintly blurred state so that it overlaps with the character information, the clear character information is not corrected, and only the unclear and blurry information code portion is subject to correction. Therefore, the accuracy of character recognition by the character recognition unit can be further improved.

Note that the characteristic configuration of the present embodiment, in which information codes with small lightness differences are excluded from the correction target, and the range in which the lightness is the same as that of the light-colored cells of the information code and the range in which the lightness is the same as that of the dark-colored cells are monochrome images. The characteristic configuration of the modified example in which the captured image is corrected so as to be within the range can also be applied to other embodiments.

[Fifth embodiment]
Next, an information reading device and program according to the fifth embodiment will be described with reference to the drawings.
The fifth embodiment is mainly different from the first embodiment in that character information to be character-recognized is arranged in the code area of the information code. Therefore, the same reference numerals are assigned to substantially the same components as in the first embodiment, and the description thereof will be omitted.

The character information to be recognized is not limited to being arranged near the information code, but may be arranged within the code area of the information code formed by arranging a plurality of bright-colored cells and dark-colored cells. For example, like the information code C of the captured image P11a illustrated in FIG. Character information W may be arranged.

When the character information arranged in the code area of the information code is imaged in this way, in the reading process in the first embodiment, the range corresponding to the successfully decoded information code is corrected to a monochrome image range. , the character information in the code area is also corrected to a monochrome image range.

Therefore, in the reading process according to the present embodiment, in the first image correction process in step S111, when an information code having a blank area is read, the range corresponding to the code area is removed from the correction target in the captured image. exclude. For this reason, for example, in an information code having a vacant area, information indicating that the vacant area is present and information indicating the range of the vacant area occupied in the code area are recorded in a readable manner so that the read information code becomes empty. The information reader 10 can easily grasp whether or not there is an area, the range of the empty area, and the like.

As a result, when character information to be character-recognized is arranged in the empty area, it is possible to prevent the character information from being corrected so as to become a monochromatic image range together with the information code.

As a modification of the present embodiment, when an information code having a blank area is read, in the first image correction process in step S111, the range of the captured image excluding the blank area from the code area is a monochrome image range. The captured image may be corrected such that In this case, when the captured image P11a illustrated in FIG. 17A is captured, the code area Ca to the empty area Cb in the captured image P11a are separated from each other like the captured image P11b illustrated in FIG. 17B. The captured image is corrected so that the range excluding is a monochrome image range.

Even in this way, when character information to be character-recognized is arranged in the empty area, it is possible to prevent the character information from being corrected so as to become a monochromatic image range. In particular, since the range excluding the blank area from the code area is corrected to be a monochrome image range, the periphery of the character information to be recognized becomes a monochrome image range (see FIG. 17B). Character recognition accuracy by the character recognition unit can be further improved.

In addition, when an information code having an empty area is read, the range corresponding to the code area is excluded from the correction target, and when an information code having an empty area is read, an empty area is removed from the code area. The characteristic configuration of the modified example in which the captured image is corrected so that the excluded range becomes a monochromatic image range can also be applied to other embodiments.

It should be noted that the present invention is not limited to the above embodiments and modifications, and may be embodied as follows, for example.
(1) The present invention may be applied to an information reader that reads only bar codes among information codes. In this case, the decoding process in the reading process is not limited to decoding only barcodes, and one-dimensional codes of other code types may be decoded. This is because the time required to decode a one-dimensional code is shorter than that of a two-dimensional code, so even if the one-dimensional code of all code types is decoded, the reading time of the main reading process is less affected. .

(2) In the reading process, the decoding result of the decoding process may be used regardless of the recognition result of the character recognition process.

(3) The present invention is not limited to being applied to an information reading device configured by installing the reading application in a portable information processing terminal such as a smartphone. It may be applied to an information reading device configured by installing the reading application or the like in a mobile terminal having other functions such as a wireless tag processing function capable of reading and writing tags, or a stationary information reading device. Apparatus may be applied.

10... Information reading device 21... Control unit (information code reading unit, character recognition unit, code determination unit, light/dark reversal determination unit, detection unit)
23... Imaging unit B... Bar code (information code)
F1... Cell arrangement direction F2... Character arrangement direction W... Character information

Claims (10)

an imaging unit;
an information code reading unit that performs a decoding process for reading the information code from the captured image in which the information code is captured by the imaging unit;
a character recognition unit that performs character recognition processing for reading the character information from an image obtained by capturing the character information by the imaging unit;
An information reader comprising
an image correction unit that corrects the captured image so that, when the information code is read by the information code reading unit, a range corresponding to the information code in the captured image becomes a monochrome image range ;
a light/dark reversal determination unit that determines whether or not the information code read by the information code reading unit is a reversed code obtained by reversing the light/darkness of a light-colored cell and a dark-colored cell;
with
When the captured image is corrected by the image correction unit, the character recognition unit performs the character recognition process on the corrected captured image ,
The information code reading unit classifies the cells into light-colored cells and dark-colored cells based on a predetermined brightness determination threshold value, and performs the decoding process,
The image correction unit
When the information code read by the information code reading unit is determined to be the inverted code by the light/dark reversal determining unit, the range corresponding to the information code in the captured image and the predetermined light/dark determination threshold value correcting the captured image so that the range where the brightness is low and the range of dark monochromatic images,
When the light/dark reversal determination unit determines that the information code read by the information code reading unit is not the reversed code, the range corresponding to the information code in the captured image is larger than the predetermined light/dark determination threshold value. Correcting the captured image so that the range where the brightness is high and the image range of a single light color system,
When the information code read by the information code reading unit is determined by the light/dark reversal determination unit to be the reverse code, the imaged image partially corrected to the dark monochromatic image range is When the performed character recognition processing fails, the range corresponding to the information code and the range in which the brightness is higher than the predetermined light/dark determination threshold in the captured image before correction become a single-color image in a bright color system. Correct the captured image so that it is within the range,
By determining that the information code read by the information code reading unit is not the reversed code by the light/darkness reversal determining unit, the imaged image partially corrected to the single color image range of the bright color system. If the performed character recognition processing fails, the range corresponding to the information code and the range in which the brightness is lower than the predetermined light/dark judgment threshold in the picked-up image before correction are dark monochromatic image ranges. An information reading device characterized by correcting the captured image so that an imaging unit;
an information code reading unit that performs a decoding process for reading the information code from the captured image in which the information code is captured by the imaging unit;
a character recognition unit that performs character recognition processing for reading the character information from an image obtained by capturing the character information by the imaging unit;
An information reader comprising
an image correction unit that corrects the captured image so that, when the information code is read by the information code reading unit, a range corresponding to the information code in the captured image becomes a monochrome image range;
When the captured image is corrected by the image correction unit, the character recognition unit performs the character recognition process on the corrected captured image ,
When the information code is read by the information code reading unit, the image correcting unit determines that the brightness difference in the captured image between the bright color cells and the dark color cells constituting the information code is equal to or greater than a predetermined brightness difference threshold. In such a case, the information reading apparatus corrects the captured image so that the range corresponding to the information code in the captured image becomes a monochromatic image range. A code determination unit that determines whether or not part of the information code is captured for the captured image for which the decoding process has failed,
When the code determination unit determines that part of the information code is imaged, the image correction unit sets a range corresponding to the part of the information code in the captured image to a monochrome image range. 3. The information reading apparatus according to claim 1, wherein the captured image is corrected as follows. The information code is a QR code in which finder patterns are arranged at three corners and two timing patterns are arranged between the finder patterns,
3. The code determination unit determines that a part of the information code is captured by capturing an image of at least one of the three finder patterns and a part of the timing pattern. The information reader according to . A detection unit that detects the cell arrangement direction of the information code,
The information according to any one of claims 1 to 4, wherein the character recognition section performs the character recognition processing on the captured image based on the cell arrangement direction detected by the detection section. reader. When the information code is read by the information code reading unit, the image correcting unit determines if the light-dark difference is less than the predetermined light-dark difference threshold value, out of the range corresponding to the information code in the captured image. , correcting the captured image so that a range having the same brightness as the light-colored cells constituting the information code and a range having the same brightness as the dark-colored cells constitute a monochrome image range. 3. The information reading device according to 2. The information code has an empty area different from the light-colored cells and the dark-colored cells in a code area in which a plurality of bright-colored cells and dark-colored cells are arranged,
When the information code having the empty area is read by the information code reading unit, the image correcting unit excludes a range corresponding to the code area in the captured image from correction targets. The information reader according to any one of claims 1 to 6. The information code has an empty area different from the light-colored cells and the dark-colored cells in a code area in which a plurality of bright-colored cells and dark-colored cells are arranged,
When the information code having the empty area is read by the information code reading unit, the image correcting unit determines that a range obtained by removing the empty area from the code area in the captured image is a monochrome image range. 7. The information reader according to any one of claims 1 to 6, wherein the captured image is corrected so that An information reader equipped with an imaging unit,
an information code reading step of performing a decoding process for reading the information code from the captured image in which the information code is captured by the imaging unit;
an image correction step of correcting the captured image so that, when the information code is read by the information code reading step, a range corresponding to the information code in the captured image becomes a monochrome image range;
a character recognition step of, when the captured image is corrected by the image correction step, performing a character recognition process for reading character information from the corrected captured image;
a light/dark reversal determining step for determining whether or not the information code read by the information code reading step is a reversed code obtained by reversing the light/darkness of a light-colored cell and a dark-colored cell;
A program that executes
In the information code reading step, the decoding process is performed by classifying the cells into light-colored cells and dark-colored cells based on a predetermined brightness determination threshold,
In the image correction step,
If the information code read by the information code reading step is determined to be the inverted code by the light/dark reversal determining step, the range corresponding to the information code in the captured image and the predetermined light/dark determination threshold value The captured image is corrected so that the range in which the brightness is low and the range in which the brightness is low is a dark monochromatic image range,
When it is determined by the light/dark reversal determination step that the information code read by the information code reading step is not the reverse code, the range corresponding to the information code in the captured image and the predetermined light/dark determination threshold The captured image is corrected so that the range in which the brightness is high and the range in which the brightness is high is a bright monochrome image range,
By determining that the information code read by the information code reading step is the reversed code by the light/darkness reversal determination step, the captured image partially corrected to the dark monochromatic image range When the performed character recognition processing fails, the range corresponding to the information code and the range in which the brightness is higher than the predetermined light/dark determination threshold in the captured image before correction become a single-color image in a bright color system. The captured image is corrected so as to be within the range,
By determining that the information code read by the information code reading step is not the reversed code by the light/darkness reversal determination step, the imaged image partially corrected to the single color image range of the bright color system. If the performed character recognition processing fails, the range corresponding to the information code and the range in which the brightness is lower than the predetermined light/dark judgment threshold in the picked-up image before correction are dark monochromatic image ranges. A program characterized in that the captured image is corrected such that An information reader equipped with an imaging unit,
an information code reading step of performing a decoding process for reading the information code from the captured image in which the information code is captured by the imaging unit;
an image correction step of correcting the captured image so that, when the information code is read by the information code reading step, a range corresponding to the information code in the captured image becomes a monochrome image range;
a character recognition step of, when the captured image is corrected by the image correction step, performing a character recognition process for reading character information from the corrected captured image;
A program that executes
In the image correcting step, when the information code is read in the information code reading step, the brightness difference in the captured image between the light-colored cells and the dark-colored cells constituting the information code is determined to be equal to or greater than a predetermined brightness difference threshold. A program , wherein the captured image is corrected such that a range corresponding to the information code in the captured image becomes a monochromatic image range when the captured image becomes a single color.

Images