JP2015127976A - Recognition device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve operability relating to switchover of modes of an operation target device in response to a gesture.SOLUTION: A recognition device comprises an acquisition unit, a calculation unit, a determination unit, and a control unit. The acquisition unit acquires positions of a specific region of an object in time series. The calculation unit calculates a feature quantity of a motion of the specific region from the positions acquired in time series. The determination unit determines a magnitude relation between the calculated feature quantity and a first threshold and a magnitude relation between the calculated feature quantity and a second threshold smaller than the first threshold. The control unit controls display processing for a notification that a recognition method for the position or motion of the specific region is possibly switched if the feature quantity is not less than the second threshold and less than the first threshold.

Description

  Embodiments described herein relate generally to a recognition apparatus.

  2. Description of the Related Art Conventionally, a recognition device that recognizes a gesture that is a predetermined action of a specific part such as a user's hand or finger is known. Such a recognition device recognizes a gesture by a user and executes an operation of an operation target device that is an operation target. An example of the operation target device is an information processing apparatus such as a PC (Personal Computer). As a technique using a recognition apparatus, there is a technique that moves a mouse cursor on a PC screen in accordance with a gesture by a user. As another technique using the recognition device, there is a technique for executing various commands such as switching a PC screen according to a gesture by a user.

JP 2010-182014 A JP 2005-352531 A

  However, the conventional technology has a problem that the user may select the operation mode of the operation target device according to the gesture in advance. For example, in the case where a mode for moving the mouse cursor and a mode for executing a command are installed, since the same or similar gesture exists in these modes, the user determines which mode to use. Must be selected in advance.

  When realizing an operation according to a gesture, it is preferable to employ a gesture that is easy for the user to operate intuitively. For this reason, for example, a gesture for moving the mouse cursor downward and a gesture for moving the screen downward may both be movements of moving the hand downward. As a result, in the related art, the user may select the operation mode of the operation target device according to the gesture in advance. In addition, although it is conceivable to switch the mode by a gesture, in such a case, a gesture for switching the mode is performed, which is not preferable because it is complicated for the user.

  Problem to be solved by the invention is providing the recognition apparatus which can improve the operativity which concerns on switching of the mode of operation of the operation target apparatus according to gesture.

  The recognition apparatus according to the embodiment includes an acquisition unit, a calculation unit, a determination unit, and a control unit. The acquisition unit acquires the position of the specific part of the subject in time series. The calculation unit calculates the feature amount of the movement of the specific part from the position acquired in time series. The determination unit determines a magnitude relationship between the calculated feature value and the first threshold value, and a magnitude relationship between the calculated feature value and a second threshold value smaller than the first threshold value. The control unit controls display processing for notifying that there is a possibility that the position or movement recognition method of the specific part may be switched when the feature amount is equal to or greater than the second threshold value and less than the first threshold value.

The block diagram which shows the recognition apparatus which concerns on 1st Embodiment. The figure which shows the information processing apparatus to which a recognition apparatus is applied. The figure which shows the position of the specific site | part acquired in time series. The figure which shows the screen display in 1st mode. The figure which shows the screen display in 2nd mode. The flowchart which shows the whole process which concerns on 1st Embodiment. The block diagram which shows the recognition apparatus which concerns on the modification 1 of 1st Embodiment. The figure which shows the control according to the feature-value which concerns on the modification 1 of 1st Embodiment. The figure which shows the provisional display of a cursor. The flowchart which shows the whole process which concerns on the modification 1 of 1st Embodiment. The block diagram which shows the recognition apparatus which concerns on the modification 2 of 1st Embodiment. The figure which shows the control according to the feature-value which concerns on the modification 2 of 1st Embodiment. The figure which shows the control according to the feature-value which concerns on the modification 2 of 1st Embodiment. The figure which shows the control according to the feature-value which concerns on the modification 2 of 1st Embodiment. The figure which shows the control according to the feature-value which concerns on the modification 2 of 1st Embodiment. The figure which shows the control according to the feature-value which concerns on the modification 2 of 1st Embodiment. The figure which shows the control according to the feature-value which concerns on the modification 2 of 1st Embodiment. The figure which shows the control according to the feature-value which concerns on the modification 2 of 1st Embodiment. The block diagram which shows the recognition apparatus which concerns on the modification 3 of 1st Embodiment. 9 is a flowchart showing overall processing according to Modification 3 of the first embodiment. The block diagram which shows the recognition apparatus which concerns on the modification 4 of 1st Embodiment. The flowchart which shows the whole process which concerns on the modification 4 of 1st Embodiment. The block diagram which shows the recognition apparatus which concerns on the modification 5 of 1st Embodiment. The flowchart which shows the whole process which concerns on the modification 5 of 1st Embodiment. The block diagram which shows the recognition apparatus which concerns on 2nd Embodiment. The figure explaining the similarity of gesture. The figure which shows the screen display of the 1st mode which concerns on 2nd Embodiment. The figure which shows the screen display of the 2nd mode which concerns on 2nd Embodiment. The flowchart which shows the whole process which concerns on 2nd Embodiment. The figure which shows the alerting | reporting to the user which concerns on mode transfer. The figure which shows the alerting | reporting to the user which concerns on mode transfer. The block diagram which shows the recognition apparatus which added the identification part. The block diagram which shows the recognition apparatus which added the shape recognition part. The figure which applies a recognition apparatus to a television receiver. The figure which applies a recognition apparatus to the information terminal for medical treatment. The figure by which a program is implement | achieved using a computer.

(First embodiment)
FIG. 1 is a block diagram illustrating a configuration example of a recognition apparatus according to the first embodiment. As illustrated in FIG. 1, the recognition apparatus 100 includes an acquisition unit 110, a calculation unit 120, a determination unit 130, and a control unit 140. For example, the recognition apparatus 100 is applied to an information processing apparatus such as a PC (Personal Computer) equipped with an image sensor such as a camera and an image sensor, recognizes a user's gesture, and operates the PC according to the recognized gesture. To control. The camera captures time-series images. The recognition apparatus 100 is connected to the display unit 1 included in the PC.

  FIG. 2 is a diagram illustrating an example of an information processing apparatus to which the recognition apparatus is applied. The recognition apparatus of any of the following embodiments is applied to the information processing apparatus 10 illustrated in FIG. For example, the information processing apparatus 10 includes a camera 11 and images a user who is a subject. The user performs a gesture with the hand 14 in order to browse the web content or the document displayed in the window 13 of the screen 12 of the information processing apparatus 10. Thereby, the scroll bar 15 moves according to the gesture, and the window 13 scrolls. Alternatively, the user performs a gesture with the hand 14 in order to move the cursor displayed on the screen 12 (display unit 1) of the information processing apparatus 10. As a result, the cursor moves according to the gesture.

  The acquisition unit 110 detects a specific part such as a user's hand or finger, which is a subject, for each image with respect to the time-series image captured by the camera 11, and tracks the time-series to thereby identify the user. The position of the part is acquired in time series. The position of the specific part to be acquired is, for example, the center of gravity of the detection region of the specific part that is a hand or a finger obtained from the image sensor. In addition, the position of the specific part to be acquired may be expressed by coordinates such as a pixel position specified on the captured image, or may be expressed by relative coordinates from the position of the detected part such as the user's face. Also good.

  Further, not only the position but also the time when each position is acquired may be stored. The time is expressed by the elapsed time from the start, the number of elapsed clocks, the number of elapsed frames, etc., when the recognition apparatus 100 is started or the position acquisition start time is “0”. For example, when the position is expressed by (x, y) coordinates on the captured image, the position (xi, yi) acquired in the i-th frame is stored in association with the time “Ti” at the time of acquisition.

FIG. 3 is a diagram illustrating an example of the position of the specific part acquired in time series. As shown in FIG. 3, the movement of the specific part is “P ₁ ”, “P ₂ ”, “P ₃ ”, “P ₄ ”, “P ₅ ”, “P ₆ ”, “P ₇ ”, “P ₈ ”followed by a continuous trajectory as indicated by the dashed arrows. The acquisition unit 110 acquires the positions of “P ₁ ” to “P ₈ ” as discrete points sampled by the image sensor. Note that “P ₃ ” and “P ₆ ” illustrated in FIG. 3 may be referred to as “folding points”.

The calculation unit 120 calculates the feature amount of the movement of a specific part such as a user's hand or finger as a subject from the position acquired by the acquisition unit 110. Here, the moving speed of the hand will be described as an example of the feature amount. For example, the calculation unit 120 sets “P ₈ ” as the latest position acquired at time “T ₈ ” for the position acquired by the acquisition unit 110, and sets the previous position acquired at time “T ₇ ”. When “P ₇ ” is assumed, the moving speed “V” of the hand is calculated by “the distance from | P ₈ to P ₇ / (T ₈ −T ₇ ) |”.

  The determining unit 130 determines the magnitude relationship between the hand moving speed calculated by the calculating unit 120 and the first threshold value representing a specific value of the moving speed, using the hand moving speed calculated by the calculating unit 120 as a feature amount. Then, the determination unit 130 outputs a determination result of the magnitude relationship between the hand movement speed and the first threshold value to the control unit 140. The first threshold is set to a first mode that uses a method for recognizing the position or movement of a specific part such as a hand or a finger when performing control according to a user's gesture, or what is the first mode? This is used to determine whether to enter the second mode using a different recognition method. That is, in one aspect of this embodiment, the first mode or the second mode is selected according to the moving speed of the hand.

  The control unit 140 receives the determination result by the determination unit 130, selects the first mode when the moving speed of the hand is less than the first threshold value, and selects the second mode when the moving speed of the hand is equal to or higher than the first threshold value. Select a mode. In addition, the control unit 140 controls display processing of a screen displayed on the display unit 1 in accordance with the user's gesture in each of the selected first mode or second mode. For example, the control unit 140 controls the movement of the cursor displayed on the display unit 1 when in the first mode, or controls the scrolling of the screen displayed on the display unit 1 when in the second mode. To do. Here, the movement of the cursor in the first mode refers to the movement of the cursor by pointing.

  FIG. 4 is a diagram illustrating an example of screen display in the first mode. In FIG. 4, as an example of the control in the first mode, a case where the cursor on the screen is moved is taken as an example. In the state of the window 13 shown in the upper part of FIG. 4, when the user performs a gesture for moving the hand downward, the cursor 16 moves downward as in the window 13 shown in the lower part of FIG. Note that the scroll bar 15 does not move in the first mode.

  FIG. 5 is a diagram illustrating an example of screen display in the second mode. In FIG. 5, as an example of the control in the second mode, a case where the scroll bar on the screen is moved is taken as an example. In the state of the window 13 shown in the upper part of FIG. 5, when the user performs a gesture for moving the hand downward, the scroll bar 15 moves downward as in the window 13 shown in the lower part of FIG. 5. The window 13 scrolls downward. FIG. 5 shows an example in which the block arrow 17 is displayed according to the moving direction of the scroll bar 15. Note that the cursor 16 does not move in the second mode.

  Next, the flow of overall processing according to the first embodiment will be described with reference to FIG. FIG. 6 is a flowchart illustrating an example of the flow of overall processing according to the first embodiment.

  As illustrated in FIG. 6, when the position of the user's hand is acquired by the acquisition unit 110 (Yes in step S101), the calculation unit 120 calculates the moving speed of the hand obtained from the hand movement (step S102). . Further, when the position of the user's hand is not acquired by the acquisition unit 110 (No in step S101), the acquisition unit 110 is in a state of waiting for acquisition of the position.

  In addition, the determination unit 130 compares the magnitude of the hand movement speed calculated by the calculation unit 120 with the first threshold value (step S103). At this time, when the determination unit 130 determines that the moving speed of the hand is less than the first threshold (Yes at Step S104), the control unit 140 selects the first mode (Step S105). On the other hand, when the determination unit 130 determines that the moving speed of the hand is equal to or higher than the first threshold (No at Step S104), the control unit 140 selects the second mode (Step S106).

  According to this embodiment, since the mode of gesture recognition related to the control of the display process of the screen displayed on the display unit is selected according to the moving speed of the specific part in the gesture by the user, the operation target device corresponding to the gesture The operability related to the switching of the operation mode can be improved. That is, according to the present embodiment, it is not necessary for the user to perform an operation for switching the operation mode of the operation target device, so that the operability can be improved.

(Modification 1 of the first embodiment)
FIG. 7 is a block diagram illustrating a configuration example of the recognition device according to the first modification of the first embodiment. In the first modification of the first embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed description of the same components may be omitted. In the first modification of the first embodiment, the functions, configurations, and processes of the units other than the determination unit 130a and the control unit 140a described below are the same as those in the first embodiment.

  As illustrated in FIG. 7, the recognition apparatus 100a includes an acquisition unit 110, a calculation unit 120, a determination unit 130a, and a control unit 140a. In addition, the recognition device 100a is connected to the display unit 1 as in the first embodiment.

  The determination unit 130a uses the moving speed of the hand calculated by the calculating unit 120 as a feature amount, and determines a magnitude relationship between the moving speed of the hand and the first threshold, and further, a value smaller than the moving speed of the hand and the first threshold. The magnitude relationship with the second threshold value to be represented is determined. Then, the determination unit 130a outputs the determination result of the magnitude relationship between the hand movement speed and the first threshold value and the determination result of the magnitude relationship between the hand movement speed and the second threshold value to the control unit 140a. The second threshold value is used to determine whether or not the movement speed of the hand in the previous stage of shifting from the first mode to the second mode corresponds to the case where the control according to the user's gesture is executed. . The stage before the transition from the first mode to the second mode is a case where the condition “second threshold ≦ moving speed of the hand <first threshold” is satisfied. That is, according to one aspect of the present embodiment, there is a possibility that the first mode may be changed to the second mode, the second mode is changed to the first mode, etc. The user can be notified.

  The control unit 140a receives the determination result by the determination unit 130a, selects the first mode when the moving speed of the hand is less than the first threshold (including the case where the moving speed is less than the second threshold), and moves the moving speed of the hand. The second mode is selected when is greater than or equal to the first threshold. In addition, the control unit 140a controls display processing of the screen displayed on the display unit 1 according to the user's gesture in each of the selected first mode or second mode, and the moving speed of the hand is equal to or higher than the second threshold value and the second mode. When it is less than one threshold, display processing of the screen displayed on the display unit 1 is controlled according to the moving speed of the hand. In the control of the display process of the screen displayed on the display unit 1 when the moving speed of the hand is equal to or higher than the second threshold value and lower than the first threshold value, for example, the cursor is displayed to the user as a temporary display that is a display different from the first mode. Inform.

  FIG. 8 is a diagram illustrating an example of control according to the feature amount according to the first modification of the first embodiment. In FIG. 8, “current state” representing the current magnitude relationship between the hand movement speed, which is a feature amount, and the first threshold value and the second threshold value, and “control” representing control according to the current state, respectively. It is shown. Among these, when the current state is “the moving speed of the hand <the second threshold value”, the recognition method and the display process correspond to the “first mode”. When the current state is “second threshold ≦ hand movement speed <first threshold”, the recognition method corresponds to “first mode”, and the display process is “provisional display”. When the current state is “first threshold ≦ hand movement speed”, the recognition method and the display process are in accordance with the “second mode”.

  FIG. 9 is a diagram illustrating an example of temporary display of the cursor. As shown in FIG. 9, when the cursor display corresponding to the first mode is the cursor 20, the provisional display includes a cursor 21 that is a cursor display whose transparency has been changed, a cursor 22 that is a cursor display whose brightness has been changed, There is a cursor 23 or the like which is a cursor display whose height has been changed. In addition to these, the saturation, the blinking speed, the moving speed of the cursor (including the stop of the movement) may be changed, or the pop-out amount may be changed if three-dimensional display is possible.

  Further, the transparency, brightness, size, etc. may be set to a predetermined ratio such as 50% transparency, or may be changed according to the moving speed of the hand as the feature amount. For example, the transparency is changed to 100% when the moving speed of the hand is the first threshold, and the transparency is changed to 0% when the moving speed of the hand is the second threshold. At this time, if the feature amount is “E”, the first threshold is “E1”, and the second threshold is “E2”, the transparency is “100 × (E−E2) / (E1−E2)”.

  Next, the flow of overall processing according to Modification 1 of the first embodiment will be described with reference to FIG. FIG. 10 is a flowchart illustrating an example of the flow of overall processing according to the first modification of the first embodiment.

  As illustrated in FIG. 10, when the position of the user's hand is acquired by the acquisition unit 110 (Yes in step S201), the calculation unit 120 calculates the moving speed of the hand obtained from the hand movement (step S202). . Further, when the position of the user's hand is not acquired by the acquisition unit 110 (No in step S201), the acquisition unit 110 is in a state of waiting for acquisition of the position.

  In addition, the determination unit 130a compares the magnitude relationship between the hand movement speed calculated by the calculation unit 120 and the first threshold value, and the magnitude relation between the hand movement speed and the second threshold value (step S203). At this time, when the determination unit 130a determines that the moving speed of the hand is less than the second threshold value (Yes at Step S204, Yes at Step S205), the control unit 140a selects the first mode (Step S206).

  Further, when the determination unit 130a determines that the moving speed of the hand is equal to or higher than the second threshold and lower than the first threshold (Yes at Step S204, No at Step S205), the control unit 140a selects the first mode, Screen display processing is controlled in accordance with the moving speed of the hand (step S207). Further, when the determination unit 130a determines that the moving speed of the hand is equal to or higher than the first threshold (No at Step S204), the control unit 140a selects the second mode (Step S208).

  According to the present embodiment, since the display of the cursor or the like is provisionally displayed before the transition from the first mode to the second mode, there is a possibility of transition from the first mode to the second mode, It is possible to make the user recognize that the mode has shifted to the first mode. Further, according to the present embodiment, it is not necessary for the user to perform an operation for switching the operation mode of the operation target device, so that the operability can be improved.

(Modification 2 of the first embodiment)
FIG. 11 is a block diagram illustrating a configuration example of the recognition device according to the second modification of the first embodiment. In the second modification of the first embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed description of the same components may be omitted. In the second modification of the first embodiment, the functions, configurations, and processes of the respective units other than the determination unit 130b and the control unit 140b described below are the same as those in the first embodiment.

  As illustrated in FIG. 11, the recognition apparatus 100b includes an acquisition unit 110, a calculation unit 120, a determination unit 130b, and a control unit 140b. In addition, the recognition device 100b is connected to the display unit 1 as in the first embodiment. In the second modification of the first embodiment, the “current state” representing the current magnitude relationship between the moving speed of the hand, which is the feature amount, and the first threshold value and the second threshold value, and the immediately preceding mode is the first mode. Alternatively, the recognition method and the display process are changed according to the “previous mode” indicating whether the mode is the second mode. Hereinafter, a recognition method and display processing in each pattern will be described with reference to FIGS. 12 to 18 are diagrams illustrating an example of a recognition method and display processing in the relationship between the current state and the immediately preceding mode according to the second modification of the first embodiment.

  In FIG. 12, the determination unit 130b determines that the moving speed of the hand is less than the second threshold value. As a result, when the immediately preceding mode is the first mode, the control unit 140b controls the recognition method and the display process of the screen displayed on the display unit 1 as corresponding to the “first mode”. In addition, when the previous mode is the second mode, the control unit 140b controls the recognition method and the display processing of the screen displayed on the display unit 1 as corresponding to the “first mode”.

  In FIG. 12, the determination unit 130b determines that the moving speed of the hand is equal to or higher than the second threshold and lower than the first threshold. As a result, when the immediately preceding mode is the first mode, the control unit 140b controls the recognition method as that corresponding to the “first mode”, and the display processing of the screen displayed on the display unit 1 is “provisional display”. To control. In addition, when the previous mode is the second mode, the control unit 140b controls the recognition method and the display processing of the screen displayed on the display unit 1 as corresponding to the “first mode”.

  In FIG. 12, the determination unit 130b determines that the moving speed of the hand is equal to or higher than the first threshold value. As a result, when the previous mode is the first mode, the control unit 140b controls the recognition method and the display process of the screen displayed on the display unit 1 according to the “second mode”. In addition, when the immediately preceding mode is the second mode, the control unit 140b controls the recognition method and the display process of the screen displayed on the display unit 1 according to the “second mode”.

  In FIG. 13, the determination unit 130b and the control unit 140b perform the same processing as the processing described in FIG. In addition to this, the calculation unit 120 calculates the movement distance of the hand, which is a specific part of the user, as the feature amount. Then, in the state where the second mode is selected, the control unit 140b recognizes the recognition method and the display unit 1 when the movement distance of the hand calculated by the calculation unit 120 is continuously within a predetermined range for a predetermined time or more. The display processing of the screen to be displayed is controlled according to the “first mode”. The case where the moving distance of the hand is continuously within a predetermined range for a predetermined time or more means that the hand stillness is detected. In other words, since the user's hand may slightly move and move, if the movement distance of the hand continues for a predetermined time or longer and falls within a predetermined range, it is detected that the hand is stationary.

  Further, in FIG. 14, the determination unit 130b determines that the moving speed of the hand is less than the second threshold value. As a result, when the immediately preceding mode is the first mode, the control unit 140b controls the recognition method and the display process of the screen displayed on the display unit 1 as corresponding to the “first mode”. In addition, when the second mode is selected, the control unit 140b recognizes when the state of being less than the first threshold (including less than the second threshold) continues for a predetermined time or longer in the determination by the determination unit 130b. And the display processing of the screen displayed on the display unit 1 is controlled according to the “first mode”. In order to notify the user that there is a possibility of shifting to the first mode when a certain percentage of the predetermined time has elapsed, a temporary display may be performed.

  In FIG. 14, the determination unit 130b determines that the moving speed of the hand is equal to or higher than the second threshold and lower than the first threshold. As a result, when the immediately preceding mode is the first mode, the control unit 140b controls the recognition method as that corresponding to the “first mode”, and the display processing of the screen displayed on the display unit 1 is “provisional display”. To control. In addition, when the second mode is selected, the control unit 140b displays a screen displayed on the recognition method and the display unit 1 when the determination by the determination unit 130b continues to be less than the first threshold for a predetermined time or more. Is controlled according to the “first mode”. In order to notify the user that there is a possibility of shifting to the first mode when a certain percentage of the predetermined time has elapsed, a temporary display may be performed.

  In FIG. 14, when the moving speed of the hand is equal to or higher than the first threshold, the determination unit 130 b and the control unit 140 b perform the same processing as the processing described in FIG. 12. In the state where the first mode is selected, if the determination by the determination unit 130b continues for a predetermined time or longer in the determination by the determination unit 130b, the recognition method and the display processing of the screen displayed on the display unit 1 are “ You may make it control as what respond | corresponds to "2nd mode".

  In FIG. 15, the determination unit 130 b determines that the value is equal to or greater than the first threshold value. Thereby, in the state where the first mode is selected, the control unit 140b changes the recognition method to “first” when the duration of the state that is equal to or greater than the first threshold in the determination by the determination unit 130b is less than the predetermined threshold. The display process of the screen displayed on the display unit 1 is controlled as “provisional display”. On the other hand, in the state where the first mode is selected, the control unit 140b displays the recognition method and the display unit 1 when the duration of the state that is equal to or greater than the first threshold in the determination by the determination unit 130b is equal to or greater than the predetermined threshold. The display process of the screen to be displayed is controlled according to the “second mode”. In addition, when the immediately preceding mode is the second mode, the control unit 140b controls the recognition method and the display process of the screen displayed on the display unit 1 according to the “second mode”. In FIG. 15, when the moving speed of the hand is less than the first threshold (including less than the second threshold), the determination unit 130b and the control unit 140b perform the same processing as the processing described in FIG. According to the example of FIG. 15, even when noise is included in the value of the moving speed of the hand as the feature amount for only one frame, it is possible to realize control that suppresses the influence of the noise.

  In FIG. 16, the determination unit 130b determines that the moving speed of the hand is equal to or higher than the second threshold and lower than the first threshold. As a result, when the immediately preceding mode is the first mode, the control unit 140b controls the recognition method as that corresponding to the “first mode”, and the display processing of the screen displayed on the display unit 1 is “provisional display”. Control as “A”. In addition, when the immediately preceding mode is the second mode, the control unit 140b controls the recognition method as that corresponding to the “first mode”, and performs the display processing of the screen displayed on the display unit 1 as “provisional display B”. To control. “Temporary display A” and “provisional display B” represent different temporary displays. In FIG. 16, when the moving speed of the hand is less than the second threshold value, or when the moving speed of the hand is equal to or higher than the first threshold value, the determination unit 130b and the control unit 140b are the same as the processing described in FIG. Perform the following process.

  In the example of FIG. 17, a third threshold value that is larger than the first threshold value is further used. In FIG. 17, the determination unit 130b determines that the moving speed of the hand is equal to or higher than the second threshold and lower than the first threshold. As a result, when the immediately preceding mode is the first mode, the control unit 140b controls the recognition method as that corresponding to the “first mode”, and the display processing of the screen displayed on the display unit 1 is “provisional display”. Control as “A”. In addition, when the previous mode is the second mode, the control unit 140b controls the recognition method and the display processing of the screen displayed on the display unit 1 as corresponding to the “first mode”.

  In FIG. 17, the determination unit 130b determines that the moving speed of the hand is equal to or higher than the first threshold and lower than the third threshold. As a result, when the previous mode is the first mode, the control unit 140b controls the recognition method and the display process of the screen displayed on the display unit 1 according to the “second mode”. In addition, when the immediately preceding mode is the second mode, the control unit 140b controls the recognition method according to the “second mode”, and performs the display process of the screen displayed on the display unit 1 as “provisional display B”. To control.

  In FIG. 17, the determination unit 130b determines that the moving speed of the hand is equal to or higher than the third threshold value. As a result, when the previous mode is the first mode, the control unit 140b controls the recognition method and the display process of the screen displayed on the display unit 1 according to the “second mode”. In addition, when the immediately preceding mode is the second mode, the control unit 140b controls the recognition method and the display process of the screen displayed on the display unit 1 according to the “second mode”. Note that when the moving speed of the hand is less than the second threshold, the determination unit 130b and the control unit 140b perform the same processing as the processing described in FIG.

  Also in FIG. 18, the third threshold is used as in the example of FIG. In FIG. 18, the determination unit 130b determines that the moving speed of the hand is less than the second threshold value. As a result, when the immediately preceding mode is the first mode, the control unit 140b controls the recognition method and the display process of the screen displayed on the display unit 1 as corresponding to the “first mode”. In addition, in the state where the second mode is selected, the control unit 140b sets the recognition method to “second mode” when the duration of the state that is less than the second threshold in the determination by the determination unit 130b is less than the predetermined threshold. ", And the display processing of the screen displayed on the display unit 1 is controlled as" provisional display B ". On the other hand, in the state where the second mode is selected, the control unit 140b recognizes the recognition method and the display unit 1 when the duration of the state that is less than the second threshold in the determination by the determination unit 130b is equal to or greater than the predetermined threshold value. The display processing of the screen to be displayed is controlled according to the “first mode”.

  In FIG. 18, the determination unit 130b determines that the moving speed of the hand is equal to or higher than the second threshold and lower than the first threshold. As a result, when the immediately preceding mode is the first mode, the control unit 140b controls the recognition method as that corresponding to the “first mode”, and the display processing of the screen displayed on the display unit 1 is “provisional display”. Control as “A”. In addition, when the second mode is selected, the control unit 140b recognizes when the duration of the state that is equal to or greater than the second threshold and less than the first threshold in the determination by the determination unit 130b is less than the predetermined threshold. Is controlled according to the “second mode”, and the display processing of the screen displayed on the display unit 1 is controlled as “provisional display B”. On the other hand, in the state where the second mode is selected, the control unit 140b recognizes when the duration of the state in which the determination by the determination unit 130b is greater than or equal to the second threshold and less than the first threshold is greater than or equal to the predetermined threshold. And the display processing of the screen displayed on the display unit 1 is controlled according to the “first mode”.

  In FIG. 18, the determination unit 130b determines that the moving speed of the hand is equal to or higher than the first threshold and lower than the third threshold. Thereby, in the state where the first mode is selected, the control unit 140b recognizes when the duration of the state that is equal to or greater than the first threshold and less than the third threshold in the determination by the determination unit 130b is less than the predetermined threshold. The method is controlled according to the “first mode”, and the display processing of the screen displayed on the display unit 1 is controlled as “provisional display A”. On the other hand, in the state where the first mode is selected, the control unit 140b recognizes when the duration of the state in which the determination by the determination unit 130b is greater than or equal to the first threshold and less than the third threshold is greater than or equal to the predetermined threshold. And the display process of the screen displayed on the display unit 1 is controlled according to the “second mode”. In addition, when the immediately preceding mode is the second mode, the control unit 140b controls the recognition method according to the “second mode”, and performs the display process of the screen displayed on the display unit 1 as “provisional display B”. To control.

  In FIG. 18, the determination unit 130b determines that the moving speed of the hand is greater than or equal to the third threshold value. Thereby, in the state where the first mode is selected, the control unit 140b changes the recognition method to “first” when the duration of the state that is equal to or greater than the third threshold in the determination by the determination unit 130b is less than the predetermined threshold. The display processing of the screen displayed on the display unit 1 is controlled as “provisional display A”. On the other hand, in the state where the first mode is selected, the control unit 140b sets the recognition method and display unit 1 to the recognition method and the display unit 1 when the duration of the state that is equal to or greater than the third threshold in the determination by the determination unit 130b is equal to or greater than the predetermined threshold. The display process of the screen to be displayed is controlled according to the “second mode”. In addition, when the immediately preceding mode is the second mode, the control unit 140b controls the recognition method and the display process of the screen displayed on the display unit 1 according to the “second mode”.

  According to the present embodiment, since the cursor display is provisionally displayed before the transition to each of the first mode and the second mode, there is a possibility of transition to each mode, or transition to each mode. This can be recognized by the user. In addition, according to the present embodiment, the mode is changed to each mode or the cursor display is temporarily displayed according to the duration of the state of the feature amount such as the moving speed of the hand. This makes it possible for the user to recognize that there is a characteristic and that the mode has been shifted to each mode. Further, according to the present embodiment, it is not necessary for the user to perform an operation for switching the operation mode of the operation target device, so that the operability can be improved.

(Modification 3 of the first embodiment)
FIG. 19 is a block diagram illustrating a configuration example of a recognition device according to Modification 3 of the first embodiment. In Modification 3 of the first embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed description of the same components may be omitted. In Modification 3 of the first embodiment, the functions, configurations, and processes of the units other than the determination unit 130c and the control unit 140c described below are the same as those in the first embodiment.

  As illustrated in FIG. 19, the recognition apparatus 100c includes an acquisition unit 110, a calculation unit 120, a determination unit 130c, and a control unit 140c. Further, as in the first embodiment, the recognition device 100 c is connected to the display unit 1. In the third modification of the first embodiment, the first threshold is different from that of the first embodiment. Specifically, in Modification 3 of the first embodiment, a value larger or smaller than the first threshold by a predetermined value is used as the threshold. Specifically, in a state where the first mode is selected, a value that is larger than the first threshold by a predetermined value is set as the fourth threshold. On the other hand, when the second mode is selected, a value that is smaller than the first threshold by a predetermined value is set as the fifth threshold.

  When the current state is the “first mode”, the determination unit 130c uses the fourth threshold that is larger than the first threshold by a predetermined value and uses the fourth moving speed calculated by the calculating unit 120 and the fourth moving speed. The magnitude relationship with the threshold is determined. Then, the determination unit 130c outputs a determination result of the magnitude relationship between the hand movement speed and the fourth threshold value to the control unit 140c. On the other hand, when the current state is the “second mode”, the determination unit 130c uses the fifth threshold value that is smaller than the first threshold value by a predetermined value and uses the hand movement speed calculated by the calculation unit 120. The magnitude relationship with the fifth threshold is determined. Then, the determination unit 130c outputs a determination result of the magnitude relationship between the hand movement speed and the fifth threshold value to the control unit 140c.

  The control unit 140c receives the determination result by the determination unit 130c, selects the first mode when the moving speed of the hand is less than the fourth threshold value, and selects the second mode when the moving speed of the hand is equal to or higher than the fourth threshold value. Select a mode. In addition, the control unit 140c receives the determination result by the determination unit 130c, selects the first mode when the moving speed of the hand is less than the fifth threshold, and when the moving speed of the hand is equal to or higher than the fifth threshold. Select the second mode. In addition, the control unit 140c controls display processing of a screen displayed on the display unit 1 in accordance with the user's gesture in each of the selected first mode or second mode.

  Next, the flow of overall processing according to Modification 3 of the first embodiment will be described with reference to FIG. FIG. 20 is a flowchart illustrating an example of the flow of overall processing according to Modification 3 of the first embodiment.

  As illustrated in FIG. 20, when the position of the user's hand is acquired by the acquisition unit 110 (Yes at Step S301), the calculation unit 120 calculates the moving speed of the hand obtained from the hand movement (Step S302). . Further, when the position of the user's hand is not acquired by the acquisition unit 110 (No at Step S301), the acquisition unit 110 is in a state of waiting for acquisition of the position.

  In addition, when the current state is the first mode (Yes at Step S303), the determination unit 130c determines to use the fourth threshold value that is a predetermined value larger than the first threshold value (Step S304). . On the other hand, when the current state is the second mode (No at Step S303), the determination unit 130c determines to use the fifth threshold value that is a predetermined value smaller than the first threshold value (Step S305). . Accordingly, the determination unit 130c compares the magnitude of the moving speed of the hand with the fourth threshold value or the fifth threshold value determined to be used (step S306).

  At this time, when the determination unit 130c determines that the moving speed of the hand is less than the fourth threshold value or less than the fifth threshold value (Yes at Step S307), the control unit 140c selects the first mode (Step S308). . On the other hand, when the determination unit 130c determines that the moving speed of the hand is equal to or higher than the fourth threshold value or the fifth threshold value (No at Step S307), the control unit 140c selects the second mode (Step S309).

  According to the present embodiment, the fourth threshold value that is larger by a predetermined value than the first threshold value is used in the first mode, and the fifth threshold value that is smaller than the first threshold value by the second mode. Since it is used, it can be made difficult to switch modes. That is, according to the present embodiment, it is possible to suppress the mode from being easily switched by an inadvertent gesture.

(Modification 4 of the first embodiment)
FIG. 21 is a block diagram illustrating a configuration example of a recognition device according to Modification 4 of the first embodiment. In the fourth modification of the first embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed description of the same components may be omitted. In Modification 4 of the first embodiment, the functions, configurations, and processes of the units other than the control unit 140d, the first recognition unit 150d, and the second recognition unit 160d described below are the same as those in the first embodiment.

  As illustrated in FIG. 21, the recognition device 100d includes an acquisition unit 110, a calculation unit 120, a determination unit 130, a control unit 140d, a first recognition unit 150d, and a second recognition unit 160d. In addition, as in the first embodiment, the recognition device 100d is connected to the display unit 1.

  The first recognition unit 150d recognizes the position and movement of a specific part such as a user's hand or finger from the position acquired in time series by the acquisition unit 110 by the recognition method corresponding to the first mode. Then, the first recognition unit 150d outputs the recognized position and movement to the control unit 140d. The recognition process by the first recognition unit 150d is executed every time the acquisition unit 110 acquires a position. Here, the first mode is a mode for recognizing a gesture for moving the cursor as one aspect.

  In the recognition process by the first recognition unit 150d, the position acquired by the acquisition unit 110 may be output as it is, or may be output after correcting the acquired position. As for the position correction, for example, a relative position from the position where the hand detection is started may be set as the output target position. Alternatively, a Kalman filter or the like may be applied to smooth the motion, and the obtained position may be set as the output target position. Further, as described above, since the first mode is a mode for moving the cursor, it is preferable to output a position with a limited movement range so that the cursor does not move outside the screen.

  The second recognizing unit 160d recognizes the position and movement of a specific part such as the user's hand or finger from the position acquired in time series by the acquiring unit 110 by the recognition method corresponding to the second mode. Then, the second recognizing unit 160d outputs a command according to the recognized position and direction of movement to the control unit 140d. The recognition processing by the second recognition unit 160d is executed every time the acquisition unit 110 acquires a position. Here, the second mode is a mode for recognizing a gesture for a command in a corresponding direction such as moving a scroll bar on the screen as one aspect.

In the recognition process by the second recognition unit 160d, the direction of the command and the progress rate (0 to 1) until command confirmation may be output. The progress rate is, for example, “0” for the time “T _j ” that is the starting point of position acquisition by the acquisition unit 110, and “1” for the time “T _j + T _0a ” when the predetermined time “T _0a ” elapses. ”, A value obtained by linearly interpolating a period from time“ T _j ”to time“ T _j + T _0a ”. In this case, the progress rate at time “t (T _j ≦ t ≦ T _j + T _0a )” is “(t−T _j ) / T _0a ”. That is, if the progress rate is “1”, the command is in a fixed state. The method is not limited to the above method, and a method of recognizing a gesture in the direction intended by the user from the movement locus of the hand including before and after the turning point and outputting a command in the corresponding direction may be used. .

  The control unit 140d selects the first mode or the second mode according to the determination result by the determination unit 130. Then, when the current state is the first mode, the control unit 140d selects the output of the recognition result by the first recognition unit 150d, and performs display processing of a screen displayed on the display unit 1 according to the recognition result. Control. On the other hand, when the current state is the second mode, the control unit 140d selects the output of the recognition result by the second recognition unit 160d, and performs display processing of a screen displayed on the display unit 1 according to the recognition result. Control. In addition, when the progress rate is output by the second recognition unit 160d when the current state is the second mode, the control unit 140d shifts to the first mode if the progress rate is continuously output. However, only the recognition result by the second recognition unit 160d may be selected. Alternatively, not only the recognition result by the second recognition unit 160d but also the recognition result by the first recognition unit 150d, and a screen displayed on the display unit 1 according to the recognition results by the first recognition unit 150d and the second recognition unit 160d The display process may be controlled.

  Next, the flow of overall processing according to Modification 4 of the first embodiment will be described with reference to FIG. FIG. 22 is a flowchart illustrating an example of the flow of overall processing according to Modification 4 of the first embodiment.

  As illustrated in FIG. 22, when the position of the user's hand is acquired by the acquisition unit 110 (Yes at Step S401), each process of Step S402, Step S403, and Step S404 is executed. In addition, when the position of the user's hand is not acquired by the acquisition unit 110 (No in step S401), the acquisition unit 110 waits for acquisition of the position.

  Here, the first recognition unit 150d recognizes the position and movement of the user's hand from the position acquired in time series by the acquisition unit 110 by the recognition method corresponding to the first mode, and the cursor position that is the recognition result. Is output to the control unit 140d (step S402). Further, the second recognition unit 160d recognizes the position and movement of the user's hand from the position acquired in time series by the acquisition unit 110 by the recognition method corresponding to the second mode, and the direction of the command that is the recognition result Is output to the control unit 140d (step S403). Further, the calculation unit 120 calculates the moving speed of the hand obtained from the hand movement (step S404).

  In addition, the determination unit 130 compares the magnitude of the hand movement speed calculated by the calculation unit 120 with the first threshold (step S405). At this time, when the determination unit 130 determines that the moving speed of the hand is less than the first threshold (Yes in step S406), the control unit 140d selects the first mode and is output by the first recognition unit 150d. The display processing of the screen according to the cursor position that is the recognition result is controlled (step S407). On the other hand, when the determination unit 130 determines that the moving speed of the hand is equal to or higher than the first threshold (No at Step S406), the control unit 140d selects the second mode and is output by the second recognition unit 160d. The display processing of the screen displayed on the display unit 1 is controlled according to the direction of the command that is the recognition result (step S408).

  According to this embodiment, according to the movement speed of the specific part in the gesture by the user, the gesture recognition method related to the control of the display processing of the screen displayed on the display unit is selected, and the gesture corresponding to the selected recognition method is selected. A recognition result is selected, and display processing of a screen displayed on the display unit is controlled according to the selected recognition result. As a result, it is possible to improve the operability related to the switching of the operation mode of the operation target device according to the gesture, and display processing of the screen displayed on the display unit according to the recognition result of the gesture according to the corresponding mode Can be controlled.

(Modification 5 of the first embodiment)
FIG. 23 is a block diagram illustrating a configuration example of a recognition device according to Modification Example 5 of the first embodiment. In the modified example 5 of the first embodiment, the same reference numerals are given to the same components as those of the first embodiment, and detailed description of the same components may be omitted. In the fifth modification of the first embodiment, the function, configuration, and processing of each unit other than the determination unit 130e, the control unit 140e, the first recognition unit 150e, and the second recognition unit 160e described below are the same as those in the first embodiment. It is the same.

  As illustrated in FIG. 23, the recognition apparatus 100e includes an acquisition unit 110, a calculation unit 120, a determination unit 130e, a control unit 140e, a first recognition unit 150e, and a second recognition unit 160e. In addition, the recognition device 100e is connected to the display unit 1 as in the first embodiment.

  The determining unit 130e determines the magnitude relationship between the moving speed of the hand and the first threshold using the moving speed of the hand calculated by the calculating unit 120 as a feature amount. Then, the determination unit 130e determines the determination result of the magnitude relationship between the hand movement speed and the first threshold value when the hand movement speed is less than the first threshold value with respect to the first recognition unit 150e. Output. On the other hand, regarding the determination result of the magnitude relationship between the moving speed of the hand and the first threshold, the determining unit 130e sends the determination result to the second recognizing unit 160e when the moving speed of the hand is equal to or higher than the first threshold. Output.

  That is, when the moving speed of the hand is less than the first threshold value, only the recognition process corresponding to the first mode, that is, the recognition process by the first recognition unit 150e, needs to be executed. To the unit 150e. Similarly, when the moving speed of the hand is equal to or higher than the first threshold value, only the recognition process according to the second mode, that is, the recognition process by the second recognition unit 160e need be executed. This is output to the recognition unit 160e.

  When receiving the determination result from the determination unit 130e, the first recognition unit 150e acquires the time-series position from the acquisition unit 110, and uses the recognition method corresponding to the first mode to recognize the user's hand or finger from the acquired position. Recognize the position and movement of specific parts. Then, the first recognition unit 150e outputs the recognized position and movement to the control unit 140e. The recognition process by the first recognition unit 150e is executed when a determination result from the determination unit 130e is received. The position correction and the like in the recognition process by the first recognition unit 150e are the same as those of the first recognition unit 150d.

  When receiving the determination result from the determination unit 130e, the second recognition unit 160e acquires a time-series position from the acquisition unit 110, and uses the recognition method corresponding to the second mode to recognize the user's hand or finger from the acquired position. Recognize the position and movement of specific parts. Then, the second recognition unit 160e outputs a command corresponding to the recognized position and direction of movement to the control unit 140e. The recognition process by the second recognition unit 160e is executed when a determination result from the determination unit 130e is received. Note that the progress rate output and the like in the recognition processing by the second recognition unit 160e is the same as that of the second recognition unit 160d.

  When the recognition result is output by the first recognition unit 150e, the control unit 140e selects the first mode and controls the display process of the screen displayed on the display unit 1 according to the output recognition result. On the other hand, when the recognition result is output by the second recognition unit 160e, the control unit 140e selects the second mode and controls the display processing of the screen displayed on the display unit 1 according to the output recognition result. .

  Next, the flow of overall processing according to Modification 5 of the first embodiment will be described with reference to FIG. FIG. 24 is a flowchart illustrating an example of the flow of overall processing according to Modification 5 of the first embodiment.

  As illustrated in FIG. 24, when the position of the user's hand is acquired by the acquisition unit 110 (Yes in step S501), the calculation unit 120 calculates the moving speed of the hand obtained from the hand movement (step S502). . In addition, when the position of the user's hand is not acquired by the acquisition unit 110 (No in step S501), the acquisition unit 110 waits for acquisition of the position.

  In addition, the determination unit 130e compares the magnitude relationship between the hand movement speed calculated by the calculation unit 120 and the first threshold value (step S503). At this time, if the determination unit 130e determines that the moving speed of the hand is less than the first threshold value (Yes in step S504), the determination unit 130e outputs a determination result to the first recognition unit 150e. On the other hand, when the determination unit 130e determines that the moving speed of the hand is equal to or higher than the first threshold (No in step S504), the determination unit 130e outputs a determination result to the second recognition unit 160e.

  In addition, the first recognition unit 150e that has received the determination result from the determination unit 130e acquires the time-series position from the acquisition unit 110, and recognizes the position and movement of the user's hand by the recognition method corresponding to the first mode. The cursor position that is the recognition result is output to the control unit 140e (step S505). Thereafter, the control unit 140e that has received the recognition result from the first recognition unit 150e selects the first mode, and controls display processing of the screen displayed on the display unit 1 in accordance with the cursor position that is the recognition result (step). S506).

  In addition, the second recognition unit 160e that has received the determination result from the determination unit 130e acquires a time-series position from the acquisition unit 110, and recognizes the position and movement of the user's hand by a recognition method corresponding to the second mode. The direction of the command that is the recognition result is output to the control unit 140e (step S507). Thereafter, the control unit 140e that receives the recognition result from the second recognition unit 160e selects the second mode, and controls display processing of the screen displayed on the display unit 1 according to the direction of the command that is the recognition result ( Step S508).

  According to the present embodiment, since only the gesture recognition process corresponding to any mode is executed after the mode determination result is output, the gesture recognition result corresponding to the corresponding mode is reduced while reducing the processing load. Accordingly, the display processing of the screen displayed on the display unit can be controlled.

(Second Embodiment)
FIG. 25 is a block diagram illustrating a configuration example of a recognition device according to the second embodiment. In the second embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed description of the same components may be omitted. In the second embodiment, the function, configuration, and processing of the acquisition unit 110 described below are the same as those in the first embodiment.

  As illustrated in FIG. 25, the recognition apparatus 200 includes an acquisition unit 110, a calculation unit 220, a determination unit 230, and a control unit 240. Further, the recognition device 200 is connected to the display unit 1 as in the first embodiment.

  The calculation unit 220 counts the number of times that a movement in which the time from the change of the sign of the inner product between a movement vector of a specific part such as a hand or a finger and a predetermined coordinate axis vector to the next change is within a predetermined time continuously occurs. Is calculated as a feature amount. For example, the calculation unit 220 calculates, as a feature amount, the number of times that a hand movement occurs within a predetermined time such as “350 ms” from the turning point (or the point at the start of detection) to the next turning point. To do. That is, when the reciprocating motion of the hand is performed as a gesture, such as when the user shakes his / her hand from side to side, the number of occurrences of the reciprocation of the reciprocating motion within a predetermined time is used as the feature amount.

  The determination unit 230 determines the magnitude relationship between the number of times and the sixth threshold value representing a specific value of the number of times, with the number of occurrences of hand reciprocation within a predetermined time calculated by the calculation unit 220 as a feature amount. To do. Then, the determination unit 230 outputs a determination result of the magnitude relationship between the number of times and the sixth threshold value to the control unit 240. For example, the sixth threshold is “4”. That is, in the determination unit 230, when the command corresponding to the second mode is executed according to the reciprocation of the hand, whether the number of occurrences of the reciprocation is greater than or less than “4”. Is determined. This is because the cursor movement by hand-wiping corresponding to the first mode and the command by reciprocation of the hand corresponding to the second mode are similar gestures.

  The control unit 240 receives the determination result by the determination unit 230, selects the first mode when the number of times is less than the sixth threshold value, and selects the second mode when the number of times is equal to or more than the sixth threshold value. In addition, the control unit 240 controls display processing of a screen displayed on the display unit 1 in accordance with the user's gesture in each of the selected first mode or second mode. That is, in the second embodiment, when a command corresponding to the second mode is executed according to the reciprocation of the hand, both modes are selected in the process of selecting either the first mode or the second mode. Has the effect of deterring the selection.

  FIG. 26 is a diagram for explaining the similarity of gestures. For example, as shown in the upper part of FIG. 26, a plurality of application icons are displayed on the screen 30. At this time, in the state where the application 32 is focused by the cursor 31, that is, in the state where the application 32 is selected, when the hand is reciprocated, the lower left part of FIG. 26 and the lower part of FIG. The recognition result is shown on the right side.

  Specifically, in the lower left part of FIG. 26, the hand 31 moves in the left-right direction with continuous hand-drawing in the right direction, so that the cursor 31 moves in the right direction on the screen 33, and the application 34 Is selected. In the lower right part of FIG. 26, the application 32 is activated on the screen 35 because the hand moves in the left-right direction due to the reciprocation of the hand. In the second embodiment, when a command corresponding to the second mode is executed according to the reciprocation of the hand, both modes are selected in the process of selecting either the first mode or the second mode. (See FIG. 26) can be suppressed. Hereinafter, display examples of screens in the respective modes according to the second embodiment will be described.

  FIG. 27 is a diagram illustrating a display example of the screen in the first mode according to the second embodiment. As shown in FIG. 27, the cursor 41 is moved in the right direction on the screen 40 when the hand is moved in the right direction. At this time, a display 42 indicating that the cursor 41 is moved in the right direction may be output to the screen.

  FIG. 28 is a diagram illustrating a display example of the screen in the second mode according to the second embodiment. As shown in FIG. 28, in the state where the application 52 is focused by the cursor 51 on the screen 50, the application 52 is activated when the hand is removed in the left-right direction. At this time, a display 53 indicating that the hand icon is reciprocating left and right (activating the application 52) may be output on the screen. The display 53 may be output while the progress rate is being output.

  Next, the flow of overall processing according to the second embodiment will be described with reference to FIG. FIG. 29 is a flowchart illustrating an example of the flow of overall processing according to the second embodiment.

  As illustrated in FIG. 29, when the position of the user's hand is acquired by the acquisition unit 110 (Yes in step S601), the calculation unit 220 calculates the number of occurrences of reciprocation of the user's hand reciprocation as a feature amount. (Step S602). In addition, when the position of the user's hand is not acquired by the acquisition unit 110 (No in step S601), the acquisition unit 110 waits for acquisition of the position.

  In addition, the determination unit 230 compares the magnitude relationship between the number of times calculated by the calculation unit 220 and the sixth threshold value (step S603). At this time, when the determination unit 230 determines that the number of times is less than the sixth threshold value (Yes in step S604), the control unit 240 selects the first mode (step S605). On the other hand, when the determination unit 230 determines that the number of times is equal to or greater than the sixth threshold (No in step S604), the control unit 240 selects the second mode (step S606).

  According to the present embodiment, when the gestures recognized in the first mode and the second mode are similar, the modes are separated by detecting the reciprocation of the hand and the like, so both modes are selected simultaneously. Can be suppressed.

(Third embodiment)
Now, although the embodiment of the recognition device has been described so far, it may be implemented in various different forms other than the above-described embodiment. Therefore, (1) the position of the specific part, (2) the feature value, (3) provisional display, (4) notification to the user that the transition is in progress, (5) threshold correction, (6) similar gesture, Different embodiments of (7) recognition result output, (8) other application examples, and (9) hardware configuration will be described.

(1) Position of Specific Part In the above embodiment, a specific part such as a user's hand or finger that is a subject is detected for each image with respect to a time-series image captured by a camera, and is tracked in time series. Thus, the case where the position of the specific part of the user is acquired in time series has been described. The position of the specific part may be a coordinate value normalized by the size of the face of the user who is the subject or the part detected first. The position of the specific part is not limited to the position on the captured image, and depth information estimated from a distance sensor or a plurality of image sensors may be used as one of the coordinate axes. Further, the specific part may be a head, a light emitting part of a laser pointer, or the like in addition to a hand or a finger. Further, the processing by the acquisition unit 110 that acquires the position of the specific part uses a pattern recognition method disclosed in Japanese Patent Laid-Open No. 2006-268825, a background difference method disclosed in Japanese Patent No. 3777650, or the like. Thus, the position of the specific part of the user may be acquired in time series.

(2) Feature Amount In the above-described embodiment, the case where the moving speed of the hand, the number of occurrences of the reciprocation of the hand, etc. are used as the feature amount has been described. As the feature amount as the moving speed “V” of the hand, the maximum one of the speeds of the section vectors from the most recent position to the immediately preceding turning point or the position acquisition start point may be used. Further, the feature amount may be a movement acceleration “α”. When obtaining the acceleration of the latest vector, the moving acceleration “α” is acquired at time “T ₇ ” with the most recent position acquired at time “T ₈ ” as “P ₈ ” (see FIG. 3). When the previous position is “P ₇ ” and the previous position acquired at time “T ₆ ” is “P ₆ ”, “[(speed from P ₈ to P ₇ ) − (from P ₇ speed up P _6)] is calculated by _{/ (T} 8 -T _7). " Further, as the movement acceleration “α”, the maximum one of the accelerations of the section vector from the most recent position to the immediately preceding turning point or the position acquisition start point may be used.

The feature amount may be a value of a function expressed by an operation distance “L”, a movement speed “V”, and a movement acceleration “α” of a specific part. The operation distance “L” is the sum of the Euclidean distances of vectors from the immediately preceding turning point or the position acquisition start point to the nearest position. If the feature quantity is “E”, it may be simply multiplied as “E = V × L × α”, or may be multiplied “E = V × √L” with a weight added. Moreover, it is good also as a value which changes according to conditions as follows.
E = V (L ≧ threshold L1)
0 (L <threshold L1)

  In the case of performing a gesture operation using an optical pointer having a light source or the like, for example, acceleration can be used as a feature amount. Of course, acceleration may also be used in hand gestures. In the case of a gesture by hand, the hand is often shaken. Therefore, when the acceleration is used as a feature amount in the gesture by the optical pointer, the accuracy is higher.

(3) Temporary display Moreover, the said embodiment demonstrated the case where a cursor was provisionally displayed according to the feature-value and the last mode. For the temporary display, both the cursor and the block arrow 17 shown in FIG. 5 are displayed, and when the feature amount is the first threshold, the display is set to 0% according to the first mode, and according to the second mode. The size and brightness may be changed at a rate of 100% display. At this time, when the feature amount is the second threshold value, the size and brightness are changed at a rate of 100% display corresponding to the first mode and 0% display corresponding to the second mode. Further, the cursor and the block arrow 17 may be morphed at the above ratio.

(4) Notification to User that Transition is in Progress In the above-described embodiment, a case has been described in which the user is notified that the mode is transitioning by temporarily displaying a cursor. The notification to the user regarding the mode transition may be made to display which mode is currently in an arbitrary position on the screen. 30 and 31 are diagrams illustrating an example of notification to a user related to mode transition. As shown in FIG. 30, any one of “first mode”, “shifting”, and “second mode” is turned on or blinked. Note that “in transition” shown in FIG. 30 indicates a state in which the cursor is temporarily displayed.

  Further, as shown in FIG. 31, any one of “first mode”, “shifting to second mode”, “second mode”, and “shifting to first mode” is turned on or blinked. . Note that “in transition to the first mode” and “in transition to the second mode” illustrated in FIG. 31 indicate a state in which provisional display of different cursors is performed. Specifically, “in transition to the first mode” corresponds to “provisional display B”, and “in transition to the second mode” corresponds to “provisional display A”.

(5) Correction of threshold In the above embodiment, a value that is larger than the first threshold by a predetermined value is set as a fourth threshold, and a value that is smaller than the first threshold by a predetermined value is set as a fifth threshold. The case where the threshold value is used has been described. As for the correction of the threshold value, a threshold value obtained by performing a correction opposite to the above may be used. Specifically, a fourth threshold value that is smaller than the first threshold value is used in the first mode, and a fifth threshold value that is larger than the first threshold value is used in the second mode. . Further, not only the first threshold value but also the second threshold value and the third threshold value may be used in the same manner. Regarding the correction of the second threshold value and the third threshold value, the same value as that added or subtracted in the correction of the first threshold value may be added or subtracted, or “fourth threshold value / first threshold value” or “ A value obtained by multiplying the ratio of the “fifth threshold value / first threshold value” by the second threshold value and the third threshold value may be used.

  As for threshold correction, the threshold may be changed according to the user. FIG. 32 is a block diagram illustrating a configuration example of a recognition device to which an identification unit according to the third embodiment is added. As illustrated in FIG. 32, the recognition device 100f includes an acquisition unit 110, a calculation unit 120, a determination unit 130f, a control unit 140, and an identification unit 170f. As in the first embodiment, the recognition device 100f is connected to the display unit 1. In FIG. 32, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed description of the same components may be omitted. In FIG. 32, the function, configuration, and processing of each unit other than the determination unit 130f and the identification unit 170f described below are the same as those in the first embodiment.

  In the recognition device 100f, an average value is obtained for each user regarding the size of the hand swing, the moving speed, and the like when performing a gesture from the past operation history of the user. The identification unit 170f identifies a user who is a subject. For example, the identification unit 170f performs face authentication of the user who made a gesture from the video imaged by the camera 11, and outputs the identification result to the determination unit 130f. As for user identification, any method capable of identifying the user, such as fingerprint authentication, may be used instead of face authentication.

  When the determination unit 130f receives the user identification result from the identification unit 170f, the determination unit 130f calculates the first threshold value based on the ratio between the average value of the moving speed of the hand corresponding to the user and the assumed standard moving speed of the hand. Correct the value of. Then, the determination unit 130f determines the magnitude relationship between the hand movement speed and the corrected first threshold using the hand movement speed calculated by the calculation unit 120 as a feature amount. Thereafter, the determination unit 130f outputs a determination result of the magnitude relationship between the hand movement speed and the corrected first threshold value to the control unit 140. The threshold correction may be performed not only on the first threshold but also on the second threshold in the same manner as described above.

  As for the correction of the threshold value, the threshold value may be changed using the shape of the specific part. FIG. 33 is a block diagram illustrating a configuration example of a recognition device to which a shape recognition unit according to the third embodiment is added. As illustrated in FIG. 33, the recognition apparatus 100g includes an acquisition unit 110, a calculation unit 120, a determination unit 130g, a control unit 140, and a shape recognition unit 180g. Further, as in the first embodiment, the recognition device 100 g is connected to the display unit 1. In FIG. 33, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed description of the same components may be omitted. In FIG. 33, the function, configuration, and processing of each unit other than the determination unit 130g and the shape recognition unit 180g described below are the same as those in the first embodiment.

  The recognition apparatus 100g corrects the threshold value using the shape of a specific part such as a hand or a finger in order to improve the accuracy of mode selection. The shape recognition unit 180g recognizes the shape of a specific part such as a hand or a finger. For example, the shape recognition unit 180g recognizes the shape of the palm of the user who made the gesture from the video imaged by the camera 11, and outputs the recognition result to the determination unit 130g.

  When the determination unit 130g receives the palm shape recognition result from the shape recognition unit 180g, the determination unit 130g calculates the degree of coincidence between the palm shape as the recognition result and the predetermined shape, and the degree of coincidence is greater than the predetermined degree of coincidence. If it is larger, the first threshold is increased by a predetermined value. In addition, the determination unit 130g decreases the first threshold value by a predetermined value when the matching degree is smaller than the predetermined matching degree. Then, the determination unit 130g determines the magnitude relationship between the hand movement speed calculated by the calculation unit 120 and the corrected first threshold value. Thereafter, the determination unit 130g outputs a determination result of the magnitude relationship between the hand movement speed and the corrected first threshold value to the control unit 140. The threshold correction may be performed not only on the first threshold but also on the second threshold in the same manner as described above. In other words, if the degree of coincidence between the shape of the specific part where the gesture is performed and the predetermined shape is high, there is a high possibility that the gesture corresponding to the first mode in which there are many simple gestures with little movement such as cursor movement is performed. For this reason, when the degree of coincidence is high, a threshold value that is larger than the first threshold value by a predetermined value is used to facilitate determination as the first mode.

(6) Similar Gestures In the above-described embodiment, a case has been described in which the number of occurrences of reciprocation of a hand within a predetermined time is used as a feature amount and a mode is suitably selected for similar gestures. As the feature amount, the number of movements in which the sign of the inner product of the movement vector of the specific part and the predetermined coordinate axis vector is switched within the past predetermined time may be used. For example, referring to the example of FIG. 3, when the past time from “P ₁ ” to “P ₈ ” is within a predetermined time, the calculation unit 220 moves each of “P ₁ ” to “P ₈ ”. An inner product of the vector and a fixed coordinate axis vector is obtained. In the example of FIG. 3, “P ₁ ” to “P ₃ ” is the left direction vector, “P ₃ ” to “P ₆ ” is the right direction, and “P ₆ ” to “P ₈ ” is the left direction vector. It has become. For this reason, the sign of the inner product of each movement vector and the fixed coordinate axis vector is switched in the order of “positive”, “negative”, “positive” or “negative”, “positive”, “negative”. Thereby, in the example of FIG. 3, the number of times becomes “2”. The processing by the determination unit 230 and the control unit 240 is the same as that in the above embodiment.

In addition, if the distance of the reciprocation of the hand is shorter, it is considered that a gesture corresponding to the second mode is often performed. For this reason, the feature amount is calculated on the condition that the distance from the change of the sign of the inner product of the movement vector and the predetermined coordinate axis vector to the next change is a reciprocating motion of the hand within a predetermined scalar. For example, in the example of FIG. 3, the calculation unit 220 has predetermined distances from “P ₁ ” to “P ₃ ”, “P ₃ ” to “P ₆ ”, and “P ₆ ” to “P ₈ ”. If it is within the scalar, the total number of times of change of the code of the inner product is counted and calculated as a feature amount.

(7) Outputting Recognition Result In the above embodiment, the case where the recognition result of the gesture is output by the first recognition unit 150d, the second recognition unit 160d, or the like has been described. Depending on the recognition method in the first mode, the progress rate may be output before the command is confirmed. In the second mode recognition method, the progress rate may be output until it is determined that the hand is reciprocating, that is, while the hand is reciprocating within a predetermined time. While the progress rate output by the first recognition unit 150d continues in the first mode, the recognition result by the first recognition unit 150d may be output even if the mode is shifted to the second mode. Similarly, in the second mode, while the output of the progress rate by the second recognizing unit 160d continues in the second mode, the recognition result by the second recognizing unit 160d even if the mode is shifted to the first mode. May be output. That is, the recognition result is output from both the first recognition unit 150d and the second recognition unit 160d. In this state, for example, when the progress rate becomes “0.8” or more on one side, the other recognition result may not be output.

(8) Other application examples Moreover, the said embodiment demonstrated the case where a recognition apparatus was applied to information processing apparatuses, such as PC. The application of the recognition device is not limited to a PC, and can be applied to various devices. FIG. 34 is a diagram illustrating an example in which the recognition device is applied to a television receiver. A television receiver 60 shown in FIG. 34 is equipped with a camera 61 and images a viewer of the television receiver 60. The television receiver 60 detects and tracks the viewer's hand 62 as a specific part and recognizes the gesture of the hand 62. On the television screen 63, “Channel 4” is currently displayed. The viewer can change to “channel 3” by performing a gesture of moving the hand 62 to the left, and can change to “channel 5” by performing a gesture of moving the hand 62 to the right. it can.

  FIG. 35 is a diagram illustrating an example in which the recognition apparatus is applied to a medical information terminal. An information terminal 70 shown in FIG. 35 is equipped with any of the recognition devices of the above-described embodiments, and is used for providing information to a doctor during surgery. Since it is preferable that the doctor during the operation keeps the hand 71 clean, it is difficult to directly touch the hand 71 to the information terminal 70 that provides the doctor with information such as patient monitoring information and test results. Therefore, by applying any of the recognition devices of the above embodiment to the information terminal 70, the doctor can operate the information terminal 70 in a non-contact manner by the gesture of the hand 71. The information terminal 70 is equipped with a camera 72 and images a doctor. The information terminal 70 detects and tracks the doctor's hand 71 as a specific part, recognizes the gesture, and commands the information terminal 70 to scroll in the direction of the detected gesture and display information. Execute command etc.

  Note that the application of the recognition device is not limited to a PC, a television receiver, and a medical information terminal, and can also be applied to other devices such as a game device. In other words, regarding the gesture interface that allows the user to easily operate the device without having to wear or wear a remote controller, sensor, marker, etc., the operation for switching the menu displayed for device control or the content displayed on the device. It can be applied to equipment that can perform

(9) Hardware Configuration FIG. 36 is a diagram illustrating that the recognition program is realized using a computer. As shown in FIG. 36, a computer 1000 as a recognition device includes a control device such as a CPU 1001, a storage device such as a ROM 1002 and a RAM 1003, an HDD (Hard Disk Drive) 1004, a disk drive 1005, and the like connected by a bus 1009. External storage device, a display device such as a display 1006, and input devices such as a keyboard 1007 and a mouse 1008, and has a hardware configuration using a normal computer.

  The recognition program executed by the recognition device is, as one form, a computer in an installable or executable format, such as a CD-ROM, flexible disk (FD), CD-R, DVD (Digital Versatile Disk), etc. Recorded on a readable recording medium. The recognition program executed by the recognition apparatus may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. The recognition program executed by the recognition device may be provided or distributed via a network such as the Internet. The recognition program may be provided by being incorporated in advance in a ROM or the like.

  The recognition program executed by the recognition device has a module configuration including the above-described units (the acquisition unit 110, the calculation unit 120, the determination unit 130, and the control unit 140). As actual hardware, a CPU (processor) is used. By reading and executing the recognition program from the storage medium, the above-described units are loaded onto the main storage device, and the acquisition unit 110, the calculation unit 120, the determination unit 130, and the control unit 140 are generated on the main storage device. It has become.

  Further, the above-described embodiments are presented as examples, and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. Moreover, each embodiment can be combined suitably as long as the contents do not contradict each other. Each embodiment and its modifications are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 100 Recognition apparatus 110 Acquisition part 120 Calculation part 130 Determination part 140 Control part

Claims (16)

An acquisition unit for acquiring the position of a specific part of the subject in time series;
From the position acquired in time series, a calculation unit that calculates a feature amount of movement of the specific part;
A determination unit that determines a magnitude relationship between the calculated feature value and the first threshold value, and a magnitude relationship between the calculated feature value and a second threshold value that is smaller than the first threshold value;
A control unit that controls a display process for notifying that there is a possibility that the position or motion recognition method of the specific part may be switched when the feature amount is equal to or greater than the second threshold and less than the first threshold. And a recognition device. The determination unit further determines a magnitude relationship between the feature amount and a third threshold value greater than the first threshold value,
A first method that is the recognition method that is selected when the feature amount is less than the first threshold when the feature amount is greater than or equal to the first threshold and less than the third threshold; The recognition device according to claim 1, wherein different second methods are selected.   The control unit controls display processing of a screen displayed on the display unit by increasing or decreasing at least one of transparency, brightness, blinking speed, and pop-out amount of an object displayed on the display unit according to the feature amount. The recognition apparatus according to claim 1 or 2.   The control unit displays on the display unit corresponding to the first method that is the recognition method selected when the feature amount is less than the first threshold when the feature amount is the first threshold. Selected when the shape change of the object is 0% and the feature value is greater than or equal to the first threshold and displayed on the display unit corresponding to the second method that is the recognition method different from the first method The display processing of the screen displayed on the display unit is controlled at a rate that the shape change of the object to be performed is 100%, and displayed on the display unit corresponding to the first method when the feature amount is the second threshold value The display processing of the screen displayed on the display unit is controlled at a rate that the shape change of the object to be displayed is 100% and the shape change of the object displayed on the display unit corresponding to the second method is 0%. Any one of claims 1 to 3 The placing of the recognition device. The position or movement of the specific part is recognized from the position acquired in time series by the acquisition unit by the first method that is the recognition method selected when the feature amount is less than the first threshold. A first recognition unit;
The specific part selected from the position acquired in time series by the acquisition unit by the second method, which is selected when the feature amount is equal to or greater than the first threshold and is different from the first method. And a second recognition unit for recognizing the position or movement of
The control unit controls display processing of a screen displayed on the display unit according to a recognition result by the first recognition unit when the first method is selected, and the second method when the second method is selected. The recognition apparatus according to claim 1, wherein a display process of a screen displayed on the display unit is controlled according to a recognition result by the recognition unit. According to the first method, which is the recognition method selected when the position acquired in time series by the acquisition unit is acquired according to the determination result by the determination unit, and the feature amount is less than the first threshold value. A first recognition unit for recognizing the position or movement of the specific part from the acquired position;
The position acquired in time series by the acquisition unit according to a determination result by the determination unit is acquired, and is selected when the feature amount is equal to or greater than the first threshold, and the recognition is different from the first method. A second recognizing unit that recognizes the position or movement of the specific part from the acquired position by the second method, which is a method;
Only one of the recognition by the first recognition unit and the second recognition unit is executed,
When the recognition by the first recognition unit is executed, the control unit controls display processing of a screen displayed on the display unit according to the recognition result by the first recognition unit, and the recognition by the second recognition unit is performed. The recognition apparatus according to claim 1, wherein when executed, the display device controls display processing of a screen displayed on the display unit according to a recognition result by the second recognition unit.   The control unit is in a predetermined time when the feature amount is equal to or greater than the first threshold value when the first method, which is the recognition method selected when the feature amount is less than the first threshold value, is selected. The recognition apparatus according to claim 1, wherein when the process continues, the second method, which is the recognition method different from the first method, is selected.   When the second method, which is the recognition method selected when the feature amount is equal to or greater than the first threshold, is selected, the control unit is in a state where the feature amount is less than the first threshold for a predetermined time. The recognition apparatus according to claim 1, wherein when the process continues, the first method, which is the recognition method different from the second method, is selected.   When the first method, which is the recognition method selected when the feature amount is less than the first threshold value, is selected, the control unit has a value that is greater than the first threshold value by a predetermined value. The recognition apparatus according to claim 1, wherein a second method that is a recognition method different from the first method is selected when a certain fourth threshold value is exceeded.   When the second method, which is the recognition method selected when the feature amount is equal to or greater than the first threshold, is selected, the control unit has a value that is smaller than the first threshold by a predetermined value. The recognition apparatus according to claim 1, wherein a first method, which is the recognition method different from the second method, is selected when it is less than a fifth threshold value. An identification unit for identifying the user as the subject;
The determination unit uses at least one of the first threshold value increased or decreased according to the identified user and the second threshold value increased or decreased according to the identified user, The recognition apparatus according to claim 1, wherein a magnitude relationship with a feature amount is determined. A shape recognition unit for recognizing the shape of the specific part;
The determination unit includes the first threshold value that increases or decreases depending on the degree of matching between the recognized shape of the specific part and the predetermined shape, and the second threshold value that increases or decreases depending on the degree of matching. The recognition apparatus according to any one of claims 1 to 10, wherein a magnitude relationship with the feature amount is determined using at least one.   The recognition apparatus according to claim 1, wherein the calculation unit calculates at least one of a distance, a speed, and an acceleration of movement of the specific part as the feature amount.   When the second method, which is the recognition method selected when the feature amount is equal to or greater than the first threshold, is selected, the control unit continues the movement distance as the feature amount for a predetermined time or more. The recognition apparatus according to claim 13, wherein the first method, which is the recognition method different from the second method, is selected. An acquisition unit for acquiring the position of a specific part of the subject in time series;
Obtained in chronological order under the condition that the distance between the position from the change of the sign of the inner product of the movement vector of the specific part and the predetermined coordinate axis vector to the next switching is within the predetermined scalar. A calculation unit that calculates the number of times that a motion in which the time from the change of the sign of the inner product of the movement vector and the predetermined coordinate axis vector to the next change is within a predetermined time from the position to be generated continuously occurs; ,
A determination unit for determining a magnitude relationship between the calculated number of times and a sixth threshold;
When the number of times is less than the sixth threshold, the first method is selected as a method for recognizing the position or movement of the specific part, and when the number of times is equal to or greater than the sixth threshold, And a control unit for selecting a second method which is a different recognition method. An acquisition unit for acquiring the position of a specific part of the subject in time series;
Obtained in chronological order under the condition that the distance between the position from the change of the sign of the inner product of the movement vector of the specific part and the predetermined coordinate axis vector to the next switching is within the predetermined scalar. A calculation unit for calculating the number of movements in which the sign of the inner product of the movement vector and the predetermined coordinate axis vector is switched within the past predetermined time from the position to be
A determination unit for determining a magnitude relationship between the calculated number of times and a sixth threshold;
When the number of times is less than the sixth threshold, the first method is selected as a method for recognizing the position or movement of the specific part, and when the number of times is equal to or greater than the sixth threshold, And a control unit for selecting a second method which is a different recognition method.

Images