JP2014056450A - Image processing device and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To simultaneously realize improvement of a detection rate of an augmented reality marker and improvement of detection processing speed thereof.SOLUTION: An LED 12 is disposed as a light emitter at the position of each corner of an AR marker 11 (four-point positions). When the AR marker 11 surrounded by a plurality of LEDs 12 is photographed, the photographed image is binarized in accordance with luminance of an LED 12, the plurality of LEDs 12 are detected from within the binarized image, an area surrounded by the plurality of LEDs 12 are then further binarized, and the AR marker 11 is detected on the basis of the binarized image.

Description

  The present invention relates to an image processing apparatus and a program for detecting an augmented reality marker from a captured image.

  In general, mobile terminal devices such as smartphones have become more multifunctional, and in addition to the camera function (imaging function), an image processing function called AR (Augmented Reality) function is installed. A virtual image (for example, a character image) is synthesized and displayed at a predetermined position in the displayed image. That is, in a state where an augmented reality marker (AR marker) as a printed matter is arranged at or near the position of the subject, when the subject is photographed by the imaging function, the mobile terminal device analyzes the photographed image. By recognizing the AR marker contained therein, a virtual image is synthesized and displayed at the position of the AR marker. At this time, in order to detect the AR marker, the photographed image is converted into a gray scale, for example, and the gray scale image is binarized using an adaptive binarization method. As a technique for binarizing an image using the adaptive binarization method as described above, for example, Patent Document 1 is proposed.

JP 7-57080 A

However, in the technique for binarizing a captured image using the adaptive binarization method as described in Patent Document 1 described above, the captured image is binarized for each frame. However, for example, in the CPU performance of a mobile terminal device such as a smartphone, the processing becomes too heavy, the processing speed becomes slow, and the AR marker cannot be detected smoothly.
In general, the AR marker has a design restriction such as arranging a white pattern (mark pattern) in a square black frame in order to specify the marker region.

  An object of the present invention is to make it possible to improve the detection rate of the augmented reality marker and the detection processing speed at the same time.

In order to solve the above-described problems, the image processing apparatus of the present invention provides:
Photographing means for photographing an augmented reality marker surrounded by a plurality of light emitters;
A first binarization unit that performs processing of binarizing a captured image captured by the imaging unit with a predetermined threshold;
First detection means for detecting the plurality of light emitters from the binarized image binarized by the first binarization means;
A second binarization unit that binarizes a region surrounded by a plurality of light emitters detected by the first detection unit;
Second detection means for detecting the augmented reality marker based on the binarized image binarized by the second binarization means;
An image processing apparatus comprising:

  According to the present invention, it is possible to simultaneously improve the augmented reality marker detection rate and the detection processing speed.

The block diagram which showed the basic component of the multifunctional mobile phone with a camera function applied as an image processing apparatus. (1)-(4) is the figure which showed the detection image after binarizing the image which image | photographed the AR marker 11 enclosed by several LED12, and image | photographed it. (1), (2) is a figure for demonstrating the case where only the predetermined AR marker 11 is detected from the multiple types of AR marker 11 imaged simultaneously. The flowchart for demonstrating the augmented reality process started execution according to switching of an augmented reality photography mode. (1)-(4) is a figure corresponding to FIG.2 (1)-(4) in the case where the brightness | luminance of each LED12 differs as a modification of 1st Embodiment. (1), (2) is a figure corresponding to Drawing 2 (1) and (2) in the case where the luminescent color (three primary colors) of each LED12 differs as a modification of a 1st embodiment. (1), (2) is a figure corresponding to Drawing 2 (1) and (2) in the case where the blinking state (lighting, light extinction, blinking) of each LED12 differs as a modification of a 1st embodiment. (1)-(4) is a figure for demonstrating the process which changes the direction of the AR mark 11 as a modification of 1st Embodiment. 9 is a flowchart for explaining augmented reality processing that is started in response to switching of an augmented reality shooting mode in the second embodiment. The figure which showed the operation | movement table T used in 2nd Embodiment.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
(Embodiment 1)
First, a first embodiment of the present invention will be described with reference to FIGS.
This embodiment exemplifies a case where the image processing apparatus is applied to a multi-function mobile phone (smart phone) with a camera function. FIG. 1 illustrates basic components of the multi-function mobile phone with a camera function. It is the block diagram shown.
This multi-function mobile phone synthesizes a predetermined virtual image (for example, a character image) in an image photographed by the camera function, in addition to the basic function of the call function, e-mail function, Internet connection function, and camera function. It has an augmented reality processing function to display.

  The CPU 1 is a central processing unit that operates by supplying power from a power supply unit 2 having a secondary battery (not shown) and controls the overall operation of the mobile phone according to various programs in the storage unit 3. . The storage unit 3 is provided with a program memory M1, a work memory M2, and the like. The program memory M1 stores a program and various applications for realizing the present embodiment in accordance with the operation procedure shown in FIG. 4, and stores information necessary for the program. The work memory M2 is a work area that temporarily stores various types of information (for example, captured images, flags, timers, and the like) necessary for the operation of the mobile phone. The storage unit 3 may be configured to include a removable portable memory (recording medium) such as an SD card or an IC card, and a part of the storage unit 3 has a predetermined external server area (not shown). It may be included.

  The wireless communication unit 4 is a wide-area communication unit used at the time of a voice call function, an e-mail function, and an Internet connection function, and outputs voice from the call speaker SP via the voice signal processing unit 5 when the call function operates. Then, the input voice data from the call microphone MC is taken from the voice signal processing unit 5 and transmitted from the antenna. The touch input display unit 6 includes a display unit 6A and a touch input unit 6B. The touch input unit 6B is arranged on the front surface of the high-definition display unit 6A, and software keys (touch keys) are allocated and arranged. This device displays a function name, detects a touch operation with a finger or the like, and inputs data corresponding to the touch operation. The display unit 6A is a high-definition liquid crystal display or the like, and serves as a finder screen that displays a captured image as a live view image (monitor image) when the camera function is used.

  The imaging unit 7 constitutes a camera unit capable of photographing a subject with high definition. The imaging unit 7 includes a lens unit 7A including a zoom lens and a focus lens (not shown), an imaging element 7B such as a CCD or a CMOS, and illuminance. In addition to having a sensor unit 7C including various sensors such as a meter, color separation and gain adjustment for each RGB color component are further performed on an image signal (analog value signal) photoelectrically converted by the image sensor 7B. An analog processing circuit 7D that performs conversion to digital value data and a digital signal processing circuit 7E that performs color interpolation processing (demosaic processing) on the image data digitally converted by the analog processing circuit 7D. ing. In this embodiment, for example, it is possible to switch to various shooting modes such as a night scene shooting mode and a sunset shooting mode, and these various shooting modes include an augmented reality shooting mode. This augmented reality shooting mode is a shooting mode in which shooting is performed using the augmented reality processing function. When switched to this augmented reality shooting mode, an autofocus (AF) adjustment function, an automatic exposure (AE) adjustment function, and an automatic white The balance (AWB) adjustment function is operated.

FIG. 2 is a diagram for explaining an augmented reality marker (AR marker) 11 used during augmented reality processing.
FIGS. 2A and 2B show captured images when the AR marker 11 is captured as a subject. The AR marker 11 is disposed on the surface of an electronic device, for example, and has a configuration in which a white area pattern (an L-shaped pattern in the illustrated example) having an arbitrary shape is disposed in a square black frame, The white area has a predetermined ratio. A light emitting diode (LED) 12 is disposed as a plurality of light emitters at each of the four corner positions (four positions) of the AR marker 11.

  In other words, the entire AR marker 11 is surrounded by a plurality (four) of light emitters (LEDs) 12. And each LED12 is a point light source with a small light emission surface, and is the same LED. That is, each LED 12 is an LED of the same size that emits light with the same luminance, and its light emitting surface is disposed at the positions of the four corners of the black frame of the AR marker 11. The electronic device in which each LED 12 is arranged controls, for example, lighting and extinguishing of each LED 12. When such an AR marker 11 with LED 12 is photographed by the imaging unit 7, the photographed image has an image content as shown in FIG. 2 (1) or (2) depending on the photographing direction with respect to the AR marker 11. .

  FIG. 2A is a diagram illustrating a captured image when the AR marker 11 with the LED 12 surrounded by the plurality of LEDs 12 is captured from the front side. FIG. 2 (2) is a diagram showing a case where the AR marker 11 with the LED 12 is photographed from an oblique right direction on the back side (upper right direction on the front in the figure). Here, FIGS. 2A and 2B are diagrams conceptually showing the AR marker 11 with the LED 12 in the captured image. In the figure, the circle indicates the light emitting surface of the LED 12, the square black frame indicates the outer frame that constitutes the AR marker 11, and the internal L-shaped pattern indicates the white region that constitutes the AR marker 11, The LED 12 and AR marker 11 are also referred to in the captured image without being distinguished from the actual LED 12 and AR marker 11 (hereinafter the same). When the AR marker 11 is photographed by the imaging unit 7 in this way, the CPU 1 detects the AR marker 11 by analyzing the photographed image (actual image) shown on the image sensor 7B. Each LED 12 is detected by binarizing the captured image by a fixed threshold binarization method using a threshold value that matches the luminance of each LED 12.

  FIG. 2 (3) is a diagram showing a detection image after binarizing the image taken from the orientation shown in FIG. 2 (1), and FIG. 2 (4) is shown in FIG. 2 (2). The figure which showed the detection image after binarizing the image image | photographed from the direction which illustrated is the case where only each LED12 is detected. In the figure, a region surrounded by a broken line indicates a region where the AR marker 11 exists (hereinafter the same), and a circle indicates a detected light emitting surface. When a plurality of LEDs 12 are detected in this way, the area surrounded by the LEDs 12 is binarized according to the adaptive binarization method, and the AR marker 11 is detected based on the binarized image. ing. As described above, in the first embodiment, the captured image is binarized according to the luminance of the LED 12, and after detecting a plurality of LEDs 12 from the binarized image, within the region surrounded by the LEDs 12. Is further binarized, and the operation of detecting the AR marker 11 based on the binarized image is repeated for each frame. When the AR marker 11 is detected in this way, the CPU 1 synthesizes a predetermined virtual image (for example, a character image) at the position of the detected AR marker 11 and displays it superimposed.

FIG. 3 is a diagram for explaining a case where only a predetermined AR marker 11 is detected from a plurality of types of AR markers 11 taken at the same time.
The above-described electronic device on the AR marker 11 side is provided with four types of AR markers 11 adjacent to each other, and only the LEDs 12 surrounding any one of the AR markers 11 are selectively lit. FIG. 3A is a diagram showing a case where four types of AR markers 11 arranged on the surface of the electronic device are photographed simultaneously. In the example shown in the figure, the LED 12 is arranged at each of the four corners (four positions) of the black frame of each AR marker 11 as in the case of FIG. The case where common LED12 is arrange | positioned is illustrated. Here, in the illustrated example, in order to selectively enable only one of the four types of AR markers 11, four LEDs 12 existing around the AR marker 11 are provided. In the case where the other LEDs 12 are turned on and all the other LEDs 12 are turned off, the filled circles in the figure indicate the LEDs 12 that are turned on.

  FIG. 3B shows a detection image after the captured image shown in FIG. 3A is binarized by a fixed threshold binarization method using a threshold that matches the luminance of the LED 12. In this case, since only the four LEDs 12 that are lit are detected, the region surrounded by the LEDs 12 that are lit (regions surrounded by broken lines in the figure) is binarized according to the adaptive binarization method. Performs the process of converting to a value. As a result, only the AR marker 11 existing in the area is detected. The CPU 1 synthesizes and displays a predetermined virtual image (for example, a character image) at the detected position of the AR marker 11.

  Next, the operation concept of the multi-function mobile phone in the present embodiment will be described with reference to the flowchart shown in FIG. Here, each function described in this flowchart is stored in the form of a readable program code, and operations according to the program code are sequentially executed. Further, it is possible to sequentially execute the operation according to the above-described program code transmitted via a transmission medium such as a network. FIG. 4 is a flowchart showing an outline of the operation of the characteristic part of the present embodiment in the overall operation of the multi-function mobile phone. When the flow of FIG. 3 is exited, the main flow of the overall operation is shown. Return to (not shown).

FIG. 4 is a flowchart for explaining augmented reality processing that is started in response to switching of the augmented reality shooting mode.
First, when the CPU 1 is switched to the augmented reality shooting mode by a user operation or the like, after starting the augmented reality shooting application, the CPU 1 acquires a shot image (actual image) reflected on the image sensor 7B (step A1), and the LED 12 The captured image (actual image) is binarized by a fixed threshold binarization method using a threshold that matches the luminance of the image (step A2). In this way, by binarizing using a threshold value that matches the luminance of the LED 12, only the LED 12 can be detected from the captured image. In this case, the luminance of the LED 12 is arbitrarily set in advance by a user operation. The binarization process is performed by reading the set value (luminance value).

  Then, based on the binarized binarized image, it is checked whether or not all four LEDs 12 can be detected (step A3), and if four LEDs 12 cannot be detected (step A3). NO), the process returns to step A1 to redo the process from the beginning, but when four LEDs 12 can be detected, for example, as shown in FIG. 2 (3), (4) or FIG. 3 (2) When all four LEDs 12 can be detected from the binarized image (YES in step A3), the four-point coordinates of the LEDs 12 are acquired (step A4). The area of the AR marker 11 in the binarized image is specified based on the acquired four-point coordinates, and the binarization method of Otsu (Mr. "Automatic threshold selection method based on paper, discrimination and least square criterion", see IEICE Transactions, Vol. J63-D, No. 4, pp. 349-356 (1980)). The binarization threshold value in the area of the AR marker 11 is calculated (step A5). In this case, in the Otsu's binarization method, the threshold that minimizes the intra-class variance and maximizes the inter-class variance is calculated as the binarization threshold based on the intra-class variance and the inter-class variance in the area of the AR marker 11. get.

  Then, it is examined whether the AR marker 11 can be detected based on the binarized image binarized as described above (step A6). In this case, the AR marker 11 is recognized by specifying the pattern of the white area drawn in the AR marker 11. As a result, if the AR marker 11 cannot be detected (NO in step A6), the process returns to step A1 to redo the process from the beginning, but when the AR marker 11 can be normally detected (step A6). YES), the virtual image is read out and superimposed and displayed on the AR marker 11 in the captured image (step A7). Thereafter, it is checked whether or not the end of augmented reality processing is instructed by a user operation (step A8). If there is an end instruction (YES in step A8), the flow of FIG. (NO in step A8), the process returns to the above-described step A1, and the above-described operation is repeated.

  As described above, in the first embodiment, when the AR marker 11 surrounded by the plurality of LEDs 12 is captured by the imaging unit 7, the CPU 1 binarizes the captured image according to the luminance of the LED 12. After detecting the plurality of LEDs 12 from the binarized image, the area surrounded by the plurality of LEDs 12 is further binarized, and the AR marker 11 is detected based on the binarized image. The marker area can be specified simply by detecting the mark, the mark recognition processing speed can be improved, and the recognition of the LED 12 is more accurate than the recognition of the mark frame (black frame), so the mark recognition rate is improved. be able to. In this way, the recognition processing speed can be improved while maintaining a high recognition rate, which is highly practical. In addition, the conventional AR marker 11 has a restriction that a black frame is arranged in order to specify a mark. However, in the present embodiment, since a mark can be specified by a plurality of LEDs 12, It also has a role, and the degree of freedom in marker design is improved.

  When the captured image is binarized, binarization is performed by a fixed threshold binarization method using a threshold value that matches the luminance of the LED 12, so that erroneous detection (false recognition) is detected when the LED 12 is detected. It is possible to significantly reduce the probability of. In this case, as shown in FIG. 3, when a plurality of types of AR markers 11 are adjacently disposed on the electronic device side, and only the LEDs 12 surrounding any one of the AR markers 11 are selectively lit. By binarizing the area surrounded by the lit LED 12 from the captured image, it is possible to detect only the AR marker 11 present in the area.

In the first embodiment described above, each LED 12 emits light with the same luminance, but the luminance of each LED 12 may be different.
FIG. 5 shows a case where the brightness of each LED 12 is different. FIG. 5 (1) shows a case where the AR marker 11 is photographed from the front side as in FIG. 2 (1), and FIG. FIG. 3 is a diagram showing a case where the AR marker 11 is photographed from a diagonally right direction on the back side, similarly to FIG. 2 (2). FIG. 5A shows the case where the luminance of the upper left and lower right LEDs 12 is “dark”, the upper right LED 12 is “bright”, and the lower left LED 12 is “intermediate”. FIG. 5B shows a case where the luminance of the upper and right LEDs 12 is “dark”, the luminance of the lower LED 12 is “bright”, and the luminance of the left LED 12 is “intermediate”.

  FIG. 5 (3) is a diagram showing a detection image after binarizing the image photographed from the direction shown in FIG. 5 (1), and FIG. 5 (4) is a diagram of FIG. 5 (2). FIG. 6 is a diagram illustrating a detection image after binarizing an image photographed from the direction shown in FIG. 6, in which only each LED 12 is detected from the binarized image. In the figure, the area surrounded by the broken line indicates the area where the AR marker 11 exists, and the circle indicates the detected light emitting surface. The difference in size is the difference in luminance of the corresponding LED 12. Is shown. It is assumed that the CPU 1 recognizes the luminance pattern according to the difference in luminance of the LED 12 as described above, and detects the LED 12 on the condition that this luminance pattern is a predetermined pattern.

  When a plurality of LEDs 12 are detected from the binarized image obtained by binarizing the captured image in this way, a luminance pattern corresponding to the difference in luminance of each LED 12 is recognized, and this luminance pattern is determined in advance. If the LED 12 is detected on the condition that it is a pattern, the detection accuracy of the LED 12 can be further improved.

In the first embodiment described above, each LED 12 emits light with the same color, but the light emission color of each LED 12 may be different.
FIG. 6 shows a case where the emission colors (three primary colors) of the LEDs 12 are different. FIG. 6A shows a case where the AR marker 11 is photographed from the front side as in FIG. (2) is the figure which showed the case where the AR marker 11 was image | photographed from the diagonal right direction of the back side like FIG. 2 (2). FIG. 6A shows a case where the emission color of the upper right and lower left LEDs 12 is “green”, the emission color of the upper left LED 12 is “red”, and the emission color of the lower right LED 12 is “blue”. FIG. 6B shows the case where the emission color of the upper and lower LEDs 12 is “green”, the emission color of the right LED 12 is “red”, and the emission color of the left LED 12 is “blue”. Assume that the CPU 1 recognizes the color pattern according to the difference in the emission color of the LED 12 as described above and detects the LED 12 on the condition that this color pattern is a predetermined pattern.

  When a plurality of LEDs 12 are detected from the binarized image obtained by binarizing the captured image in this way, a color pattern corresponding to the difference in the emission color of each LED 12 is recognized, and this color pattern is determined in advance. If the LED 12 is detected on the condition that it is a pattern, the detection accuracy of the LED 12 can be further improved.

In the first embodiment described above, each LED 12 emits light and emits light. However, the blinking states (lighting, extinguishing, and blinking) of each LED 12 may be different.
FIG. 7 shows a case where the blinking states (lighting, extinguishing, blinking) of each LED 12 are different. FIG. 7A shows a case where the AR marker 11 is photographed from the front side as in FIG. FIG. 7 (2) is a diagram showing a case where the AR marker 11 is photographed from the oblique right direction on the back side thereof, similarly to FIG. 2 (2). The captured image in this case is an image for a plurality of frames corresponding to a predetermined time (for example, 1 second), and the blinking of the LED 12 is repeatedly turned on / off several times per second, for example. FIG. 7A shows the case where the upper left and upper right LEDs 12 are “flashing” and the lower left and lower right LEDs 12 are “lit”. FIG. 7B shows the case where the right and lower LEDs 12 are “flashing” and the left and upper LEDs 12 are “off”. Assume that the CPU 1 recognizes the blinking pattern corresponding to the difference in the blinking state of the LED 12 as described above, and detects the LED 12 on the condition that this blinking pattern is a predetermined pattern.

  When a plurality of LEDs 12 are detected from the binarized image obtained by binarizing the captured image in this way, the blinking pattern corresponding to the difference in the blinking state of each LED 12 is recognized, and this blinking pattern is determined in advance. If the LED 12 is detected on the condition that it is a pattern, the detection accuracy of the LED 12 can be further improved.

  Further, as described above, when the light emission pattern (luminance pattern, color pattern, blinking pattern) corresponding to the light emission state of the LED 12 is recognized, the direction of the AR mark 11 is specified according to the light emission pattern, and the AR mark 11 If is not a predetermined orientation, the entire AR mark 11 including the light emission pattern may be rotated to change the orientation so that the AR mark 11 has a predetermined orientation. FIG. 8 is a diagram for explaining the process of changing the direction of the AR mark 11. FIG. 8A is a diagram showing the light emission pattern (color pattern) of the LED 12 and the direction of the AR mark 11 stored in advance by a user operation as the predetermined direction described above, and FIG. Similarly, the AR mark 11 is taken from the front.

  FIG. 8 (2) shows an actual AR mark 11 with an LED 12, and FIG. 8 (3) shows a light emission pattern (color pattern) in a photographed image when the actual AR mark 11 with an LED 12 is photographed from an oblique direction. FIG. The light emission pattern (color pattern) shown in FIG. 8 (1) is “square”, whereas the light emission pattern (color pattern) shown in FIG. 8 (3) is “diamond”. When the entire light emission pattern (including the AR mark 11) is rotated so as to be in a predetermined direction and the direction is changed, the changed light emission pattern becomes “square” as shown in FIG. 8 (4). .

  In this way, the direction of the AR mark 11 is specified according to the light emission pattern corresponding to the light emission state of the LED 12, and when the AR mark 11 is not in a predetermined direction, the light emission is performed so that the AR mark 11 is in the predetermined direction. If the process of rotating the entire AR mark 11 including the pattern and changing the direction thereof is performed, the process of detecting the mark by binarizing the area surrounded by the light emission pattern becomes easy. .

(Second Embodiment)
Hereinafter, a second embodiment of the present invention will be described with reference to FIGS.
In the first embodiment described above, after the captured image is binarized according to the luminance of the LED 12 and a plurality of LEDs 12 are detected from the binarized image, the region surrounded by the LEDs 12 is determined. Further, the operation of binarizing and detecting the AR marker 11 based on the binarized image is repeated for each frame. In the second embodiment, the AR marker 11 is detected by the series of operations described above. If it is possible, only the LED detection operation is repeated after the next frame. Furthermore, when a predetermined virtual image (for example, a character image) is compositely displayed, the display operation of the virtual image is controlled according to the type of light emission pattern specified by the plurality of LEDs 12. Here, in both the embodiments, the same or the same names are denoted by the same reference numerals, the description thereof will be omitted, and the following description will focus on the features of the second embodiment. .

FIG. 9 is a flowchart for explaining augmented reality processing that is started in response to switching of the augmented reality shooting mode in the second embodiment.
First, the CPU 1 performs a similar process corresponding to steps A1 to A6 of FIG. 4 described above (steps B1 to B6). That is, a photographed image (actual image) is acquired (step B1), and the photographed image is binarized by a fixed threshold binarization method using a threshold matched to the luminance of the LED 12 (step B2). Then, it is checked whether or not the four LEDs 12 can be detected based on the binarized binarized image (step B3).

  As a result, if four LEDs 12 cannot be detected (NO in step B3), the process returns to step B1 to restart the process from the beginning, but when four LEDs 12 can be detected (step B3). YES), the four-point coordinates of the LED 12 are acquired (step B4). The area of the AR marker 11 in the binarized image is specified on the basis of the four-point coordinates acquired in this way, and in the area of the specified AR marker 11, using the binarization method of Otsu, A binarization threshold value in the area of the marker 11 is calculated (step B5). Then, it is checked whether or not the AR marker 11 can be detected based on the binarized binarized image (step B6). If the AR marker 11 cannot be detected (NO in step B6), from the beginning. Return to step B1 to redo the process.

Here, when the AR marker 11 can be detected (YES in step B6), the light emission pattern is specified from each detected AR marker 11, and the type of the light emission pattern is temporarily stored (step B7). ). Then, the display operation of the virtual image is specified by searching the operation table T (see FIG. 10) based on the type of the light emission pattern (step B8).
FIG. 10 is a diagram for explaining the operation table T. The operation table T is configured to store the “display operation” of the virtual image in association with the “type of light emission pattern”. In the illustrated example, as the type of light emission pattern (color pattern), one red color and one green color are displayed. “Operation A” is stored corresponding to two light emission patterns, blue, and “Operation B” is stored corresponding to one red, one green, and two blue light emission patterns. “Operation C” is stored corresponding to the light emission patterns of two red, one green, and one blue.

  The virtual image of the display operation specified in this way is superimposed and displayed on the AR marker 11 in the captured image (step B9). After detecting the AR marker 11 and displaying the virtual image at the position in this way, a captured image (actual image) is acquired (step B10), and a fixed threshold value 2 using a threshold value that matches the luminance of the LED 12 is acquired. The captured image is binarized by the binarization method (step B11). Then, it is examined whether or not four LEDs 12 can be detected based on the binarized binarized image (step B12).

  Here, when four LEDs 12 can be detected (YES in step B12), the four-point coordinates of the LEDs 12 are acquired (step B13). The area of the AR marker 11 in the binarized image is specified based on the four-point coordinates acquired in this way, and the virtual image is superimposed and displayed on the center position of the specified area of the AR marker 11 (step B14). In this case, by searching the operation table T1 corresponding to the type of the light emission pattern temporarily stored in step B7 described above, the display operation of the virtual image associated therewith is read, and the virtual image is synthesized and displayed.

  Thereafter, it is checked whether or not the end of augmented reality processing is instructed by a user operation (step B15). If there is an end instruction (YES in step B15), the flow of FIG. If not (NO in step B15), the process returns to step B10 described above, and the operation for detecting the AR marker 11 is repeated. In such a state that the operation of detecting the AR marker 11 is repeatedly executed, when the AR marker 11 is lost and cannot be detected (NO in step B12), the process is restarted from the beginning. Return to Step B1.

  As described above, in the second embodiment, when the AR marker 11 is detected, only the operation of detecting only the plurality of LEDs 12 (light emission patterns) is repeated after the next frame. The detecting operation can be omitted. Therefore, the recognition processing speed can be improved while maintaining a high recognition rate.

  In addition, when a predetermined virtual image is synthesized and displayed, the display operation of the virtual image is controlled according to the type of the light emission pattern specified by the plurality of LEDs 12, so that the virtual image can be displayed only by changing the light emission pattern. The operation can be changed, and a display rich in change can be easily realized.

  In each of the above-described embodiments, the AR marker 11 has a configuration in which a pattern of a white region having an arbitrary shape is arranged in a square black frame, but the configuration of the AR marker 11 is not limited to this and is arbitrary. Further, although the LEDs 12 are arranged at the four corners of the AR marker 11, the arrangement position and the number of the LEDs 12 are arbitrary according to the shape of the AR marker 11 and the like. That is, if the area of the AR marker 11 can be specified by arranging the LEDs 12, the arrangement position and the number of the LEDs 12 are arbitrary.

  In the above-described embodiments, when binarizing the area of the AR marker 11 in the binarized image based on the four-point coordinates of the LED 12, the binarization method of Otsu is used. However, the binarization in this case is not limited to this and is arbitrary.

  In the above-described embodiment, the case where the present invention is applied to a multi-function mobile phone (smart phone) with a camera function has been illustrated. -It may be a music player, an electronic game machine, etc. Of course, it may be a digital camera itself.

  Further, the “apparatus” and “unit” shown in each of the above-described embodiments may be separated into a plurality of cases by function, and are not limited to a single case. In addition, each step described in the above-described flowchart is not limited to time-series processing, and a plurality of steps may be processed in parallel or separately.

The embodiment of the present invention has been described above. However, the present invention is not limited to this, and includes the invention described in the claims and the equivalent scope thereof.
Hereinafter, the invention described in the claims of the present application will be appended.
(Appendix)
(Claim 1)
The invention described in claim 1
Photographing means for photographing an augmented reality marker surrounded by a plurality of light emitters;
A first binarization unit that performs processing of binarizing a captured image captured by the imaging unit with a predetermined threshold;
First detection means for performing processing for detecting the plurality of light emitters from the binarized image binarized by the first binarization means;
Second binarizing means for binarizing an area surrounded by a plurality of light emitters detected by the first detecting means;
Second detection means for performing processing for detecting the augmented reality marker based on the binarized image binarized by the second binarization means;
An image processing apparatus comprising:
(Claim 2)
The invention according to claim 2 is the image processing apparatus according to claim 1,
The first binarization unit uses a threshold value that matches the captured image with the luminance of the light emitter when the augmented reality marker surrounded by the plurality of light emitters is captured by the image capturing unit. Perform binarization processing by the fixed threshold binarization method,
This is an image processing apparatus characterized by being configured as described above.
(Claim 3)
The invention according to claim 3 is the image processing apparatus according to claim 1 or 2,
A luminance pattern recognizing unit for recognizing a luminance pattern according to a difference in luminance of each light emitter from the binarized image binarized by the first binarizing unit;
The first detection means detects the plurality of light emitters on the condition that the brightness pattern recognized by the brightness pattern recognition means is a predetermined pattern.
This is an image processing apparatus characterized by being configured as described above.
(Claim 4)
The invention according to claim 4 is the image processing apparatus according to claim 1 or 2, wherein
A color pattern recognizing unit for recognizing a color pattern corresponding to a difference in emission color of each light emitter from the binarized image binarized by the first binarizing unit;
The first detection means detects the plurality of light emitters on the condition that the color pattern recognized by the color pattern recognition means is a predetermined pattern.
This is an image processing apparatus characterized by being configured as described above.
(Claim 5)
The invention according to claim 5 is the image processing apparatus according to claim 1 or 2, wherein
A blinking pattern recognition means for recognizing a blinking pattern corresponding to a blinking state of each of the light emitters from the binarized image binarized by the first binarization means;
The first detection means detects the plurality of light emitters on the condition that the blinking pattern recognized by the blinking pattern recognition means is a predetermined pattern.
This is an image processing apparatus characterized by being configured as described above.
(Claim 6)
The invention according to claim 6 is the image processing apparatus according to any one of claims 1 to 5,
A light emission pattern recognition means for recognizing a light emission pattern corresponding to the light emission state of each light emitter from the binarized image binarized by the first binarization means;
Direction recognition means for recognizing the direction of the augmented reality marker based on the light emission pattern recognized by the light emission pattern recognition means;
Changing means for changing the augmented reality marker to a predetermined direction based on a recognition result recognized by the orientation recognition means;
Further comprising
This is an image processing apparatus characterized by being configured as described above.
(Claim 7)
The invention according to claim 7 is the image processing apparatus according to any one of claims 1 to 6,
Augmented reality by the second detection means in the state where the operations of the first binarization means, the first detection means, the second binarization means, and the second detection means are sequentially executed. When the marker is detected, thereafter, the light emitter detection operation of the first binarizing unit and the first detection unit is repeatedly executed, and the first detection unit is repeatedly executed during the repetition of the light emitter detection operation. If the illuminant cannot be detected by the detecting means, the operations of the first binarizing means, the first detecting means, the second binarizing means, and the second detecting means are performed again. Are further provided with control means for sequentially executing
This is an image processing apparatus characterized by being configured as described above.
(Claim 8)
The invention according to claim 8 is the image processing apparatus according to any one of claims 1 to 7,
Virtual image display means for displaying a virtual image at a mark position in the captured image when an augmented reality marker is detected by the second detection means;
Operation storage means for storing the display operation of the virtual image in association with a light emission pattern according to a light emission state of the plurality of light emitters;
A light emission pattern recognition means for recognizing a light emission pattern corresponding to the light emission state of each light emitter from the binarized image binarized by the first binarization means;
A virtual image control unit that reads a display operation corresponding to the light emission pattern recognized by the light emission pattern recognition unit from the operation storage unit and controls the virtual image according to the display operation;
Further comprising
This is an image processing apparatus characterized by being configured as described above.
(Claim 9)
The invention according to claim 9 is:
Against the computer,
A function of shooting an augmented reality marker surrounded by a plurality of light emitters;
A function of performing a process of binarizing the captured image with a predetermined threshold;
A function of performing processing for detecting the plurality of light emitters from the binarized binarized image;
A function of binarizing a region surrounded by the detected plurality of light emitters;
A function of performing processing for detecting the augmented reality marker based on a binarized image in which the region is binarized;
It is a program for realizing.

1 CPU
3 Storage Unit 6 Touch Input Display Unit 6A Display Unit 7 Imaging Unit 11 AR Marker 12 LED
M1 program memory

Claims (9)

Photographing means for photographing an augmented reality marker surrounded by a plurality of light emitters;
A first binarization unit that performs processing of binarizing a captured image captured by the imaging unit with a predetermined threshold;
First detection means for performing processing for detecting the plurality of light emitters from the binarized image binarized by the first binarization means;
Second binarizing means for binarizing an area surrounded by a plurality of light emitters detected by the first detecting means;
Second detection means for performing processing for detecting the augmented reality marker based on the binarized image binarized by the second binarization means;
An image processing apparatus comprising: The first binarization unit uses a threshold value that matches the captured image with the luminance of the light emitter when the augmented reality marker surrounded by the plurality of light emitters is captured by the image capturing unit. Perform binarization processing by the fixed threshold binarization method,
The image processing apparatus according to claim 1, which is configured as described above. A luminance pattern recognizing unit for recognizing a luminance pattern according to a difference in luminance of each light emitter from the binarized image binarized by the first binarizing unit;
The first detection means detects the plurality of light emitters on the condition that the brightness pattern recognized by the brightness pattern recognition means is a predetermined pattern.
The image processing apparatus according to claim 1, wherein the image processing apparatus is configured as described above. A color pattern recognizing unit for recognizing a color pattern corresponding to a difference in emission color of each light emitter from the binarized image binarized by the first binarizing unit;
The first detection means detects the plurality of light emitters on the condition that the color pattern recognized by the color pattern recognition means is a predetermined pattern.
The image processing apparatus according to claim 1, wherein the image processing apparatus is configured as described above. A blinking pattern recognition means for recognizing a blinking pattern corresponding to a blinking state of each of the light emitters from the binarized image binarized by the first binarization means;
The first detection means detects the plurality of light emitters on the condition that the blinking pattern recognized by the blinking pattern recognition means is a predetermined pattern.
The image processing apparatus according to claim 1, wherein the image processing apparatus is configured as described above. A light emission pattern recognition means for recognizing a light emission pattern corresponding to the light emission state of each light emitter from the binarized image binarized by the first binarization means;
Direction recognition means for recognizing the direction of the augmented reality marker based on the light emission pattern recognized by the light emission pattern recognition means;
Changing means for changing the augmented reality marker to a predetermined direction based on a recognition result recognized by the orientation recognition means;
Further comprising
The image processing apparatus according to claim 1, wherein the image processing apparatus is configured as described above. Augmented reality by the second detection means in the state where the operations of the first binarization means, the first detection means, the second binarization means, and the second detection means are sequentially executed. When the marker is detected, thereafter, the light emitter detection operation of the first binarizing unit and the first detection unit is repeatedly executed, and the first detection unit is repeatedly executed during the repetition of the light emitter detection operation. If the illuminant cannot be detected by the detecting means, the operations of the first binarizing means, the first detecting means, the second binarizing means, and the second detecting means are performed again. Are further provided with control means for sequentially executing
The image processing apparatus according to claim 1, wherein the image processing apparatus is configured as described above. Virtual image display means for displaying a virtual image at a mark position in the captured image when an augmented reality marker is detected by the second detection means;
Operation storage means for storing the display operation of the virtual image in association with a light emission pattern according to a light emission state of the plurality of light emitters;
A light emission pattern recognition means for recognizing a light emission pattern corresponding to the light emission state of each light emitter from the binarized image binarized by the first binarization means;
A virtual image control unit that reads a display operation corresponding to the light emission pattern recognized by the light emission pattern recognition unit from the operation storage unit and controls the virtual image according to the display operation;
Further comprising
The image processing apparatus according to claim 1, wherein the image processing apparatus is configured as described above. Against the computer,
A function of shooting an augmented reality marker surrounded by a plurality of light emitters;
A function of performing a process of binarizing the captured image with a predetermined threshold;
A function of performing processing for detecting the plurality of light emitters from the binarized binarized image;
A function of binarizing a region surrounded by the detected plurality of light emitters;
A function of performing processing for detecting the augmented reality marker based on a binarized image in which the region is binarized;
A program to realize

Images