JP2012003583A - Position detector and touch panel - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a position detector capable of accurately detecting a position of a pointer even under the outdoors etc. and a touch panel having the position detector.SOLUTION: The position detector comprises two or more imaging parts 11L and 11R which take images of the top of a detection surface 21 from different locations to produce images. And when a plurality of objects to be recognized D and S including a pointer D are recognized from produced images and a plurality of candidate positions D, S, Pa and Pb including a position of the pointer D on the detection surface 21 can be obtained, the position of the pointer D on the detection surface 21 is detected on the basis of a positional relation of the candidate positions D, S, Pa and Pb.

Description

  The present invention relates to a position detection device that detects the position of an indicator on a detection surface, and a touch panel that includes the position detection device.

  A so-called touch panel that includes a display device that displays an image and a position detection device that detects a position designated by a user and detects a position on the display device designated by the user is widely used. The touch panel is intuitive and allows a user who has visually recognized an image displayed on the display device to touch a predetermined position on the display device (on the image) with a pointer such as a finger or a stylus, using the image as a clue. It is operated by a simple operation.

  As a method of the position detection device provided in the touch panel, for example, there is a capacitance method. The capacitance method is a method in which a conductive film is provided on a display device and a change in capacitance caused by a finger touching or approaching the conductive film is detected as a current change through a current. It detects the close position.

  However, in the capacitance method, since it is necessary to provide a detection member such as a conductive film on the display device, the light transmittance on the display device is lowered, and the visibility of the image is deteriorated outdoors. It becomes. In addition, since the capacitance method requires a detection member having a size corresponding to the display device, there is a problem that the larger the display device, the more complicated the manufacturing process and the higher the manufacturing cost. Arise.

  As a method for solving the above problem, Patent Document 1 discloses an infrared retroreflection method (equipped with an infrared light emitting unit and an image sensor at two corners of the detection surface and an infrared reflector at the end of the detection surface, There has been proposed a position detection apparatus of a type that detects a position touched by a user based on an image generated by an image sensor.

  Moreover, in patent document 2, while each of indicators, such as a user's hand and a finger, and the shadow of an indicator are detected from the image data obtained by imaging of a video camera, based on the position of an indicator and a shadow. Thus, a position detection device for obtaining the position and orientation of the indicator has been proposed. In this position detection device, the indicator and shadow are detected by differentiating the parameter (hue value and saturation value) for detecting the indicator and the parameter (luminance value) for detecting the shadow. Then, the position and orientation of the indicator are detected based on the detected indicator and the position of the shadow.

Japanese Patent No. 4229964 JP 2008-59283 A

  In the infrared retroreflection method as described in Patent Document 1, since it is not necessary to provide a detection member such as a conductive film on the display device, the light transmittance on the display device is reduced (the visibility is deteriorated outdoors). It is possible to prevent the manufacturing process from becoming complicated and the manufacturing cost from increasing. However, since the detection accuracy is deteriorated outdoors due to the influence of infrared rays radiated from other than the light emitting unit (for example, the sun) and the influence of the shadow of the indicator, it is problematic.

  On the other hand, with the position detection device described in Patent Document 2, it is possible to distinguish and detect the indicator and the shadow. However, in this position detection device, since the hand or finger that is the indicator is detected based on the hue value and the saturation value, for example, when the indicator becomes dark under an indefinite light source such as outdoors, the indicator and the background Cannot be distinguished. Similarly, when the luminance value of the indicator becomes close to the luminance value of the shadow, the indicator and the shadow cannot be distinguished. Furthermore, if a shadow is formed at a position away from the indicator, the position and orientation of the indicator cannot be accurately detected. Therefore, in the above-described position detection device, it is necessary to irradiate light having a constant light amount from a fixed position, and if it is used outdoors where the illumination is indefinite, the detection accuracy deteriorates, which causes a problem.

  Therefore, an object of the present invention is to provide a position detection device that can accurately detect the position of an indicator even under outdoor conditions and a touch panel including the position detection device.

  In order to achieve the above object, a position detection device according to the present invention is a position detection device that detects the position of a pointer on a detection surface, and generates an image by imaging the detection surface from different locations. Two or more imaging units, and a position detection unit that detects the position of the indicator on the detection surface by recognizing the indicator from an image generated by imaging of the imaging unit, and at least When a plurality of recognition objects including the indicator are recognized from a plurality of images generated by imaging by one imaging unit, and a plurality of candidate positions including the position of the indicator on the detection surface can be obtained. The position detection unit detects the position of the indicator on the detection surface based on a positional relationship between the plurality of candidate positions.

  In the following embodiments, a left image processing unit, a right image processing unit, and a pointer position calculation unit are described as examples of the position detection unit.

  Further, in the position detection device having the above configuration, the recognized object recognized by the position detection unit from the image generated by the imaging of the imaging unit may include the indicator and the shadow of the indicator. It does not matter.

  With this configuration, even when the indicator and the shadow of the indicator generated by light irradiation become a recognized object and a plurality of candidate positions can be obtained, the indicator can be selected from the candidate positions. Corresponding candidate positions can be obtained. Therefore, it is possible to detect the position of the indicator with high accuracy even under conditions where shadows such as outdoors can occur.

  Further, in the position detection device having the above configuration, the imaging unit is arranged on one end side of the detection surface, and the position detection unit preferentially selects a candidate position on the end side of the detection surface. Then, the position of the indicator on the detection surface may be set.

  With this configuration, each angle formed by a line connecting the two imaging units and each line connecting each imaging unit and a non-indicator (for example, a shadow) is a line connecting the two imaging units. When the angle is equal to or greater than the corresponding angle formed by each of the imaging unit and each line connecting the indicators, the candidate position corresponding to the indicator is preferentially selected and set as the indicator position. It becomes possible. In the following embodiments, a candidate position on the upper side of the display area is described as an example of the candidate position on the edge side. In addition, the position detection device may emit light from the end side toward the detection surface.

  In the position detection device having the above configuration, the position detection unit is configured to obtain the candidate position in the candidate position located outside the range of the detection surface and each image obtained by imaging of the imaging unit. Other candidate positions excluding candidate positions obtained by different recognized objects from the recognized objects may be preferentially selected as the position of the indicator on the detection surface.

  With this configuration, it is possible to preferentially select other candidate positions excluding candidate positions that do not correspond to either the indicator or a non-indicator (for example, a shadow), and use them as the indicator positions. It becomes. In addition, at least one of the angles formed by the lines connecting the two imaging units and the lines connecting the imaging units and the non-indicator is the line connecting the two imaging units, the imaging units, The candidate positions that are smaller than the corresponding angles formed by the respective lines connecting the indicators and that do not correspond to either the indicator or the non-indicator than the candidate positions corresponding to the indicator When it is formed on the side, it is possible to prevent the candidate position that does not correspond to either the indicator or the non-indicator from being preferentially selected to become the position of the indicator.

  Further, in the position detection device having the above configuration, the position detection unit preferentially selects a candidate position obtained from a recognized object having a large variation in image signal value, and the position of the indicator on the detection surface It does not matter as

  If comprised in this way, it will become possible to preferentially select the candidate position calculated | required from the to-be-recognized body with high possibility of being a pointer, and let it be the position of a pointer.

  In the position detection device having the above configuration, the position detection unit preferentially selects a candidate position obtained from a recognition object having a large variance of image signal values, and the indicator on the detection surface is selected. It does not matter as the position.

  In addition, a touch panel according to the present invention includes a display device that displays an image in a display area, and the position detection device that uses the display area as a detection surface.

  With the configuration of the present invention, even if a plurality of recognition objects are recognized from an image generated by imaging of the imaging unit and a plurality of candidate positions can be obtained, the positional relationship between the plurality of candidate positions The candidate position corresponding to the indicator can be obtained. Therefore, it is possible to detect the position of the indicator with high accuracy even under outdoor conditions.

  The significance or effect of the present invention will be further clarified by the following description of embodiments. However, the following embodiment is merely one of the embodiments of the present invention, and the meaning of the terminology of the present invention or each constituent element is limited to those described in the following embodiments. is not.

These are the top view and side view which show an example of a structure of the touchscreen which is one Embodiment of this invention. These are block diagrams which show an example of a structure of a position detection apparatus. These are figures explaining a left side imaging part. These are figures explaining a right side imaging part. These are figures which show the 1st example of a situation where a pointer exists on the surface of a display area, and a shadow is reflected on a frame. These are figures which show the 2nd example of a situation where a pointer exists on the surface of a display area, and a shadow is reflected on a frame. These are figures which show the 3rd example of a situation where a pointer exists on the surface of a display area, and a shadow is reflected on a frame. These are figures which show the 4th example of a situation where a pointer exists on the surface of a display area, and a shadow is reflected on a frame. These are figures which show the 5th example of a situation where a pointer exists on the surface of a display area, and a shadow is reflected on a frame. These are flowcharts which show an example of the whole operation | movement of a position detection apparatus. These are flowcharts which show an example of the operation | movement which calculates a left side difference image. These are figures which show an example of the setting method of a processing range. These are flowcharts which show an example of the operation | movement which recognizes a to-be-recognized body from a left difference image. These are flowcharts which show an example of the operation | movement which recognizes the position of a pointer.

<Configuration of touch panel>
A touch panel according to an embodiment of the present invention will be described below with reference to the drawings. First, an example of the configuration of the touch panel and the position detection device will be described with reference to the drawings.

  1A and 1B are a plan view and a side view illustrating an example of a configuration of a touch panel according to an embodiment of the present invention. FIG. 1A is a plan view showing a surface on which an image of the touch panel is displayed. On the other hand, FIG.1 (b) is a side view which shows the left side surface of a touchscreen when the surface shown to Fig.1 (a) is turned to the right. FIG. 2 is a block diagram illustrating an example of the configuration of the position detection device.

  As shown in FIG. 1, the touch panel 1 of the present embodiment includes a display device 20 in which a frame body 22 that protrudes from the surface of the display area 21 is arranged around a display area 21 that displays an image. In addition, the touch panel 1 includes a position detection device 10 (see FIG. 2) including a left imaging unit 11L and a right imaging unit 11R that images the display area 21 of the touch panel 1. Further, the position detection device 10 detects the position of the indicator on the surface of the display area 21 using the surface of the display area 21 as a detection surface.

  FIG. 1 illustrates an example in which the left imaging unit 11L and the right imaging unit 11R are arranged at both ends of one side of the display device 20 that is rectangular on the plane illustrated in FIG. Hereinafter, in the plane shown in FIG. 1A, the side on which the left imaging unit 11L and the right imaging unit 11R are arranged is the upper side, the side facing the upper side of the display device 20 is the lower side, the upper side is the upper side, and the lower side is the lower side. When arranged, the left side is the left side and the right side is the right side. Further, the frame body 22 has a width of a predetermined length from each side of the display device 20 toward the display area 21.

  The left imaging unit 11L is disposed at an end portion on the left side of the upper side of the display device 20, and images the lower side and the right side. Further, the right imaging unit 11R is disposed at an end portion on the right side of the upper side of the display device 20, and images the lower side and the left side. In FIG. 1, the directions in which the left imaging unit 11L and the right imaging unit 11R capture images are indicated by white arrows. The arrangement mode of the left imaging unit 11L and the right imaging unit 11R shown in FIG. For example, a part of the left imaging unit 11L and the right imaging unit 11R (for example, a part of an imaging element or a lens) may be arranged so as to be embedded around the display area 21.

  As shown in FIG. 2, the position detection device 10 includes the left imaging unit 11L and the right imaging unit 11R, and a left image processing unit 12L that processes a left image generated by imaging of the left imaging unit 11L. A left background image storage unit 13L that temporarily stores a left background image used for processing in the left image processing unit 12L, a right image processing unit 12R that processes a right image generated by imaging of the right imaging unit 11R, and Right-side background image storage unit 13R that temporarily stores the right-side background image used for processing in the right-side image processing unit 12R, left-side image processing results by the left-side image processing unit 12L, and right-side image processing by the right-side image processing unit 12R And a pointer position calculation unit 14 that calculates the position of the pointer on the surface of the display area 21 based on the result and outputs it as pointer position information. The indicator is used by the user to indicate an arbitrary position in the display area 21 such as the user's finger or stylus. In the following description, a finger is exemplified as an indicator for the purpose of concrete description.

  The left image generated by the imaging of the left imaging unit 11L and the right image generated by the imaging of the right imaging unit 11R each have an angle of view of about 90 °, and the entire surface of the display area 21 falls within the angle of view. Is preferable. Further, it is preferable that the left imaging unit 11L and the right imaging unit 11R include a light receiving element that detects visible light. In addition, the left imaging unit 11L and the right imaging unit 11R preferably include an image sensor that can generate a color image, such as a single-plate Bayer array image sensor or a three-plate image sensor. An image sensor that can generate a monochrome image, such as an image sensor, may be provided.

  Note that the left background image storage unit 13L and the right background image storage unit 13R may be unnecessary depending on the processing contents (details will be described later) of the left image processing unit 12L and the right image processing unit 12R. In such a case, the left background image storage unit 13L and the right background image storage unit 13R may be omitted.

  Further, when the user uses the touch panel 1, the position detection device 10 and the touch panel 1 are designed so that the left imaging unit 11 </ b> L and the right imaging unit 11 </ b> R are arranged on the light source side with respect to the display area 21 ( For example, the direction in which the image is displayed in the display area 21 and the positions and directions of the pedestals and mounting members that support the touch panel 1 are those that realize the above state). Specifically, for example, when the light source is a general illumination device such as an electric light or the sun, the position detection device is configured such that the left imaging unit 11L and the right imaging unit 11R are above the display region 21 in the vertical direction. 10 or the touch panel 1 is preferable.

  If comprised as mentioned above, the light of a light source will be irradiated toward the display area 21 from the side in which the left imaging part 11L and the right imaging part 11R of the display apparatus 20 were provided. Therefore, it is possible to prevent the light of the light source from directly entering the left imaging unit 11L and the right imaging unit 11R and capturing an unclear left image and right image. Although detailed description will be given later, the relative positional relationship between the indicator and a disturbance (for example, a shadow or reflected light of the indicator) caused by the indicator appearing on the surface of the display area 21. Can be limited to such an extent that the indicator can be detected with high accuracy. In the detailed description to be described later, for the sake of specific description, a case where the touch panel 1 has the above-described configuration and a shadow of an indicator is generated as a disturbance is illustrated.

<Method for calculating the position of the indicator>
A method for calculating the position of the indicator based on the left image and the right image will be described with reference to FIGS. FIG. 3 is a diagram illustrating the left imaging unit, and FIG. 4 is a diagram illustrating the right imaging unit. FIG. 3A is a plan view showing an angle of view captured by the left imaging unit 11L, and shows the same plane as FIG. FIG. 3B is a diagram illustrating a left image obtained by imaging by the left imaging unit 11L illustrated in FIG. Similarly, FIG. 4A is a plan view showing an angle of view captured by the right imaging unit 11R, and shows the same plane as FIG. FIG. 4B is a diagram illustrating a right image obtained by imaging by the right imaging unit 11R illustrated in FIG.

  As shown in FIGS. 3A and 3B, the left imaging unit 11L generates a left image including the upper side to the left side of the display device 20 around itself. Here, considering an angle centered on the upper left of the display device 20 where the left imaging unit 11L is arranged, if the upper side is defined as 0 ° and the left side is defined as 90 °, the direction in which the indicator D within the angle of view is It is expressed by the clockwise angle α. In addition to the display area 21, the left image in this example is displayed with a frame 22 (the portion displayed in gray in the figure, the same applies to other figures) and the side of the display device (lattice hatching in the figure). The same applies to other figures.) The left end corresponds to α = 90 °, and the right end corresponds to α = 0 °. Note that the angle α may be interpreted as an angle formed by the upper side of the display device 20 (a line connecting the left imaging unit 11L and the right imaging unit 11R) and a line connecting the left imaging unit 11L and the indicator D. it can.

  The same applies to the right imaging unit 11R. As shown in FIGS. 4A and 4B, the right imaging unit 11R generates a right image including the upper side to the right side of the display device 20 around itself. Here, considering an angle centered on the upper right side of the display device 20 in which the right imaging unit 11R is arranged and defining the upper side as 0 ° and the right side as 90 °, the direction in which the indicator D within the angle of view is It is represented by an angle β counterclockwise (counterclockwise). Further, the right image in this example includes the frame 22 and the side surface of the display device in addition to the display area 21, and the left end corresponds to β = 0 ° and the right end corresponds to β = 90 °. The angle β may be interpreted as an angle formed by the upper side of the display device 20 (a line connecting the left imaging unit 11L and the right imaging unit 11R) and a line connecting the right imaging unit 11R and the indicator D. it can.

  The angles α and β correspond to the horizontal position of the indicator D in the left image and the right image, as shown in FIGS. 3B and 4B. Therefore, the left image processing unit 12L calculates the angle α by recognizing the indicator D from the left image and obtaining the position thereof. Similarly, the right image processing unit 12R calculates the angle β by recognizing the indicator D from the right image and obtaining the position thereof. Note that the left image processing unit 12L and the right image processing unit 12R may recognize the indicator D from the left image and the right image, or an image obtained by processing the left image and the right image (for example, a left side to be described later). The indicator D may be recognized from the difference image and the right difference image. In any case, the indicator D is recognized from each of the left image and the right image. Details of how the left image processing unit 12L and the right image processing unit 12R recognize the indicator D will be described later.

The indicator position calculation unit 14 performs, for example, triangulation using the angle α and the angle β calculated by the left image processing unit 12L and the right image processing unit 12R and the known upper side length k of the display device 20. Thus, the position of the indicator D on the surface of the display area 21 is calculated. Specifically, the calculated position is expressed by coordinates (x, y) indicating an arbitrary position in the plane including the display area 21. Here, when the upper left coordinate of the display device 20 is defined as (0, 0) and the value of the x coordinate is larger toward the right and the value of the y coordinate is larger toward the lower side, by solving the following simultaneous equations, The coordinates (x ₁ , y ₁ ) of the indicator D on the surface of the display area 21 can be calculated.

x ₁ × tan α = y ₁
(K−x ₁ ) × tan β = y ₁

  The indicator position calculation unit 14 generates and outputs indicator position information indicating the position of the indicator D on the surface of the display area 21 calculated as described above. Then, the display device 20 performs various operations such as switching an image to be displayed in the display area 21 based on the indicator position information (that is, based on the operation content of the touch panel 1 by the user).

  Note that the above calculation method is merely an example, and other calculation methods may be applied. For example, the left imaging unit 11L is arranged at the upper left of the display area 21, the right imaging unit 11R is arranged at the upper right of the display area 21, and the clockwise angle from the upper side of the display area 21 to the indicator D is α The counterclockwise angle from the upper side of the display area 21 to the indicator D is β, the upper left of the display area 21 is (0, 0), the x-coordinate value increases toward the right, and the y-coordinate value decreases toward the bottom. You may define so that it may become large. Also in this case, the coordinates of the indicator D on the surface of the display area 21 can be calculated according to the above simultaneous equations.

<Candidate position selection method>
As described above, if the left image processing unit 12L and the right image processing unit 12R can recognize the indicator D from each of the left image and the right image, the position of the indicator D on the surface of the display region 21 is detected. can do. However, the left image processing unit 12L and the right image processing unit 12R recognize the recognition target (the left image processing unit 12L and the right image processing unit 12R exist on the surface of the display area 21 from the left image and the right image. It does not matter whether there is an entity, that is, it is not limited to an object such as the indicator D, and may include disturbances such as shadows and reflected light, and so on. As well as disturbances such as shadows may be included. In particular, when a shadow appears in the frame 22, there is a high possibility that the shadow is recognized as a recognition object in at least one of the left image processing unit 12L and the right image processing unit 12R.

  In the position detection device 10 of the present example, the accuracy is obtained by selecting a candidate position corresponding to the indicator D from among a plurality of candidate positions that can be obtained by recognizing a plurality of recognition objects as described above. The position of the indicator D is often detected. Specifically, even if the left image processing unit 12L and the right image processing unit 12R recognize a plurality of recognized objects, even if a plurality of directions (angles) where the recognized objects exist are obtained, the pointer position calculation is performed. The unit 14 detects the position of the indicator D with high accuracy by selecting a correct direction (angle) in which the indicator D is present. Hereinafter, a specific example of this candidate position selection method will be described by exemplifying a case where a shadow appears in the frame 22 on the lower side of the display device 20.

[First example]
First, the selection method of the first example will be described with reference to the drawings. FIG. 5 is a diagram illustrating a first situation example in which the indicator is present on the surface of the display area and the shadow is reflected in the frame. Fig.5 (a) is a top view of a touchscreen, and shows the same plane as Fig.1 (a). FIG. 5B is a diagram showing a left image obtained by imaging of the left imaging unit 11L shown in FIG. 5A, and FIG. 5C is a diagram of the right imaging unit 11R shown in FIG. It is a figure which shows the right side image obtained by imaging. In FIGS. 5A to 5C, in order to clearly indicate the positions of the indicator D and the shadow S, a circle (the indicator D is a white circle and the shadow S is a black circle) is shown. Is displayed.

  As shown in FIG. 5, in the first situation example, an angle αD indicating the indicator D (an angle indicating a direction in which the indicator D exists with respect to the left imaging unit 11L, and is expressed according to the definition of the angle α described above. Than the angle αS indicating the shadow S (the angle indicating the direction in which the shadow S is present with respect to the left imaging unit 11L, the angle expressed according to the definition of the angle α described above, and so on). Get smaller. In addition, an angle βD indicating the indicator D (an angle indicating the direction in which the indicator D is present with respect to the right imaging unit 11R, and an angle expressed according to the definition of the angle β described above, hereinafter the same), The angle βS is smaller than the angle βS (the angle indicating the direction in which the shadow S exists with respect to the right imaging unit 11R, the angle expressed according to the definition of the angle β described above, and so on).

  In the first situation example, each of the left image processing unit 12L and the right image processing unit 12R recognizes each of the two objects to be recognized (indicator D and shadow S). Therefore, the indicator position calculation unit 14 selects the position of the indicator D from the following four candidate positions.

(1) Candidate position (position of indicator D) defined by angle αD and angle βD
(2) Candidate positions defined by the angle αD and the angle βS (the indicator when the left image processing unit 12L recognizes the indicator D as the indicator and the right image processing unit 12R recognizes the shadow S as the indicator. The position of the false candidate Pa where neither D nor shadow S exists)
(3) Candidate positions defined by the angles αS and βD (indicator in the case where the left image processing unit 12L recognizes the shadow S as an indicator and the right image processing unit 12R recognizes the indicator D as an indicator The position of the false candidate Pb in which neither D nor shadow S exists)
(4) Candidate position (position of shadow S) defined by angle αS and angle βS

  In the selection method of the first example, the indicator position calculation unit 14 is located above the candidate positions (1) to (4) (in the direction in which the left imaging unit 11L, the right imaging unit 11R, and the light source exist), candidate position (1). Is preferentially selected as the position of the indicator D. In other words, the indicator position calculation unit 14 determines that the angle (angle αD or angle αS) with respect to the left imaging unit 11L and the angle (angle βD or angle βS) with respect to the right imaging unit 11R from the candidate positions (1) to (4). The candidate position (1) that is both small is preferentially selected and set as the position of the indicator D.

  Therefore, when the selection method of the first example is performed in the first situation example, the indicator position calculation unit 14 preferentially selects the candidate position (1) corresponding to the indicator D, and the position of the indicator D is determined. It becomes possible to do.

  Another example of the situation will be described with reference to the drawings. FIG. 6 is a diagram illustrating a second situation example in which the indicator is present on the surface of the display area and the shadow is reflected in the frame. FIG. 6A to FIG. 6C correspond to FIG. 5A to FIG. 5C, respectively, showing the first situation example. Moreover, the same code | symbol is attached | subjected to the thing similar to a 1st example of a situation, and the detailed description is abbreviate | omitted.

  As shown in FIG. 6, in the second situation example, the angle αD is smaller than the angle αS. Further, the angle βD and the angle βS are equal. Further, in the second situation example, the left image processing unit 12L recognizes each of the two recognition objects (indicator D and shadow S), but the right image processing unit 12R has one recognition object (in front of the shadow S). Is recognized. Therefore, the indicator position calculation unit 14 selects the position of the indicator D from the following two candidate positions.

(1) Candidate position defined by angle αD and angle βD (βS) (position of indicator D)
(4) Candidate position (position of shadow S) defined by angle αS and angle βD (βS)

  Therefore, when the selection method of the first example is performed in the second situation example, the indicator position calculation unit 14 preferentially selects the candidate position (1) corresponding to the indicator D, and the position of the indicator D is determined. It becomes possible to do.

  The case where the angle αD is equal to the angle αS and the angle βD is smaller than the angle βS is the same as in the second situation example. That is, the indicator position calculation unit 14 can preferentially select the candidate position (1) corresponding to the indicator D and set it as the position of the indicator D.

[Second example]
Next, the selection method of the second example will be described with reference to the drawings. FIG. 7 is a diagram illustrating a third situation example in which the indicator exists on the surface of the display area and the shadow appears in the frame. FIGS. 7A to 7C correspond to FIGS. 5A to 5C, respectively, illustrating the first situation example. Moreover, the same code | symbol is attached | subjected to the thing similar to a 1st example of a situation, and the detailed description is abbreviate | omitted.

  As shown in FIG. 7, in the third situation example, the angle αD is smaller than the angle αS. Further, the angle βD is larger than the angle βS. Furthermore, in the third situation example, each of the left image processing unit 12L and the right image processing unit 12R recognizes each of the two recognition objects (indicator D and shadow S). Therefore, the indicator position calculation unit 14 selects the position of the indicator D from the following four candidate positions.

(1) Candidate position (position of indicator D) defined by angle αD and angle βD
(2) Candidate position (position of false candidate Pa) defined by angle αD and angle βS
(3) Candidate position defined by angle αS and angle βD (position of false candidate Pb)
(4) Candidate position (position of shadow S) defined by angle αS and angle βS

  In the third situation example, since the angle βS is larger than the angle βD, the uppermost candidate position (the direction in which the left imaging unit 11L, the right imaging unit 11R, and the light source exist) is the candidate position corresponding to the false candidate Pa ( 2). In other words, the candidate position where the angle with respect to the left imaging unit 11L (angle αD or angle αS) and the angle with respect to the right imaging unit 11R (angle βD or angle βS) are the smallest is the candidate position corresponding to the false candidate Pa (2 )become.

  In such a case, in the selection method of the first example described above, it is difficult for the indicator position calculation unit 14 to select the candidate position (1) corresponding to the indicator D as the position of the indicator D. Become. Therefore, using the selection method of the second example, the indicator position calculation unit 14 prevents the candidate position (2) corresponding to the false candidate Pa from being preferentially selected as the position of the indicator D. In other words, the indicator position calculation unit 14 preferentially selects other candidate positions excluding the candidate position (2), and sets it as the position of the indicator D.

  Specifically, in the selection method of the second example, the indicator position calculation unit 14 first confirms the candidate position (3) located outside the display area 21. Then, the indicator position calculation unit 14 determines the candidate position (3) and the recognition target for obtaining the candidate position (3) (the left image processing unit 12L has a shadow S, and the right image processing unit 12R has an indicator D. ), The candidate position (2) obtained by a different recognition object (the left image processing unit 12L is the indicator D and the right image processing unit 12R is the shadow S) corresponds to both the indicator D and the shadow S. Can be confirmed (corresponding to false candidate Pa and false candidate Pb).

  When the candidate position (2) corresponding to the false candidate Pa is positioned above the candidate position (1) of the indicator D, the candidate position (3) of the other false candidate Pb is placed on the frame body 22. This is because it is located below the candidate position (4) of the shadow S to be reflected (that is, outside the range of the display area 21). Therefore, by confirming the candidate position (3) located outside the display area 21 as described above, the candidate position (2) located above the candidate position (1) of the indicator D is confirmed. Can do.

  Then, the indicator position calculation unit 14 indicates that the candidate positions (2) and (3) that are confirmed to correspond to neither the indicator D nor the shadow S (corresponding to the false candidate Pa and the false candidate Pb) are indicated. The position of the body D is not preferentially selected. In other words, the indicator position calculation unit 14 preferentially selects the other candidate positions (1) and (4) excluding the candidate positions (2) and (3) as the position of the indicator D. It should be noted that only the candidate position (2) that is not confirmed to be located outside the range of the display area 21 among the candidate positions (2) and (3) is not preferentially selected as the position of the indicator D. It doesn't matter.

  Therefore, when the selection method of the second example is performed in the third situation example, the indicator position calculation unit 14 does not preferentially select the candidate position (2) that does not correspond to the indicator D as the position of the indicator D. It becomes possible to. Furthermore, by performing the selection method of the first example in the third situation example, the indicator position calculation unit 14 preferentially selects the candidate position (1) corresponding to the indicator D, and the position of the indicator D It becomes possible.

  The case where the angle αD is larger than the angle αS and the angle βD is smaller than the angle βS is the same as in the third situation example. That is, the indicator position calculation unit 14 can preferentially select the candidate position (3) that does not correspond to the indicator D as the position of the indicator D. Furthermore, by performing the selection method of the first example in the third situation example, the indicator position calculation unit 14 preferentially selects the candidate position (1) corresponding to the indicator D, and the position of the indicator D It becomes possible.

Further, when the indicator position calculation unit 14 can obtain the four candidate positions (1) to (4), it is confirmed whether or not all the candidate positions (1) to (4) are out of the display area 21. It doesn't matter. Also, the lowest candidate position (the side away from the left imaging unit 11L and the right imaging unit 11R, the side away from the light source) among the four candidate positions (1) to (4), in other words, the left imaging unit 11L. Whether or not the candidate position has a large angle with respect to (angle αD or angle αS) and the angle with respect to the right imaging unit 11R (angle βD or angle βS) may be confirmed. Further, the indicator position calculation unit 14 may check whether or not the display area 21 is out of the range by calculating the coordinates (x ₁ , y ₁ ) of the candidate position.

  A situation such as the third situation example may occur when the light source exists at a position inclined from the upper side of the display device 20 to the left side or the right side. Therefore, when the light source cannot be in such a position, the selection method of the second example may not be performed. However, by performing the selection method of the second example in addition to the selection method of the first example, the indicator position calculation unit 14 selects the candidate position (1) corresponding to the indicator D more preferentially, The position of the indicator D can be set.

  When four candidate positions can be obtained as in the first situation example and the third situation example, the candidate position (1) of the indicator D, the candidate position (4) of the shadow S, and the candidate position of the false candidate Pa Each set of (2) and the candidate position (3) of the false candidate Pb is located at the diagonal of a quadrangle composed of candidate positions (1) to (4). Therefore, if a candidate position corresponding to one recognized body in a certain set is known, a candidate position corresponding to the other recognized object in a certain set is known. Specifically, for example, if the candidate position (1) corresponding to the indicator D is known, it can be seen that the candidate position (4) located on the diagonal corresponds to the shadow S. Further, for example, if it is known that the candidate position (3) corresponds to the false candidate Pb, it can be understood that the candidate position (2) located on the diagonal also corresponds to the false candidate Pa.

  The above-described first to third situation examples to which the selection methods of the first example and the second example can be applied are all in the case where a shadow S is reflected in the frame 22 on the lower side of the display device 20, In addition to this, there may be a situation in which a shadow S appears in the left side frame and the right side frame 22 of the display device 20. In this situation, for example, when the pointer position calculation unit 14 performs the selection method of the first example described above, the candidate position (1) corresponding to the pointer D is preferentially selected, and the position of the pointer D is determined. It becomes possible to do. Hereinafter, the case where the shadow S appears in the left side frame and the right side frame 22 of the display device 20 will be described with reference to the drawings.

  FIG. 8 is a diagram illustrating a fourth situation example in which the indicator exists on the surface of the display area and the shadow appears in the frame. FIG. 8A to FIG. 8C correspond to FIG. 5A to FIG. 5C, respectively, showing the first situation example. Moreover, the same code | symbol is attached | subjected to the thing similar to a 1st example of a situation, and the detailed description is abbreviate | omitted.

  As shown in FIG. 8, in the fourth situation example, the angle αD is smaller than the angle αS. In addition, a shadow S appears in the frame 22 on the right side of the display device 20. Thereby, the left image processing unit 12L recognizes each of the two recognized objects (indicator D and shadow S), but the right image processing unit 12R recognizes one recognized object (indicator D). Therefore, the indicator position calculation unit 14 selects the position of the indicator D from the following two candidate positions.

(1) Candidate position (position of indicator D) defined by angle αD and angle βD
(3) Candidate position defined by angle αS and angle βD (position of false candidate Pb)

  Therefore, when the selection method of the first example is performed in the fourth situation example, the indicator position calculation unit 14 preferentially selects the candidate position (1) corresponding to the indicator D, and the position of the indicator D is determined. It becomes possible to do.

  Further, the case where the shadow S appears in the frame 22 on the left side of the display device 20 and the angle βD is smaller than the angle βS is the same as in the fourth situation example. That is, the indicator position calculation unit 14 can preferentially select the candidate position (1) corresponding to the indicator D and set it as the position of the indicator D.

  Another example of the situation will be described with reference to the drawings. FIG. 9 is a diagram illustrating a fifth situation example in which the indicator exists on the surface of the display area and the shadow appears in the frame. FIGS. 9A to 9C correspond to FIGS. 5A to 5C, respectively, showing the first situation example. Moreover, the same code | symbol is attached | subjected to the thing similar to a 1st example of a situation, and the detailed description is abbreviate | omitted.

  As shown in FIG. 9, in the fifth situation example, a shadow S appears in the frame 22 on the right side of the display device 20, and the angle αD is equal to the angle αS. Therefore, the left image processing unit 12L recognizes one object to be recognized (indicator D existing in front of the shadow S), and the right image processing unit 12R also recognizes one object to be recognized (indicator D).

  In this case, the indicator position calculation unit 14 calculates only the candidate position (1) corresponding to the indicator D. Therefore, the candidate position (1) corresponding to the indicator D can be selected and set as the position of the indicator D.

  Further, the case where the shadow S appears in the frame 22 on the left side of the display device 20 and the angle βD and the angle βS are equal is the same as in the fifth situation example. That is, the candidate position (1) corresponding to the indicator D can be selected and set as the position of the indicator D.

  By the way, when the shadow S is reflected on the left side frame and the right side frame 22 of the display device 20, it is considered that the following two situations may occur in addition to the fourth situation example and the fifth situation example described above. It is done.

1. The shadow S is reflected in the frame 22 on the right side of the display device 20, and the angle αD is larger than the angle αS.
2. The shadow S is reflected in the frame 22 on the left side of the display device 20, and the angle βD is larger than the angle βS.

  However, such a situation cannot occur unless the light source exists on the lower side of the display device 20. Therefore, the position detection device 10 and the touch panel 1 designed so that the left imaging unit 11L and the right imaging unit 11R as described above are arranged closer to the light source than the display area 21 cannot be generated.

[Third example]
Next, the selection method of the third example will be described. The selection method of the third example is based on the state of the left image and the right image, unlike the selection methods of the first example and the second example described above. Specifically, the indicator D is discriminated from the plurality of recognized objects using the difference in characteristics between the image showing the indicator D and the image showing the disturbance such as the shadow S. Then, a candidate position obtained from a recognized object distinguished from the indicator D is preferentially selected and used as the position of the indicator D.

  The image includes, for example, R (red), G (green), and B (blue) color signals, Y (luminance), UV (color difference) signals, H (hue), S (saturation), and B (luminance). Consists of various types of signals. In the image showing the shadow S, the luminance (for example, Y (luminance) and B (luminance) described above) and the color signal (for example, the signals other than the luminance) are likely to be uniform, and the indicator D In an image indicating, the luminance and color signals are unlikely to be uniform.

  Therefore, in the selection method of the third example, the pointer position calculation unit 14 uses a recognition object having a small variation (for example, a dispersion value indicating variation with respect to the average value) of image signal values (for example, luminance and color signals). The candidate position to be obtained is not preferentially selected, but the candidate position obtained by the recognition object having a large variation in the signal value of the image is preferentially selected to be the position of the indicator D.

  Therefore, by performing the selection method of the third example, it is possible to preferentially select a candidate position obtained from a recognized object that is highly likely to be the indicator D and set it as the position of the indicator D. .

  Note that the image for obtaining the variation in the signal value may be a left image and a right image, and an image obtained by processing the left image and the right image (for example, a luminance component from each of the left image and the right image). Each luminance image obtained by extraction) may be used. In any case, the variation in the signal value is obtained from each of the left image and the right image. Further, for example, the left image processing unit 12L and the right image processing unit 12R obtain the variation in the signal value.

  Further, in order to confirm that the variation of the signal value is large or small, it may be confirmed that it is absolutely larger or smaller than a predetermined threshold value. In addition, it may be confirmed that it is relatively large or small as compared with the variation in the signal value of the image showing another recognition target.

  When configured as described above, even if a plurality of recognized objects are recognized from the images generated by the imaging of the left imaging unit 11L and the right imaging unit 11R and a plurality of candidate positions can be obtained, Candidate positions corresponding to the indicator D can be obtained from the positional relationship between the plurality of candidate positions. Therefore, it is possible to detect the position of the indicator with high accuracy even under outdoor conditions.

  In particular, even when the indicator and the shadow of the indicator generated by light irradiation become a recognized object and a plurality of candidate positions can be obtained, candidate positions corresponding to the indicator from the candidate positions Can be obtained. Therefore, it is possible to detect the position of the indicator with high accuracy even under conditions where shadows such as outdoors can occur.

  The selection methods of the first to third examples described above can be implemented in combination. For example, the indicator position calculation unit 14 is configured such that a higher priority is set for a candidate position that is preferentially selected by each selection method, and the candidate position that finally has the highest priority is The position of the indicator D may be used.

Further, the candidate positions are expressed by angles (angle αD, angle αS, angle βD, and angle βS) with respect to the left imaging unit 11L and the right imaging unit 11R, and a candidate position that becomes the position of the indicator D is selected based on the angles. As an example, the candidate position may be expressed by coordinates (x ₁ , y ₁ ), and the candidate position that becomes the position of the indicator D may be selected based on the coordinates.

<Operation of position detection device>
Next, an example of the overall operation of the position detection device 10 when the pointer position calculation unit 14 can perform the selection methods of the first to third examples described above will be described with reference to the drawings. FIG. 10 is a flowchart illustrating an example of the operation of the position detection device.

  As shown in FIG. 10, when the position detection apparatus 10 starts operation, first, the left image processing unit 12L calculates a left difference image that is an image for recognizing the recognition target (STEP 1L). Similarly, the right-side image processing unit 12R calculates a right-side difference image that is an image for recognizing the recognition target (STEP 1R).

  The respective operations for calculating the left difference image and the right difference image of STEP1L and STEP1R will be described with reference to FIG. However, the operation for calculating the right side difference image is the same as the operation for calculating the left side difference image, and thus will be omitted. However, the operation of calculating the right difference image can be described by appropriately replacing “left” and “L” in the following description of the calculation method of the left difference image with “right” and “R”. .

  FIG. 11 is a flowchart illustrating an example of an operation for calculating the left side difference image. As illustrated in FIG. 11, first, the left image processing unit 12L acquires a left image generated by imaging of the left imaging unit 11L (STEP 1L1). At this time, the left image processing unit 12L may convert the type of the signal value of the left image acquired from the left imaging unit 11L as necessary for subsequent processing. In the left image processed by the left image processing unit 12L, for example, one pixel may include signal values for each color of RGB, each signal value for each type of YUV, and each HSB You may provide a kind of signal value.

  The left image processing unit 12L determines whether or not the acquired left image is the first frame (that is, the first left image acquired after the operation of the position detection device 10), or the display in the acquired left image. It is determined whether or not the area 21 has deviated from the display area 21 in the left image acquired previously (for example, in the previous frame) (STEP 1L2).

  When the left image processing unit 12L determines that the acquired left image is the first frame, or when it is determined that the display area 21 in the acquired left image is shifted (STEP 1L2, YES), the left image processing unit 12L displays the left image from the left image. The area 21 is detected (STEP 1L3). At this time, for example, the position of the display area 21 may be confirmed by detecting a predetermined mark or the like, or the position of the display area 21 may be confirmed based on an image (light) displayed in the display area 21. It doesn't matter.

  The left image processing unit 12L sets a range to be processed in the left image (hereinafter referred to as a processing range) based on the detected display area 21 (STEP 1L4). The processing range will be described with reference to FIG. FIG. 12 is a diagram showing an example of a processing range setting method, and corresponds to FIG. 3B showing an example of the left image. FIG. 12 shows an example in which the range on the surface of the display area 21 in the left image is set as the processing range A.

  If the left image processing unit 12L sets the processing range A in STEP1L4, or if it is determined in STEP1L2 that the left image is not in the first frame and the display area 21 is not shifted, then the left image processing unit 12L determines that the left difference An image is calculated (STEP 1L5).

  For example, the left image processing unit 12L obtains a difference value of the signal value for each pixel at a corresponding position between the left background image stored in the left background image storage unit 13L and the left image acquired in the most recent STEP1L1. The left side difference image is calculated using the difference value as the signal value of each pixel. In addition, when each pixel of the left background image and the left image includes a plurality of types of signal values, it is preferable to calculate a difference value for each type of signal value. Further, the difference value may be calculated for all the pixels included in the left background image and the left image, but if the pixel to be calculated is limited to the above processing range A, the amount of calculation can be reduced. preferable.

  By the way, the left background image may be stored in the left background image storage unit 13L in advance before, for example, the operation shown in FIG. For example, the left image processing unit 12L monitors a plurality of consecutively acquired images for a predetermined time, and uses the left background image storage unit 13L as a left background image with a frame having a small signal value variation (difference value). You may memorize. For example, the left image of the first frame may be stored as the left background image. Note that it is preferable to limit the left background image stored in the left background image storage unit 13L to an image within the processing range A because the storage amount can be reduced.

  Further, in this example, it is monitored whether or not the imaging environment (for example, the light intensity in the entire angle of view) has changed so that the recognition target can be accurately detected from the left difference image in STEP 2L described later. Then, an operation for maintaining the left background image at an appropriate level is performed (STEP 1L6 to 1L8).

  First, with respect to the left side difference image calculated in STEP 1L5, it is determined whether or not a predetermined number or more of pixels within the processing range A have a difference value equal to or greater than a threshold value (STEP 1L6). That is, it is determined whether there is a change in the spatial direction of the imaging environment.

  At this time, when each pixel of the left difference image has a difference value of a plurality of types of signal values, it may be compared with each threshold value for each type of signal value, or the difference values of a plurality of types of signal values are added together. You may compare a comprehensive difference value and a threshold value. The “threshold value” is set to a value large enough to assume that the imaging environment has changed. The “predetermined number” is a number that can be regarded as a change in the entire processing range A (that is, not a local change in the image by the recognition target, but a change in the imaging environment).

  When the number of pixels whose difference value is equal to or greater than the threshold is less than the predetermined number (STEP 1L6, NO), it is determined that the imaging environment has not changed. Then, the operation of STEP1L is terminated. On the other hand, when the number of pixels whose difference value is equal to or greater than the threshold value is equal to or greater than the predetermined number (STEP1L6, YES), the frame that satisfies the condition of STEP1L6 (STEP1L6 is YES) is continuously equal to or greater than the predetermined number. It is determined whether or not there is (STEP 1L7). That is, it is determined whether there is a change in the imaging environment in the time direction. Note that the “predetermined number” is a number that allows the imaging environment to be continuously changed and regarded as not transient.

  When the number of frames satisfying the condition of STEP1L6 is less than the predetermined number (STEP1L7, NO), it is determined that the imaging environment has not changed. Then, the operation of STEP1L is terminated. On the other hand, when the number of frames satisfying the condition of STEP1L6 is equal to or greater than the predetermined number (STEP1L7, YES), it is determined that the imaging environment has changed. Then, the left image processing unit 12L updates the left background image stored in the left background image storage unit 13L (STEP 1L8). For example, the left image acquired in the latest STEP1L1 is stored again as the background image. Then, the operation of STEP1L is terminated.

  As illustrated in FIG. 10, when the left image processing unit 12L performs the above-described STEP1L, the left image processing unit 12L recognizes the recognition target from the left difference image obtained in STEP1L (STEP2L). Similarly, when STEP1R is performed, the right image processing unit 12R recognizes the recognition target from the right difference image obtained in STEP1R (STEP2R).

  Each operation for recognizing the recognition target object from the left side difference image and the right side difference image of STEP2L and STEP2R will be described with reference to FIG. However, the operation for recognizing the recognition target from the right side difference image is the same as the operation for recognizing the recognition target from the left side difference image, and is therefore omitted. However, in the following description of the operation for recognizing the recognition target from the left difference image, “left” and “L” are appropriately replaced with “right” and “R” to recognize the recognition target from the right difference image. It can be an explanation of the operation to be performed.

  FIG. 13 is a flowchart illustrating an example of an operation for recognizing the recognition target object from the left side difference image. As illustrated in FIG. 13, the left image processing unit 12L displays the pixel of the left difference image calculated in STEP 1L as a pixel that displays a recognized object (hereinafter referred to as a recognized object display pixel) and a background. It is divided into pixels (hereinafter referred to as background display pixels). That is, the left difference image is binarized (STEP2L1).

  At this time, for example, by the same method as in STEP1L6, a pixel whose difference value is a certain threshold value or more is set as a recognition object display pixel, and a pixel whose difference value is smaller than the threshold value is set as a background display pixel. The “threshold value” is set to a value large enough to accurately determine that the indicator is the indicator D. Further, the left image processing unit 12L may generate an image (so-called binarized image) in which the recognition target display pixel and the background display pixel of the left background image are expressed by different signal values.

  Further, the left image processing unit 12L recognizes an area (that is, a recognized object) composed of a predetermined number or more of recognized object display pixels within the detection range of the left difference image (including the binarized image). (STEP2L2). The detection range is preferably at least within the processing range. For example, the detection range may be a range in which it can be determined that the indicator D has touched the display area 21. Further, the “predetermined number” is a number that can be accurately determined as the indicator D, and is, for example, a number that can eliminate a pixel-unit variation caused by noise or the like.

  If the number of regions (recognized bodies) detected in STEP2L2 is 0 (STEP2L3, YES), it is determined that there are no recognized bodies in the left difference image (STEP2L4). Then, the operation of STEP2L is terminated.

  On the other hand, if the number of recognized objects is not 0 (STEP2L3, NO), it is determined that the recognized object exists in the left difference image (STEP2L5). And it is confirmed whether the number of to-be-recognized bodies is 2 or less (STEP2L6).

  When the number of recognition objects is larger than 2 (STEP2L6, NO), the left image processing unit 12L determines that the recognition object recognized from the left difference image is inappropriate (STEP2L7). Then, the operation of STEP2L is terminated.

  On the other hand, if the number of recognition objects is 2 or less (STEP2L6, YES), it is determined that the recognition object recognized by the left image processing unit 12L from the left difference image is appropriate (STEP2L8). Then, as described in the third example of the candidate position selection method for each recognized object in the left image, the left image processing unit 12L varies the signal value of the image indicating the recognized object (for example, (Dispersion) is calculated (STEP2L9). Then, the operation of STEP2L is terminated.

  As shown in FIG. 10, when the left image processing unit 12L and the right image processing unit 12R perform the above-described STEP2L and STEP2R, it is next determined whether or not a recognition object exists in STEP2L and STEP2R. Is confirmed (STEP 3).

  Here, if at least one of the left image processing unit 12L and the right image processing unit 12R determines that the recognition target is not present (see STEP2L4) (STEP3, NO), the process returns to STEP1L and STEP1R, and the left side The difference image and the right difference image are calculated.

  On the other hand, if both the left image processing unit 12L and the right image processing unit 12R determine that a recognition target exists (see STEP 2L5) (STEP 3, YES), then STEP 2L and STEP 2R In step 4, it is confirmed whether or not it is determined that the recognition target is appropriate.

  Here, if at least one of the left image processing unit 12L and the right image processing unit 12R determines that the recognition object is inappropriate (see STEP2L7) (STEP4, NO), error processing is performed (STEP5). . Any processing may be performed as error processing. For example, the operation may be terminated (STEP 8, YES), a warning indicating that the user cannot be detected may be displayed, and the left background image and the right background image are updated in STEP 1L and STEP 1R ( (Refer to STEP1L8). Moreover, when the determination that the recognition target is inappropriate is continuously performed, the error processing as described above may be performed.

  On the other hand, if both the left image processing unit 12L and the right image processing unit 12R determine that the recognition target is appropriate (see STEP 2L8) (STEP 4, YES), then the pointer position calculation is performed. The unit 14 recognizes the position of the indicator D based on the recognition target recognized by the left image processing unit 12L and the right image processing unit 12R (STEP 6). Specifically, for example, the direction (angle α and angle β, specifically, angle αD and angle βD) in which the indicator D exists is recognized with respect to the left imaging unit 11L and the right imaging unit 11R.

  The operation in which the indicator position calculation unit 14 of STEP 6 recognizes the position of the indicator will be described with reference to FIG. FIG. 14 is a flowchart illustrating an example of an operation for recognizing the position of the indicator. As shown in FIG. 14, the indicator position calculation unit 14 includes one to-be-recognized object recognized by the left image processing unit 12 </ b> L and the right image processing unit 12 </ b> R, for example, as illustrated in FIGS. 3, 4, and 9. If there is one (STEP 601, YES), the position obtained from the recognized object is recognized as the position of the indicator (STEP 602). Then, the operation of STEP 6 is terminated.

On the other hand, the indicator position calculation unit 14 does not recognize the recognized bodies recognized by the left image processing unit 12L and the right image processing unit 12R as shown in FIGS. 5 and 7, for example (STEP601, NO ) If there are two (STEP 603, YES), the candidate position obtained from each recognition object is recognized (STEP 604). At this time, in order to determine whether or not the candidate position is located outside the range of the display area 21, for example, the coordinates (x ₁ , y ₁ ) of the candidate position are obtained.

  Then, for example, as shown in FIG. 7, when there is a candidate position located outside the range of the display area 21 (STEP 605, YES), as described in the second example of the candidate position selection method, the candidate position of the false candidate Are excluded from the position of the indicator D (STEP 606). Then, the next operation (STEP 607) is performed. On the other hand, when there is no candidate position located outside the display area 21 (STEP 605, NO), the next operation (STEP 607) is performed without performing the operation of STEP 606.

  Further, for example, as shown in FIGS. 6 and 8, the recognition target recognized by the left image processing unit 12L and the right image processing unit 12R is not one by one (STEP 601, NO), but not by two (STEP 603). , NO), the next operation (STEP 607) is performed without performing the operations of STEP 604 to STEP 606.

  In STEP 607, as described in the first and third examples of the candidate position selection method, the candidate position is the upper candidate position, and the candidate position obtained from the recognition target whose signal value variation is equal to or larger than a predetermined size is determined. The position is selected and recognized as the position of the indicator D. If there is no candidate position that satisfies both of the above conditions, the above error processing (see STEP 5) may be performed, and a candidate position that satisfies one of the conditions (for example, the upper side) is selected as the position of the indicator D. It may be recognized, or the candidate position having the highest priority set by each selection method may be selected and recognized as the position of the indicator D.

  If two candidate positions are obtained (YES in STEP 608), the candidate position that is not the candidate position selected as the position of the indicator D is recognized as the position of the shadow S (STEP 609). Then, the operation of STEP 6 is terminated.

  On the other hand, when the number of candidate positions to be obtained is not two (STEP 608, NO), as described above, the candidate position located at the diagonal of the candidate position selected as the position of the indicator D is the position of the shadow S. (STEP 610). Then, the operation of STEP 6 is terminated. If the position of the shadow S cannot be obtained or if it is not necessary to recognize the position of the shadow S, only the position of the indicator D is recognized in STEP 607 and the position of the shadow S may not be recognized. .

As shown in FIG. 10, when the indicator position calculation unit 14 performs the operation of STEP 6 described above, the coordinates (x of the indicator D as described above are then used using the position of the indicator D recognized in STEP 6. ₁ , y ₁ ) is calculated to detect the position of the indicator D on the surface of the display area 21 (STEP 7). Then, the indicator position calculation unit 14 outputs the calculated position of the indicator D to the display device 20 as indicator position information.

  Here, when the operation of the position detection device 10 is to be ended (STEP 8, YES), the operation is ended. On the other hand, when the operation is continued without ending (STEP 8, NO), the process returns to STEP 1L and STEP 1R to calculate the left side difference image and the right side difference image.

  If comprised as mentioned above, it will become possible to detect the position of the indicator D on the surface of the display area 21 by recognizing the position of the indicator D accurately according to the situation. In particular, the position of the indicator can be detected with high accuracy even under outdoor conditions.

  FIG. 14 shows an operation example in which candidate positions corresponding to the indicator D are selected by sequentially performing the first to third examples of the candidate position selection method, but some or all of the candidates are selected. The above-described priority may be set by applying each selection method to the position, and the candidate position where the priority finally becomes the largest may be selected as the position of the indicator D.

  Further, the definition of the angle α and the angle β shown in FIGS. 3 and 4 and the image capturing method (for example, a lens to be used) by the left imaging unit 11L and the right imaging unit 11R are merely examples, and the display area 21 Any other definition and imaging method may be applied as long as the position of the indicator D on the surface can be detected. For example, an imaging method in which an image in which the edge of the display area 21 is curved may be applied.

  Further, the left image processing unit 12L and the right image processing unit 12R have been described as performing processing on a partial range (processing range, detection range) of the left image and the right image. The imaging unit 11R may perform imaging in which only an image in the range is generated.

  Moreover, although illustrated about the structure provided with two imaging parts, 11L of left side imaging parts, and 11R of right side imaging parts, it is good also as a structure provided with three or more imaging parts. The same applies to the image processing unit and the background image storage unit.

  Further, the left image processing unit 12L, the right image processing unit 12R, the left background image storage unit 13L, and the right background image storage unit 13R are displayed as separate bodies on the block diagram shown in FIG. It does not matter.

  Further, the left image processing unit 12L calculates a left difference image from the left image and the left background image, and the right image processing unit 12R calculates a right difference image from the right image and the right background image, whereby the left image and the right side image are calculated. Although the method for recognizing the recognition object from each of the images has been illustrated, the recognition method for the recognition object is not limited to this. For example, at least two left images generated by continuous imaging are compared (for example, the difference image as described above is calculated), and similarly, at least two right images are compared, thereby recognizing the recognition target. It doesn't matter. Further, the recognition target may be recognized directly (without generating the left difference image and the right difference image) based on the signal values of the left image and the right image.

  In addition, when the processing range A of STEP1L4 in FIG. 11 is set for at least one of the left image and the right image, or the background image of STEP1L8 is updated for at least one of the left background image and the right background image. The left image processing unit 12L reacquires the left image newly captured by the left imaging unit 11L, and the right image processing unit 12R acquires the right image newly captured by the right imaging unit 11R. You can do it again. Further, when one of the left image processing unit 12L and the right image processing unit 12R sets the processing range A of STEP1L4 or updates the left background image or the right background image of STEP1L8, the other performs the same operation. It doesn't matter.

  The frame body 22 may be omitted. However, from the viewpoint of suppressing the fluctuation of the imaging environment on the surface of the display area 21 and recognizing the indicator accurately from the left image and the right image, the frame body. A configuration including 22 is preferable. Further, the frame body 22 may be provided at any position as long as it is around the display area 21, and may have any color. However, a color with little light reflection (for example, black or gray) or a color opposite to the indicator D (that is, a complementary color, for example, the frame 22 and the indicator D are displayed in the left image and the right image, respectively. It is preferable to use a color in which a difference value of signal values such as RGB, YUV, and HSB of the pixel is large, because the indicator is more easily recognized. In addition, it is preferable to change the color of the frame body 22 to a color with little light reflection (for example, black or gray), because it is possible to further suppress fluctuations in the imaging environment.

<Modification>
The position detection apparatus 10 described above can recognize the position by distinguishing the indicator and the disturbance (for example, a shadow). For this reason, by applying the indicator detection method of the position detection device 10 described above to a position detection device that can detect the positions of a plurality of indicators (for example, two fingers) on the detection surface, the accuracy can be improved. It becomes possible to detect the positions of a plurality of indicators.

  The position detection device 10 according to the embodiment of the present invention may be partially or wholly operated by a control device such as a microcomputer. Further, all or part of the functions realized by such a control device is described as a program, and the program is executed on a program execution device (for example, a computer) to realize all or part of the functions. It doesn't matter if you do.

  In addition to the above-described case, the touch panel 1 shown in FIG. 1 and the position detection device 10 shown in FIG. 2 can be realized by hardware or a combination of hardware and software. Moreover, when comprising a part of touch panel 1 or the position detection apparatus 10 using software, the block about the site | part implement | achieved by software shall represent the functional block of the site | part.

  As mentioned above, although embodiment in this invention was described, the range of this invention is not limited to this, It can add and implement various changes in the range which does not deviate from the main point of invention.

  INDUSTRIAL APPLICABILITY The present invention can be used for a position detection device that detects the position of an indicator on a detection surface and a touch panel that includes the position detection device.

DESCRIPTION OF SYMBOLS 1 Touch panel 10 Position detection apparatus 11L Left side imaging part 11R Right side imaging part 12L Left side image processing part 12R Right side image processing part 13L Left side background image memory | storage part 13R Right side background image memory | storage part 14 Pointer position calculation part 20 Display apparatus 21 Display area 22 Frame Body D Indicator S Shadow

Claims (6)

A position detection device for detecting the position of an indicator on a detection surface,
Two or more imaging units that generate images by imaging the detection surface from different locations;
A position detection unit that detects the position of the indicator on the detection surface by recognizing the indicator from an image generated by imaging of the imaging unit;
A plurality of recognized objects including the indicator can be recognized from among images generated by imaging of the at least one imaging unit, and a plurality of candidate positions including the position of the indicator on the detection surface can be obtained. When
The position detection unit detects a position of the indicator on the detection surface based on a positional relationship between the plurality of candidate positions.   The recognized object recognized by the position detection unit from an image generated by imaging of the imaging unit can include the indicator and a shadow of the indicator. Position detection device. The imaging unit is disposed on one end side of the detection surface,
The position detection unit preferentially selects a candidate position on the edge side of the detection surface and sets it as the position of the indicator on the detection surface. The position detection device described. The position detection unit is
Candidate positions located outside the range of the detection surface;
In each image obtained by imaging by the imaging unit, a candidate position obtained by a recognized object different from a recognized object for obtaining the candidate position;
4. The position detection device according to claim 1, wherein other candidate positions excluding the first position are preferentially selected and set as the position of the indicator on the detection surface. 5.   2. The position detection unit preferentially selects a candidate position obtained from a recognition object having a large variation in image signal values, and sets the position as the position of the indicator on the detection surface. The position detection device according to claim 4. A display device for displaying an image in a display area;
The position detection device according to any one of claims 1 to 5, wherein the display area is a detection surface;
A touch panel comprising:

Images