JP2021022155A - Digital display reader and program - Google Patents

Abstract

To provide a digital display reader that can derive the degree of certainty of the contents read from a digital display.SOLUTION: A digital display reader according to the present invention has a plurality of segments and reads numbers from an image of a digital display device that shows numbers by combinations of the presence or absence of lighting of each segment. The digital display reader includes: reading means for reading numbers from the image; and certainty degree deriving means for deriving the degree of certainty of the contents read by the reading means on the basis of the lighting state of the segment forming numbers known from the image.SELECTED DRAWING: Figure 1

Description

The present invention relates to a digital display reader and a program for reading the display contents from a captured image of a digital display.

There is a device that photographs a digital display in which each number is represented by seven segments and reads a display showing a measurement result or the like based on the analysis of the captured image (see, for example, Patent Document 1).

Japanese Patent Application No. 2018-197201

When reading numbers expressed by multiple segments from the captured image of the digital display, the light reflected on the glass surface that protects the digital display, the image of surrounding objects, or the shadow of the edge of the digital display, etc. It may be mixed in the captured image and the display may be misread.

However, with the conventional reading device, only the reading result by image analysis can be obtained, and in order to detect the reading error, it is necessary to visually collate the actual display with the reading result by the reading device.

An object of the present invention is to provide a digital display reading device capable of deriving the accuracy of the content read from the digital display. As a result, the credibility of the content read from the digital display can be easily determined.

(1) The digital display / reading device of the present invention is a digital display / reading device that includes a plurality of segments and reads a numerical value from a captured image of a digital display in which a numerical value is expressed by a combination of lighting / non-lighting of each segment. The reading means for reading the numerical value from the captured image and the accuracy derivation means for deriving the accuracy of the content read by the reading means based on the lighting state of the segment constituting the numerical value grasped from the captured image are provided.

(2) The accuracy derivation means determines the accuracy of each digit of each digit constituting the numerical value for each of a plurality of segments expressing the numerical value, and determines the relationship between the lighting area and the non-lighting area grasped from the captured image. It may be derived based on the lighting ratio based on.

(3) For example, the accuracy of the number may be derived as the lowest lighting ratio among the lighting ratios of each of the plurality of segments expressing the number.

(4) The accuracy derivation means sets the lighting ratio of each of the plurality of segments expressing the numbers on the display portion of the captured image of the numbers read by the reading means, and the plurality of segments expressing the numbers on the digital display. It may be obtained based on the relationship between the lit area and the non-lit area in the contour of each segment, which is grasped by superimposing the arrangements.

(5) The accuracy derivation means may derive the accuracy of a numerical value based on the accuracy of each numerical value constituting the numerical value.

(6) For example, the accuracy of the numerical value may be derived as the lowest accuracy among the accuracy of each numerical value constituting the numerical value.

According to the digital display reading device of the present invention configured as described above, the accuracy of the content read from the digital display is derived, so that the user trusts the content such as numerical values and numbers read based on this accuracy. It is possible to decide whether to adopt it as it is or to read it again.

(7) When the accuracy derivation means determines that the accuracy of any of the numbers constituting the numerical value is lower than a predetermined threshold value, the reading means another photographing of the digital display on which the numerical value is displayed. At least, the number determined to be lower than the predetermined threshold value may be read again from the image.

(8) Further, when the accuracy deriving means determines that the accuracy of the numerical value is lower than a predetermined threshold value, the reading means rereads the numerical value from another captured image of the digital display on which the numerical value is displayed. You may do so.

(9) Further, the accuracy deriving means may further include a notification unit for notifying the user of predetermined information prompting input to another captured image reading means when the accuracy is determined to be lower than a predetermined threshold value. Good.

By adopting the configurations shown in (7) to (9) for re-reading another captured image based on the accuracy derived by the accuracy derivation means 120, the derivation of the accuracy and the re-reading process can be linked. , The reread result can be obtained quickly.

(10) The reading means reads a numerical value from each of a plurality of captured images in which the same numerical value is captured, and the accuracy derivation means derives the accuracy for the numerical value of each captured image and identifies the numerical value having the highest accuracy. It may be.

In this way, by configuring to specify the numerical value with the highest accuracy among the numerical values read from the plurality of captured images, a highly reliable reading result can be obtained.

(11) The numerical value consists of a number and a decimal point, and the decimal point is expressed by the presence or absence of lighting of the segment for the decimal point of each digit provided at a predetermined position with respect to the display position of the number of each digit on the digital display. In this case, when the reading means reads the numbers of each digit from the captured image, the rectangular area corresponding to the display position of the numbers is specified in the captured image, and the decimal point area is set at a predetermined position with respect to the rectangular area. The lighting state of the segment for the decimal point is read by specifying, and the accuracy derivation means derives the accuracy based on the lighting state for the segment for the decimal point of each digit, and the decimal point is at the position of the segment for the decimal point with the highest accuracy. May be specified as being displayed.

When reading a numerical value including a decimal point from a captured image, it is also necessary to detect the position of the decimal point. The position of the decimal point has a great effect on the final reading value, but the decimal point is difficult to read due to its small size, and if the image is easily filtered, it may disappear or appear.
In the present invention, the position of the segment for the decimal point of each digit is specifically specified, and the digit where the decimal point is displayed is specified based on the lighting state of each segment for the decimal point, so that the position of the decimal point is detected more accurately. can do.

(12) A target closed contour contained in an original image including a digital display as a subject and whose aspect ratio is substantially equal to the aspect ratio of the digital display is detected, and the area inside the target closed contour is used as a reading means as a captured image. A closed contour detecting means for input may be further provided.

(13) When the closed contour detecting means detects a plurality of closed contours having an aspect ratio substantially equal to the aspect ratio of the digital display and having an inclusion relationship, the innermost closed contour may be regarded as the target closed contour.

When shooting a digital display, various objects around the digital display are usually included in the shooting range, so the digital display that is the target of reading numerical values from the original image obtained by shooting is shot. The image needs to be extracted. By adopting the configurations shown in (12) and (13), among various closed contours included in the original image, a closed contour having an aspect ratio substantially equal to that of the digital display on which the numerical value to be read is displayed is searched. By this simple process, it is possible to quickly extract the captured image to be read.

(14) The original image covers a predetermined range including the digital display, centered on the line-of-sight position of the photographer at the time of shooting, and the photographing unit that captures the extraction source image including the digital display as the subject. The original image extraction means for extracting the image may be further provided.

(15) The line-of-sight detection unit for detecting the line-of-sight direction of the photographer may be further provided, and the original image extraction means may specify the line-of-sight position based on the line-of-sight direction of the photographer at the time of shooting detected by the line-of-sight detection unit.

(16) Further, the original image extraction means may specify the center of the extraction source image as the line-of-sight position.

By adopting the configurations shown in (14) to (16), the wearer can see through a certain range of the field of view of the wearer and use a head-mounted display device provided with a photographing unit capable of photographing the field of view. The original image to be input to the closed contour detection means can be extracted from the extraction source image taken by looking at the digital display to be read.

The digital display / reading device of the present invention may be realized by describing the function of each means in a program and causing a computer to execute the function.

It is a figure which shows the structural example of the digital display reading apparatus 100 of this invention. It is a figure explaining the process which superimposes the arrangement of a segment on a photographed image. It is a figure explaining the reading method of the decimal point. It is a figure which shows the structural example of the digital display reading apparatus 200 of this invention. It is a figure explaining the significance of the closed contour detecting means 210. It is a figure which shows the structural example of the digital display reading apparatus 300 of this invention. It is a figure which shows the configuration example for realizing the function of each means by executing the program which described the function of each means of the digital display reading apparatus of this invention in a CPU.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following description and drawings, the same functional parts are designated by the same reference numerals, and the functional parts once described will be omitted or described to the extent necessary.

<First Embodiment>
FIG. 1 shows a configuration example of the digital display reading device 100 of the present invention. The digital display reader 100 includes a plurality of segments, for example, seven segments per digit, and a photographed image of a digital display in which a numerical value including at least one digit number is expressed by a combination of lighting and non-lighting of each segment. It is a digital display reading device that reads a numerical value from the above, and includes a reading means 110 and an accuracy derivation means 120.

The reading means 110 receives an input of a captured image of the digital display, and reads a numerical value from the captured image by image analysis. A known method may be adopted as the method for reading the numerical value.

The accuracy derivation means 120 derives the accuracy of the content read by the reading means 110 based on the lighting state of the segment constituting the numerical value grasped from the captured image. The accuracy may be derived for each digit constituting the numerical value, or may be derived as a whole numerical value.

When deriving the accuracy for each digit, for example, for each of the plurality of segments expressing the numbers, the lighting ratio based on the relationship between the lighting area and the non-lighting area grasped from the captured image is obtained, and based on this. It may be derived. The specific derivation method is arbitrary, and for example, it may be derived as the lowest lighting ratio among the lighting ratios obtained for each of the plurality of segments, or may be derived as an average value.

The lighting ratio of each segment is grasped, for example, by superimposing the arrangement of a plurality of segments expressing the number on the digital display on the display portion of the captured image of the number read by the reading means 110. It may be obtained based on the relationship between the lit area and the non-lit area in the contour of each segment.

Since the arrangement and shape of the segment expressing each number is known depending on the model of the digital display, for example, the accuracy derivation means 120 stores the arrangement and shape in the model in advance or stores it in any storage means. The stored items may be read out, the arrangement and shape of each of a plurality of models may be stored in advance, or the items stored in any storage means may be read out, and the operator may select and input the models. You may read it.

The processing in the accuracy derivation means 120 will be specifically described with reference to FIG. When a certain digit of the 7-segment digital display is photographed, for example, when the photographed image shown in FIG. 2A is obtained and the reading means 110 reads this as "5", the accuracy derivation means 120 , The arrangement of the contours of the five segments S1 to S5 of FIG. 2B showing "5" on the 7-segment digital display is superimposed on the captured image as shown in FIG. 2C.

The superposition may be performed so that, for example, the lighting area ratio to the total area in the contour of all five segments is the largest, or the lighting area ratio to the total area in the contour of some segments is the largest. You may go.

In addition, lighting and non-lighting here mean a relative lighting state, the background color of the digital display corresponds to non-lighting, and the color of the segment constituting the number when the number is expressed is Corresponds to lighting. FIG. 2 illustrates a case where the non-lit part is represented by black and the lit portion is represented by white. For example, when the background is white and the numbers are represented by black, the non-lit portion is white and lit. The part becomes black.

After superimposing as shown in FIG. 2 (c), for each of the segments S1 to S5, based on the relationship between the area of the lit portion represented by white and the area of the non-lit portion represented by black in the contour. Find the lighting ratio. The lighting ratio may be defined as, for example, the ratio of the area of the lit portion to the total area in the contour, or may be defined as the ratio of the area of the non-lit portion to the area of the lit portion. The threshold value of the color indicating the lit portion and the color indicating the non-lit portion may be arbitrarily set according to the frequency of rereading and the tolerance for the occurrence of reading errors.

Assuming that the lighting ratios of the segments S1 to S5 are 90%, 100%, 81%, 100%, and 75%, respectively, the accuracy of the digit "5" is derived as the lighting ratio of the segment having the lowest lighting ratio. In that case, the accuracy is 75%.

When determining the necessity of rereading based on the accuracy of each digit, the accuracy derived in this way for each digit is arbitrarily determined according to the frequency of rereading and the tolerance for the occurrence of reading errors. It can be judged whether or not it is below the threshold value.

On the other hand, when determining the necessity of rereading by the accuracy of the entire numerical value consisting of a plurality of digits, the accuracy needs to be derived from the numerical value. The specific method of deriving the accuracy of the numerical value is arbitrary. For example, the accuracy of each digit may be derived as the lowest accuracy or the average value.

It is a numerical value consisting of four digits, and if the accuracy of each digit is 95%, 81%, 85%, 100%, when deriving as the accuracy of the lowest accuracy, it is relevant. The accuracy of the numbers is 81%.

When determining the necessity of re-reading based on the numerical accuracy, whether or not the accuracy derived in this way is below a threshold value arbitrarily determined according to the frequency of re-reading and the tolerance for the occurrence of reading errors. Can be judged by.

According to the digital display reading device 100 of the present invention configured as described above, the accuracy of the content read from the digital display is derived, so that the user can trust the content such as numerical values and numbers read based on this accuracy. It is possible to easily determine whether to adopt it as it is or to read it again.

[Modification 1]
When the accuracy deriving means 120 determines that the accuracy of any of the numbers constituting the numerical value is lower than a predetermined threshold value, the reading means 110 determines that the numerical value is displayed on another digital display. At least, the number determined to be lower than the predetermined threshold value may be reread from the captured image.

Further, when it is determined that the accuracy of the numerical value is lower than the predetermined threshold value, the reading means 110 may read the numerical value again from another captured image of the digital display on which the numerical value is displayed.

Further, when the accuracy deriving means 120 determines that the accuracy is lower than the predetermined threshold value, the user is notified of predetermined information such as voice, characters, and images prompting the input of another captured image to the reading means 110. A notification unit 121 may be further provided.

In this way, by adopting a configuration in which another captured image is read again based on the accuracy derived in the accuracy derivation means 120, the derivation of the accuracy and the rereading process can be linked, so that the reread result can be read quickly. Can be obtained.

[Modification 2]
The reading means 110 reads a numerical value from each of a plurality of captured images in which the same numerical value is captured, and the accuracy deriving means 120 derives the accuracy for the numerical value of each captured image and identifies the numerical value having the highest accuracy. You may.

In this way, by configuring to specify the numerical value with the highest accuracy among the numerical values read from the plurality of captured images, a highly reliable reading result can be obtained.

[Modification 3]
A numerical value with a decimal point is generally displayed on the digital display, and in this case, it is necessary to detect the position of the decimal point from the captured image of the digital display. The position of the decimal point varies depending on the number of effective digits and the number of integer copies. Therefore, the position of the decimal point has a great influence on the final reading value, but the segment for the decimal point that expresses the decimal point is small in size, so it is difficult to read the lighting state, and if you easily filter the image, it will disappear or appear. There is a risk.

Therefore, the lighting accuracy of the segment for the decimal point of each digit provided at a predetermined position with respect to the display position of the number of each digit may be obtained, and the display position of the decimal point may be specified by comparing the lighting accuracy of each.

Specifically, first, when the reading means 110 reads each digit number from the captured image, the rectangular area corresponding to the display position of the number is specified in the captured image, and at a predetermined position with respect to the rectangular area. After specifying the decimal point area, read the lighting state of the segment for the decimal point.

Since the predetermined position of the decimal point region with respect to the rectangular region is known depending on the model of the digital display, for example, the reading means 110 stores the position in the model in advance or reads out what is stored in any storage means. Alternatively, the positions of each of the plurality of models may be stored in advance, or those stored in any storage means may be read out, and the positions may be read out by the model selection input by the operator.

Subsequently, the accuracy derivation means 120 derives the accuracy of each digit segment for the decimal point based on the lighting state, and identifies that the decimal point is displayed at the position of the segment for the decimal point with the highest accuracy. The method of deriving the accuracy based on the lighting state is arbitrary. For example, the accuracy may be derived in a form of giving a high accuracy in proportion to the lighting intensity, or may be derived as the ratio of the lighting area to the area of the decimal point region. ..

The reading of the decimal point will be specifically described with reference to FIG. When the captured image P is, for example, an image of a numerical value including a decimal point composed of four-digit numbers as shown in FIG. 3A, the reading means 110 first displays the captured image P of the digital display. The rectangular regions RA1 to RA4 corresponding to the display position of the number of each digit are specified. Subsequently, the decimal point regions NA1 to NA4 are specified at predetermined positions with respect to the rectangular regions RA1 to RA4, and the lighting state of the segment for the decimal point existing in each decimal point region is read.

Then, for example, the lighting intensity of each segment for the decimal point is compared, and it is specified that the decimal point is displayed at the position of the segment having the highest intensity.

According to the configuration of the modification 3 described above, in order to specifically specify the position of the segment for the decimal point of each digit and to specify the digit in which the decimal point is displayed based on the lighting state of each segment for the decimal point. The position of the decimal point can be detected more accurately.

<Second Embodiment>
FIG. 4 shows a configuration example of the digital display reader 200 of the present invention. The digital display reading device 200 includes a configuration in which the closed contour detecting means 210 is added to the digital display reading device 100 shown in FIG.

The closed contour detecting means 210 receives an original image including a digital display as a subject, detects a target closed contour included in the original image and having an aspect ratio substantially equal to that of the digital display, and inside the target closed contour. The area of is input to the reading means 110 as a captured image.

When shooting a digital display, various objects around the digital display are usually included in the shooting range, so the digital display that is the target of reading numerical values from the original image obtained by shooting is shot. The image needs to be extracted.

By providing the closed contour detecting means 210, among various closed contours included in the original image, a closed contour having an aspect ratio substantially equal to that of the digital display on which the numerical value to be read is displayed is searched by a simple process. It is possible to quickly extract the captured image to be read.

The closed contour detecting means 210 has an aspect ratio substantially equal to that of a digital display displaying a numerical value to be read, and when a plurality of closed contours having an inclusion relationship are detected, the innermost closed contour is targeted. The captured image may be extracted as if it were. For example, when closed contours C1 and C2 having an aspect ratio substantially equal to that of the digital display are detected in the original image as shown in FIG. 5, the inner closed contour C2 is regarded as the target closed contour.

As a result, the aspect ratio is substantially equal to that of the digital display on which the numerical value to be read is displayed, and even when a plurality of closed contours having an inclusion relationship are detected, the target closed contour can be quickly identified.

[Modification example]
An original image input to the closed contour detecting means 210 can be viewed, for example, by using a transmissive head-mounted display device including a photographing unit 220 capable of seeing through a certain range of the wearer's field of view and capturing the field of view. It may be extracted from the extraction source image taken by the wearer looking at the digital display to be read.

To extract the original image from the extraction source image captured by the photographing unit 220, for example, the original image extracting means 230 for extracting a predetermined range including the digital display centered on the line-of-sight position of the photographer at the time of photographing. It may be provided.

The method of specifying the line-of-sight position is arbitrary. For example, the line-of-sight detection unit 240 for detecting the line-of-sight direction of the photographer may be further provided, and the original image extracting means 230 may specify the image based on the line-of-sight direction of the photographer at the time of shooting detected by the line-of-sight detection unit 240. Further, the center of the extraction source image may be specified as the line-of-sight position without using the line-of-sight detection unit 240.

With the above configuration, the wearer can read a target by using a transmissive head-mounted display device including a photographing unit 220 capable of seeing through a certain range of the wearer's field of view and capturing the view. The original image to be input to the closed contour detecting means 210 can be extracted from the extraction source image taken by looking at the digital display of the above.

[Application example]
A method of quickly extracting the captured image of the digital display from the original image by a simple process of searching for a closed contour with an aspect ratio that is almost equal to that of the digital display that displays the numerical value to be read is shooting. It is also effective in the conventional configuration in which the accuracy is not derived only by reading the numerical value from the image.

FIG. 6 shows a configuration example of the digital display / reading device 300 excluding the portion related to the derivation of accuracy from the configuration of the second embodiment.

According to the digital display reading device 300 including the reading means 110 and the closed contour detecting means 210, the aspect ratio of the closed contour detecting means 210 is substantially equal to that of the digital display on which the numerical value to be read is displayed from the original image. By a simple process of searching for a closed contour, the captured image of the digital display can be quickly extracted and input to the reading means 110.

In addition, an extraction taken by the wearer looking at the digital display to be read by using a head-mounted display device including a photographing unit 220 capable of seeing through a certain range of the wearer's field of view and photographing the field of view. The original image to be input to the closed contour detecting means 210 can be extracted from the original image by providing the original image extracting means 230 (and the line-of-sight detecting unit 240).

<Third Embodiment>
The functions of the respective means may be realized by describing the functions of the means constituting the digital display readers 100 and 200 in a program and causing the computer to execute the functions. A configuration example is shown in FIG.

The digital display readers 100, 200, and 300 include, for example, a CPU 101, a storage unit 102, an input unit 103, a display unit 104, a communication unit 105, and a photographing unit 220.

The CPU 101 executes a program in which the functions of the means stored in the storage unit 102 are described, and realizes the functions of the means. The storage unit 102 is an arbitrary storage unit that stores information used for processing in each means, such as a program in which the functions of each means are described, captured images, segment arrangement / shape information, and position information of a decimal point area with respect to a rectangular area. It is a means. For the storage unit 102, for example, in addition to a storage medium such as an HDD or a flash memory, a non-volatile memory, a volatile memory, or the like can be adopted. The captured image and appearance data are read out by executing a program in which the functions of each means are described. The storage unit 102 may be realized by adopting cloud storage connected via the communication unit 105 instead of being provided in the devices of the digital display / reading devices 100, 200, and 300. The input unit 103 is an arbitrary input interface when a captured image or the like is captured from the outside of the device. The display unit 104 displays various information under the control of the CPU 101. The communication unit 105 is an interface for connecting to a wireless network or a wired network, and sends and receives information to and from a cloud storage connected to the network under the control of the CPU 101. The photographing unit 220 photographs a digital display or the like and stores it in the storage unit 102.

The present invention is not limited to the above embodiments and modifications. The above-described embodiment and modification are examples, and any one having substantially the same configuration as the technical idea described in the claims of the present invention and exhibiting the same effect and effect. However, it is included in the technical scope of the present invention. That is, it can be appropriately changed within the scope of the technical idea expressed in the present invention, and the form in which such a change or improvement is added is also included in the technical scope of the present invention.

100, 200, 300 ... Digital display reader 101 ... CPU
102 ... Storage unit 103 ... Input unit 104 ... Display unit 105 ... Communication unit 110 ... Reading means 120 ... Accuracy deriving means 121 ... Notification unit 210 ... Closed contour detecting means 220 ... Imaging unit 230 ... Original image extracting means 240 ... Line of sight detecting unit NA1 to NA4 ... Decimal point area RA1 to RA4 ... Rectangular area P ... Captured image

Claims (17)

A digital display reader that has a plurality of segments and reads the numerical value from a captured image of a digital display in which the numerical value is expressed by a combination of lighting and non-lighting of each of the segments.
A reading means for reading the numerical value from the captured image,
An accuracy deriving means for deriving the accuracy of the content read by the reading means based on the lighting state of the segment constituting the numerical value, which is grasped from the captured image.
A digital display reader equipped with. The accuracy derivation means determines the accuracy of each digit of each digit constituting the numerical value for each of the plurality of segments expressing the numerical value, and the lighting area and the non-lighting area grasped from the captured image. The digital display / reading device according to claim 1, wherein the digital display / reading device is derived based on a lighting ratio based on a relationship. The digital display reading according to claim 2, wherein the accuracy derivation means derives the accuracy of the number as the lowest lighting ratio among the lighting ratios of the plurality of segments expressing the numbers. apparatus. The accuracy derivation means expresses the lighting ratio of each of the plurality of segments expressing the numbers on the display portion of the numbers read by the reading means in the captured image and the numbers on the digital display. The digital according to claim 2 or 3, which is obtained based on the relationship between the lighting area and the non-lighting area in the contour of each of the segments, which is grasped by superimposing the arrangements of the plurality of the segments. Display reader. The digital display reading device according to any one of claims 2 to 4, wherein the accuracy derivation means derives the accuracy of the numerical value based on the accuracy of each of the numerical values constituting the numerical value. The digital display reading device according to claim 5, wherein the accuracy derivation means derives the accuracy of the numerical value as the lowest accuracy among the accuracy of each of the numerical values constituting the numerical value. When the accuracy derivation means determines that the accuracy of any of the numbers constituting the numerical value is lower than a predetermined threshold value, the reading means determines the digital display on which the numerical value is displayed. The digital display reading device according to any one of claims 2 to 4, wherein at least the number determined to be lower than a predetermined threshold value is read again from the captured image of the above. When the accuracy derivation means determines that the accuracy of the numerical value is lower than a predetermined threshold value, the reading means rereads the numerical value from another captured image of the digital display on which the numerical value is displayed. The digital display reading device according to claim 5 or 6. Claim 7 or claim 7, further comprising a notification unit that notifies the user of predetermined information prompting the user to input the captured image to the reading means when the accuracy is determined to be lower than the predetermined threshold value in the accuracy deriving means. 8. The digital display reader according to 8. The reading means reads the numerical value from each of the plurality of captured images in which the same numerical value is captured.
The digital display reading according to claim 5 or 6, wherein the accuracy derivation means derives the accuracy for each of the numerical values read from each of the captured images, and identifies the numerical value having the highest accuracy. apparatus. The numerical value is composed of a number and a decimal point, and the decimal point is a lighting of the segment for the decimal point of each digit provided at a predetermined position with respect to the display position of the number of each digit on the digital display. Expressed by the presence or absence,
When the reading means reads the number of each digit from the photographed image, the reading means identifies a rectangular area corresponding to the display position of the number in the photographed image, and at the predetermined position with respect to the rectangular area. Specify the decimal point area and read the lighting state of the segment for the decimal point.
The accuracy derivation means derives the accuracy based on the lighting state for the segment for the decimal point of each digit, and specifies that the decimal point is displayed at the position of the segment for the decimal point with the highest accuracy. The digital display / reading device according to claim 1. The reading means includes a target closed contour whose aspect ratio is substantially equal to the aspect ratio of the digital display, which is included in the original image including the digital display as a subject, and the region inside the target closed contour is used as the captured image. The digital display reading device according to any one of claims 1 to 11, further comprising a closed contour detecting means for inputting to. The closed contour detecting means is characterized in that when a plurality of closed contours having an aspect ratio substantially equal to the aspect ratio of the digital display and having an inclusion relationship are detected, the innermost closed contour is set as the target closed contour. The digital display reading device according to claim 12. An imaging unit that captures an extraction source image that includes the digital display as a subject, and
An original image extraction means for extracting a predetermined range including the digital display centered on the line-of-sight position of the photographer at the time of shooting in the extraction source image as the original image.
The digital display reading device according to claim 12 or 13, further comprising. Further equipped with a line-of-sight detection unit that detects the line-of-sight direction of the photographer,
The digital display reading device according to claim 14, wherein the original image extracting means specifies the line-of-sight position based on the line-of-sight direction of the photographer at the time of shooting detected by the line-of-sight detection unit. The digital display reading device according to claim 14, wherein the original image extracting means specifies the center of the extraction source image as the line-of-sight position. A program for operating a computer as each means according to any one of claims 1 to 16.

Images