JP2010217997A - Device, program, and method for recognizing character - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the level of satisfaction of a user by providing a portable terminal 1 enabling the user to easily photograph an image of a character which the user wants to recognize. <P>SOLUTION: A central control part 5 of a portable terminal 1 equipped with a photographic part 2 for obtaining a photographic image 40 by photographing and a central control part 5 for executing a character recognition program 23 for recognizing a character in the photographic image 40 is configured to execute a step S6 for detecting a static state and steps S6 to S8 for starting character recognition, wherein the detection of the static state in the step S6 is defined as one condition of character recognition start. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a character recognition device, a character recognition program, and a character recognition method that recognize characters included in image data, for example.

  Conventionally, a portable information terminal with a camera equipped with a character recognition function has been proposed (see Patent Document 1). This camera-equipped mobile information terminal can recognize characters from captured images, and can recognize English words and display English vocabulary from the dictionary function.

  Here, as a trigger for starting shooting by the camera, it is conceivable to press the shooting button.

  However, if the shooting button is pressed as a trigger, the two operations of adjusting the position of the camera-equipped portable information terminal so that the character to be recognized enters the shooting range and pressing the shooting button must be performed simultaneously. However, there is a problem that the operation is complicated.

  In addition, when the camera-operated portable information terminal is blurred due to the pressing operation of the shooting button, the shot image is also blurred, resulting in an image that is not suitable for character recognition.

JP 2005-316912 A

  SUMMARY OF THE INVENTION In view of the above-described problems, the present invention provides a character recognition device, a character recognition program, and a character recognition method that allow a user to easily capture an image of a character that the user wants to recognize, and to improve user satisfaction. And

  The present invention is a character recognition device comprising a photographing means for acquiring image data by photographing, and a character recognition means for recognizing characters in the image data, the stationary detection means for detecting a stationary state, and the character A character recognition start unit that starts character recognition by the recognition unit, and the character recognition start unit is a character recognition device that uses a detection of a stationary state by the stationary detection unit as a condition for starting character recognition. Features.

  The photographing means can be constituted by an appropriate means capable of acquiring image data, such as a CCD camera or a CMOS camera.

  The image data may be appropriate image data such as a color image, a gray scale image, or a monochrome two-tone image.

  The character recognizing means may be, for example, means for extracting feature data from a character image in image data and matching the feature data with feature data of matching data to recognize which character.

  The stillness detection means is a calculation means for detecting that the movement of the images continuously taken by the photographing means is within a predetermined range, and that the acceleration is not changed (or the change is predetermined). Acceleration sensor that detects that the camera is stationary when it is within the range, and a gyroscope that detects that the sensor is stationary when the angular velocity has not changed (or the change is within the specified range). It can be configured by an appropriate means for detection. Detection of a stationary state based on an image is detection of a relative stationary state between the photographing unit and the subject (character print medium), and detection by a physical sensor such as an acceleration sensor or a gyro sensor is detection of a physical stationary state. . In any case, the stationary state can be detected not only in a state where it is stationary on the earth, but also in a state where acceleration is not applied during movement at a constant speed by transportation or the like.

  The character recognition starting means is suitably a control means for starting character recognition if the stationary state is detected by the stationary state detecting means, or a control means for starting character recognition if the stationary state is detected and other conditions are met. The control means can be used. The other conditions may be, for example, that the subject is in focus, the character is present, the character is not tilted, the character size is appropriate, or a plurality of these conditions.

  According to the present invention, the user can easily take an image of a character to be recognized, and the satisfaction of the user can be improved.

  As an aspect of the present invention, there is provided suitability determining means for determining whether or not a character string is properly photographed in the image data, and the character recognition starting means is characterized in that a character is properly photographed by the suitability distinguishing means. This determination can be made a condition for starting character recognition.

  The suitability judging means, for example, whether the subject is in focus, the character is present, the character is not tilted, the character size is appropriate, or whether these multiple conditions are satisfied. It can be configured by a computing means for discrimination.

  According to this aspect, it is possible to start character recognition when the mobile terminal is in a stationary state and the character is properly captured. Therefore, it is possible to prevent character recognition from starting based on image data that cannot be properly recognized, and to improve user satisfaction.

  Further, as an aspect of the present invention, an input means for receiving an input of a character recognition start instruction is provided, and the character recognition start means inputs a start instruction by the input means regardless of whether or not a character recognition start condition is satisfied. Can be configured to start character recognition by the character recognition means.

  The input means is an appropriate means for accepting an input operation, such as a push button for accepting a user's push operation, a touch panel for detecting the user's touch operation, or a voice recognition means for accepting a user's instruction voice such as “start”. Can be configured.

  This aspect enables the user himself / herself to start character recognition at an arbitrary timing. Therefore, for example, the user can give a start instruction when the stationary state cannot be detected, or the user can give a start instruction when the character in the image data cannot be determined to be appropriate based on the background pattern of the character.

  Further, as an aspect of the present invention, the photographing unit is configured to continuously acquire images in a preview mode, and the stillness detecting unit is a first image in a first region determined in advance from the captured image in the preview mode. For a predetermined number of images subsequent to the preview mode, if the first image is recognized in a second area wider than the first area, it is detected that the image is stationary. it can.

  According to this aspect, it is possible to detect whether the image is still using the image data photographed by the photographing means, and the character recognition device can be simplified.

  According to another aspect of the present invention, there is provided a character recognition program for causing a computer to execute an image data acquisition step of acquiring image data by a photographing unit and a character recognition step of recognizing characters in the image data by a character recognition unit. Character recognition for causing a computer to execute a stationary detection step for detecting a stationary state by means, and a character recognition starting step for starting the character recognition step on the condition that the stationary state is detected by the stationary detection step. It can be a program.

The character recognition program can be stored in an appropriate storage medium such as a CD (Compact Disc), a DVD (Digital Versatile Disk), a flexible disk, a memory, a chip, or a hard disk.
According to the present invention, if a character recognition program is installed in a computer, high-speed character recognition can be realized with a simple operation on image data.

  According to another aspect of the present invention, there is provided a character recognition method for executing an image data acquisition step of acquiring image data by a photographing unit and a character recognition step of recognizing a character in the image data by a character recognition unit. A character recognition method for executing a stationary detection step for detecting a stationary state and a character recognition starting step for starting the character recognition step on the condition that the stationary state is detected by the stationary detection step as a condition for starting the character recognition. Can do.

  According to the present invention, the user can easily take an image of a character to be recognized, and the satisfaction of the user can be improved.

  According to the present invention, the user can easily take an image of a character to be recognized, and the satisfaction of the user can be improved.

The external perspective view of a portable terminal. The block diagram which shows the structure of a portable terminal. Explanatory drawing of the detection method of a stationary state. Explanatory drawing of a projection histogram. The flowchart of the operation | movement which the central control part of a portable terminal performs. Explanatory drawing of the screen displayed on the display part of a portable terminal.

  An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows a perspective view of the mobile terminal 1, and FIG. 2 shows a block diagram of the configuration of the mobile terminal 1.
As shown in FIG. 1, the portable terminal 1 is provided with a display unit 3 and an operation unit 4 (key operation units 4 a to 4 e) on the front surface, and an imaging unit 2 on the back surface.

  The photographing unit 2 is a digital camera that photographs an image. The display unit 3 is a liquid crystal display or an organic EL display that displays an image. The photographing unit 2 and the display unit 3 are provided back to back. Thereby, when the portable terminal 1 is positioned between the photographing target and the user, the photographing target (character) is photographed by the photographing unit 2 on the back of the portable terminal 1, and the user is displayed on the display unit 3 on the front of the portable terminal 1. The captured image can be checked.

  The operation unit 4 is a button to be pressed. Note that a touch panel may be provided on the display unit 3 as a touch operation unit, and the operation unit 4 may be configured by the touch operation unit and the key operation units 4a to 4e.

  As shown in FIG. 2, the mobile terminal 1 is provided with a central control unit 5, a RAM 6, and a ROM 7 in addition to the photographing unit 2, the display unit 3, and the operation unit 4. The photographing unit 2, the display unit 3, the operation unit 4, the RAM 6, and the ROM 7 are all connected to the central control unit 5.

  The photographing unit 2 is a digital camera such as a CCD camera or a CMOS camera. A photographed image obtained by photographing by the photographing unit 2 is transmitted to the central control unit 5 as digital image data. This digital image data may be finally stored in a separate storage device (nonvolatile memory or the like). The photographing unit 2 starts photographing upon receiving a photographing signal from the central control unit 5.

  The display unit 3 has a function as an operation display unit that displays an operation screen and a function as a recognition result display unit that displays a recognition result obtained by character recognition. When image data is sent from the central control unit 5, the display unit 3 displays an image of this image data.

  The operation unit 4 transmits to the central control unit 5 a pressing signal indicating that the user has pressed the key operation units 4a to 4e (FIG. 1).

  The central control unit 5 is a processor that performs various operations using the RAM 6 as a temporary storage area in accordance with programs stored in the RAM 6 and the ROM 7. The central control unit 5 includes a preview function unit 11, a stillness detection function unit 12, a focus determination function unit 13, a character suitability determination function unit 14, a character cutout function unit 15, a character recognition function unit 16, and a translation function unit 17. Function.

  The preview function unit 11 performs preview display in which the photographing unit 2 continuously captures still images and displays them as moving images. In this preview image, as shown in the screen explanatory diagram of FIG. 3A, a character position adjustment guide 41 and an explanation display unit 42 are displayed so as to overlap the photographed image 40. Since the character position adjustment guide 41 and the explanation display unit 42 are stopped and the captured images 40 are continuously displayed as a moving image, the character that the user wants to read can be aligned with the character position adjustment guide 41. .

  The stillness detection function unit 12 determines whether or not the subject to be photographed and the mobile terminal 1 are relatively stationary based on the still images continuously photographed for preview display. If it is determined that there is, the process for detecting stillness is executed.

  More specifically, as shown in FIG. 3A, an image of a predetermined still determination image acquisition area 51 is cut out from the captured image 40, and this is used as a reference image 51a as shown in FIG. 3B. Then, it is determined whether or not the same image as the reference image 51a is included in the shake allowable area 52 with respect to a predetermined number (for example, three) of the captured images 40 continuously captured after the still image.

  In the example of FIG. 3B, the first matching image 51b, the second matching image 51c, the third matching image 51d, and the same matching images (51b to 51d) as the reference image 51a are continuously three times. Has been confirmed within. When the third matching image 51d is obtained, it is determined that the photographing target and the mobile terminal 1 are relatively stationary, and this is determined as stationary detection.

  In this example, the same image as the reference image 51a is continuously detected a plurality of times in the shake allowable area 52. For example, the reference image is included in the shake allowable area 52 of 3 out of 5 shot images. If the same image as 51a is included, it may be determined to be in a stationary state by detection at a predetermined ratio, such as being recognized as a stationary state.

  The focus determination function unit 13 executes a process for determining whether or not a captured still image is in focus. The determination of whether or not the image is in focus is performed by performing edge extraction using the image of the still determination image acquisition area 51, and determining that the image is in focus when the edge is successfully extracted.

  The character suitability determination function unit 14 executes a process of determining whether a character exists, whether the character size is appropriate, and whether it is not tilted. More specifically, as shown in the projection histogram explanatory diagram of FIG. 4A, the image in the character position adjustment guide 41 is binarized to create a projection histogram 61 in the character string direction.

  If the character to be recognized exists in the character position adjustment guide 41 with an appropriate size without tilting, a clear peak and valley are shown in the projection histogram 61 in the character string direction as shown in FIG. Appears. In particular, a mountain appears at the center in the vertical direction, and a valley appears above and below it.

  If the characters are inclined, it can be seen that peaks and valleys do not appear in the projection histogram 61 as shown in FIG. Similarly, when the character is large enough to protrude above and below the character position adjustment guide 41, the valley disappears, and when the character is too small, the sense of the valley and the valley becomes narrow. In addition, a clear valley does not appear when there is no character.

  In this way, suitability can be determined from the projection histogram 61 for the inclination, size, and presence of the characters. Note that the determination criteria and parameters can be adjusted for each model of the mobile terminal 1.

  The character cutout function unit 15 executes a process of cutting out a character portion from the captured image. The character portion is cut out by recognizing the height of the character using the projection histogram 61 in the character string direction (left and right character width direction in the figure), and similarly projected in the direction perpendicular to the character string (upper and lower character height direction in the figure). One word may be cut out by recognizing the width of the word using the histogram.

  The character recognition function unit 16 collates the characters within the range cut out by the character cut-out function unit 15 with the character recognition dictionary unit 31, calculates the matching reliability, and recognizes the character.

  The translation function unit 17 collates a word made up of recognized characters with the translation dictionary unit 32 and translates the meaning of the word into Japanese or a foreign language.

The ROM 7 stores a camera control program 21, a character recognition start determination program 22, a character recognition program 23, and a translation program 24.
The camera control program 21 includes a preview mode for continuously acquiring low-resolution images for preview, a recognition shooting mode for acquiring high-resolution images for character recognition, and a display shooting mode for acquiring low-resolution images for confirmation display images. For example, it is a program that executes appropriate photographing control.
The character recognition start determination program 22 is a program for starting character recognition by causing the stillness detection function unit 12 to determine whether or not it has stopped.
The character recognition program 23 is a program that causes the character recognition function unit 16 to perform character recognition.
The translation program 24 is a program that causes the translation function unit 17 to perform translation.

  The RAM 6 temporarily stores data used by the central control unit 5 for various controls and calculations, and stores various data including the character recognition dictionary unit 31 and the translation dictionary unit 32.

The character recognition dictionary unit 31 is a character recognition dictionary composed of character codes and matching information. The matching information may be composed of character feature data or the like, and the character can be recognized by matching the matching information with the character image.
The translation dictionary unit 32 is an appropriate dictionary for translating words, such as a Japanese-English dictionary, an English-Japanese dictionary, a Japanese-Chinese dictionary, and a Chunichi dictionary.

  Note that the mobile terminal 1 may also be provided with a communication antenna for performing wireless communication, a speaker for outputting sound, a microphone for collecting sound, and the like. In this case, it can be used as a mobile phone.

  Further, the central control unit 5, the RAM 6 and the ROM 7 may be mounted on one chip and configured as electronic components. In this case, if this chip is provided in the portable terminal 1, a character recognition function can be easily added.

  FIG. 5 is a flowchart of operations executed by the central control unit 5 of the mobile terminal 1 according to the camera control program 21, the character recognition start determination program 22, the character recognition program 23, and the translation program 24. Here, an example of English-Japanese translation will be described.

  The central control unit 5 waits until the character recognition translation mode is started by a user operation input (step S1: No). When the character recognition translation mode is started (for example, the start button of the English-Japanese word translation function is pressed) (step S1: Yes), the central control unit 5 displays the initial screen shown in FIG. Then, the subsequent processing is executed until the processing is completed (step S2: No).

On this initial screen, a character position adjustment guide 41 for guiding the position of the character that the operator wants to recognize is overlaid on the photographed image 40 so as to be overlaid. The character position adjustment guide 41 is displayed in order to position the character in the guide, and the explanation display unit 42 guides the character to be read on the left side of the guide. With this display, the user can move the mobile terminal 1 and match the character to be translated to the character position adjustment guide 41.
When the user performs an end operation (step S2: Yes), the central control unit 5 ends the character recognition translation mode.

  When the user presses the photographing button assigned to one of the key operation units 4a to 4e (step S3: Yes), the central control unit 5 skips the process and proceeds to step S11. It should be noted that the skip to step S11 may be executed regardless of the timing at which the shooting button is pressed after the start of the preview mode up to step S11.

  If the photographing button is not pressed by the user (step S3: No), the central control unit 5 starts the preview mode by the preview function unit 11 (step S4). In this preview mode, it is preferable to perform necessary adjustments such as focus adjustment, white balance adjustment and exposure adjustment of the photographing unit 2, or to perform necessary adjustments in real time while displaying a preview. As a result, the subsequent character recognition photographing (step S9) can be quickly executed at an appropriate timing.

  The central control unit 5 captures and stores a low-resolution still image for preview every Nms (step S5). The central control unit 5 displays the still image acquired at this time on the display unit 3 in real time as shown in FIG. As a result, the user can see the captured image 40 displayed on the display unit 3 and align the head of the character to be read with the left end in the character position adjustment guide 41.

  In addition, this screen also displays a still determination in progress display 43 and functions as a still determination in progress screen. This still-in-determination display 43 is displayed so that the bar extends every time it is determined that the same image as the reference image 51a is contained in the shake allowable area 52 (see FIG. 3). If it is determined that the same image as 51a is contained, the bar extends to the end, indicating that the stationary state has been recognized, and informing the user that the next processing is to be performed. If it is determined that the same image as the reference image 51a is not included in the shake allowable area 52 (see FIG. 3), the bar is returned to zero and the process is started again.

The central control unit 5 determines whether or not the stationary state is detected by the stillness detection function unit 12, and returns the process to step S2 until the stationary state is detected (step S6: No).
When the stationary state is detected (step S6: Yes), the central control unit 5 determines whether or not the focus is determined by the focus determination function unit 13, and if not (step S7: No), the process proceeds to step S2. To return.

  If the image is in focus (step S7: Yes), the central control unit 5 determines whether or not the character is appropriate by the character suitability determination function unit 14, and if the character is not appropriate (step S8: No), step S2 Return processing to. The determination of whether or not the character is appropriate is made by determining whether or not the character exists, whether or not the character size is appropriate, and whether or not the character inclination is appropriate. To do.

  If the character is appropriate (step S8: Yes), the central control unit 5 captures a character recognition rectangular image by the photographing unit 2 (step S9). In photographing at this time, an image having a higher resolution than that in steps S5 and S10 is acquired for use in character recognition. Further, in the photographing at this time, only the rectangular image in the character position adjustment guide 41 is acquired. Thereby, only the image in the character position adjustment guide 41 used for character recognition can be acquired and processed efficiently.

  The central control unit 5 further captures a character string cutting result display image (step S10). Since the character string cutout result display image is for display on the display unit 3 for the result confirmation, an image having a resolution lower than that in step S9 (the same resolution as the photographed image for preview in step S5) is acquired. To do. The character string cutout result display image may also be acquired only in the character position adjustment guide 41.

  The central control unit 5 displays the character recognition in-progress display 44 on the display unit 3 as shown in FIG. 6C (step S11). At the same time, sound output such as “shooting” sound may be performed. Thus, the user can be notified that photographing has been completed and character recognition has started. When the user sees this screen or listens to the voice, the user can finish the position adjustment operation so that the character to be read is held in the character position adjustment guide 41 with the portable terminal 1, and relax and wait for the recognition completion. Can do.

  The central control unit 5 performs character segmentation by the character segmentation function unit 15 and character recognition by the character recognition function unit 16 for the segmented character (step S12).

The central control unit 5 performs a translation process by the translation function unit 17 on the recognized character (step S13).
When the translation is completed, the central control unit 5 displays the result on the display unit 3 as shown in FIG. 6D (step S14). At this time, the photographed image 40 photographed in step S 10 is displayed on the display unit 3, and the extracted word portion G of the photographed image 40 is surrounded by the recognition target frame 45 and displayed. In addition, a recognized character display 47 and a translation display 48 are displayed below the captured image 40. As a result, it is possible to display how the character is recognized and to display the translation of the word composed of the recognized character.

  The central control unit 5 continues to display the result until a predetermined time elapses (No in step S15), and when the predetermined time elapses, the process returns to step S2 and is repeated (step S15: Yes).

  With the above configuration and operation, character recognition can be started by detecting the stationary state of the mobile terminal 1. Thus, the user can perform character recognition and translation only by aligning the position of the portable terminal 1 with the character to be read without performing a complicated operation such as pressing the shooting button while aligning the position of the portable terminal 1 with the character to be read. It can be performed. In particular, it can be used very conveniently for a user such as an elderly person who has difficulty in hand operation.

  More specifically, when a user wants to read a word, the user tries to fit the word within the frame of the character position adjustment guide 41 on the preview screen. As soon as you want to start the character recognition well within the frame, the camera is still stationary and the focus and character position are appropriate. The mobile terminal 1 described above automatically detects the state, and can automatically start reading when the user wants to read. For this reason, usability is very good, and the user can use it conveniently.

  In addition, since shooting can be started without pressing the shooting button, it is possible to prevent the mobile terminal 1 from blurring due to the pressing of the shooting button, thereby causing blurring of the acquired still image and lowering of character recognition accuracy.

  Moreover, since character recognition is not started until the stationary state of the portable terminal 1 is detected, it is possible to prevent characters from being recognized by photographing an image or character that is not intended by the user simply by focusing for a moment. As a result, it is possible to prevent unintentionally starting character recognition and having to wait unnecessarily until the character recognition is completed, thereby greatly reducing unnecessary character recognition processing.

  In addition, since the stationary state of the mobile terminal 1 is determined by a captured image captured by the imaging unit 2, it is not necessary to provide a separate device such as an acceleration sensor or a gyro sensor, and the mobile terminal 1 can be simplified. Further, since the stationary state is determined based on the captured image, the character recognition can be started if the character print medium as the subject and the mobile terminal 1 are relatively stationary even when the user is moving.

  In addition, since character recognition is performed after detecting that a character is present in step S8, that the character is in an appropriate direction, and that the character is in an appropriate size, the state where the character does not exist is inappropriate. It is possible to prevent a user from waiting in vain by starting character recognition in a state of orientation or an inappropriate size.

  Also, since the character recognition can be forcibly started with the shooting button, even if there is a pattern on the background of the character and the character suitability cannot be determined as appropriate in step S8, the operator presses the shooting button and moves to any camera position. Character recognition can be started.

In addition, since the character recognition can be forcibly started with the shooting button, the character recognition can be started even when the user is unfamiliar with the operation and cannot focus well or when the stationary state cannot be detected.
Even if the camera lens is not in good focus due to camera lens limitations, character recognition can be forcibly started with the shooting button.

  In addition, the determination of the start of character recognition (stillness determination, focus determination, character presence determination, character orientation determination, character size determination, or a plurality of these) is not the entire preview image, but a portion of the still determination image Since only the acquisition area 51 is used, the determination can be made at high speed and with high accuracy. By making the still determination image acquisition area 51 the minimum necessary area for determination, the determination accuracy and speed can be improved to the maximum. Accordingly, a word reading function and a translation function can be realized without stress even on an inexpensive terminal having a relatively low processing speed.

  In addition, since the stillness determination is performed in the shake allowable area 52 wider than the stillness determination image acquisition area 51, it is determined that the camera is stationary even if there is some blurring (the user is trying to stop). Can do. Further, since the shake allowable area 52 is a partial area of the captured image 40, the calculation process for stillness determination can be lightly processed at high speed.

  The captured image for character recognition acquired in step S9 is an image having a higher resolution than the preview image in step S5, and is a captured or read image of only the rectangular image in the character position adjustment guide 41. Character recognition can be performed accurately and at high speed.

  Compared to the method in which characters cannot be recognized unless the shooting button is pressed, the user can adjust the position many times and press the shooting button to perform focus adjustment, shooting, and character recognition when character recognition is not successful. There is no need to repeat, it can be used conveniently. That is, when the focus has already been adjusted in the preview mode and character recognition is performed again, if the position is adjusted, shooting and character recognition can be started and repeated in a short time without stress.

  In addition, if a method of automatically acquiring a captured image and recognizing characters without performing stillness determination or the like, the characters will be captured and recognized at a position not intended by the user, which is inconvenient. That is, when character recognition is started by photographing at an unintended position, the user must wait until character recognition is completed. Meanwhile, since the portable terminal is performing the character recognition calculation process, the preview display is stopped and the user cannot position the character to be read. Then, alignment is started again after the character recognition process is completed. If shooting and character recognition are started in the middle of this process, the user must wait again, which is very inconvenient.

  On the other hand, the above-described portable terminal 1 can improve convenience. That is, when the user wants to recognize or translate characters, the user moves the mobile terminal 1 and performs an operation for adjusting the characters to be translated within the photographing range of the photographing unit 2. At that time, it is a normal operation that the user tries to stop the portable terminal 1 in the aligned state. In the above embodiment, since this operation is detected (static state is detected) and character recognition is started, it is possible to prevent character recognition by photographing at a position not intended by the user, and it is excellent in operability and easy to use without stress. A character recognition function and a translation function can be provided.

  Note that a mode in which only the character inclination determination is not executed in step S8 may be added, and the operator may select whether or not to execute the character inclination determination. This makes it possible to show particularly good results with text on paper when character inclination is determined, and when character inclination is poor, such as when there is a pattern in the background, when character inclination is not determined. In addition, character recognition and translation can be performed after the stationary state is detected.

  Further, an algorithm that is resistant to blurring may be adopted for the character recognition program 23 so that the focus determination in step S7 is not performed. Even in this case, the characters can be recognized properly.

  The character recognition program 23 and the translation program 24 are stored in storage means of a server connected to a telecommunication line such as the Internet, and can be downloaded through the telecommunication line, or ASP (Application Service) through the telecommunication line. It is also possible to adopt a configuration that enables use as a provider service. In this case, the calculation processing load of the mobile terminal 1 can be reduced.

  In addition, the image may not be taken in step S10, and the preview image acquired in step S5 may be used as the character cutout result display table image.

  The stationary state can be detected by an acceleration sensor, a gyro sensor, or the like. In this case as well, it is possible to realize that character recognition starts when it is stationary.

In correspondence between the configuration of the present invention and the above-described embodiment,
The character recognition device and computer of this invention correspond to the portable terminal 1 of the embodiment,
Similarly,
The photographing means corresponds to the photographing unit 2,
The input means corresponds to the shooting button assigned to one of the key operation units 4a to 4e,
The stationary detection means corresponds to the central control unit 5 that executes step S6,
The character recognition starting means corresponds to the central control unit 5 that executes steps S6 to S8,
The character recognition means corresponds to the central control unit 5 that executes step S12,
The suitability determination means corresponds to the central control unit 5 that executes steps S7 to S8,
The character recognition program corresponds to the camera control program 21, the character recognition start determination program 22, and the character recognition program 23.
The image data corresponds to the captured image 40,
The first area corresponds to the still determination image acquisition area 51,
The first image corresponds to the reference image 51a,
The second area corresponds to the shake allowable area 52,
The image data acquisition step corresponds to step S5,
The stationary detection step corresponds to step S6,
The character recognition start step corresponds to steps S6 to S8,
The character recognition step corresponds to step S12,
Input of the start instruction corresponds to pressing of the shooting button,
The present invention is not limited only to the configuration of the above-described embodiment, and many embodiments can be obtained.

  The present invention can be used for a device for recognizing characters, and in particular, can be used for a portable terminal that performs character recognition when a user captures a character to be recognized by holding the device.

DESCRIPTION OF SYMBOLS 1 ... Portable terminal, 2 ... Imaging | photography part, 4a-4e ... Key operation part, 5 ... Central control part, 21 ... Camera control program, 22 ... Character recognition start determination program, 23 ... Character recognition program, 40 ... Captured image, 51 Image acquisition area for stillness determination 51a Reference image 52 Allowable blur area

Claims (6)

Photographing means for acquiring image data by photographing;
A character recognition device comprising character recognition means for recognizing characters in the image data,
A stationary detection means for detecting a stationary state;
Character recognition start means for starting character recognition by the character recognition means,
The character recognition starting means
A character recognition device that makes one condition of character recognition start that the stationary state is detected by the stationary detection means. Suitability determination means for determining whether or not a character string is properly captured in the image data,
The character recognition start means
The character recognition device according to claim 1, wherein a condition for starting character recognition is that the character determination unit determines that the character is properly photographed. An input means for receiving an input of a character recognition start instruction;
The character recognition start means
3. The character recognition device according to claim 1, wherein the character recognition unit starts character recognition when an input of a start instruction is received by the input unit regardless of whether a character recognition start condition is satisfied. The photographing means is configured to continuously acquire images in a preview mode,
The stationary detection means includes
Obtaining a first image of a predetermined first area from the captured image in the preview mode;
2. A configuration in which a predetermined number of images subsequent to the preview mode are detected as being stationary when the first image is recognized in a second area wider than the first area. Or the character recognition apparatus of 3 description. A character recognition program that causes a computer to execute an image data acquisition step of acquiring image data by a photographing unit and a character recognition step of recognizing characters in the image data by a character recognition unit,
A stationary detection step for detecting a stationary state by a stationary detection means;
A character recognition program for causing a computer to execute a character recognition start step for starting the character recognition step on the condition that a stationary state is detected by the stillness detection step as a condition for starting character recognition. A character recognition method for executing an image data acquisition step of acquiring image data by a photographing means and a character recognition step of recognizing characters in the image data by a character recognition means,
A stationary detection step for detecting a stationary state by a stationary detection means;
A character recognition method for executing a character recognition start step for starting the character recognition step on the condition that the stationary state is detected by the stillness detection step as a condition for starting character recognition.

Images