JP2011113439A - Input device, input method, and input program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an input device, an input method, and an input program, for accurately determining input data in accordance with an operation state of a virtual operation member. <P>SOLUTION: In the input device for determining input data in accordance with movement of user's hands operating one or more virtual operation members, detection data within an acquisition period including input timing at which operation members have been operated by hands, and having a first prescribed length and assigned data are stored in a storage means as learning data in association with each other (S160) for each operation state including a normal operation wherein operation members are operated continuously for a time equal to or shorter than a threshold, a start operation of starting a continuous operation wherein operation members are operated continuously for a time longer than the threshold, and a release operation of stopping the continuous operation. Detection data within the acquisition period and learning data for each operation state, which is stored in the storage means, are compared with each other to determine assigned data corresponding to the detection data as input data (S210). <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an input device, an input method, and an input program for determining input data based on detection data indicating a user's hand movement.

  2. Description of the Related Art Conventionally, an input device for a user to input data without using a keyboard is known. For example, Patent Document 1 proposes a data input device (input device) capable of inputting information by hand gestures. In a conventional input device, detection data output within a predetermined period from a sensor attached to a user's finger is compared with learning data registered in advance, and input data corresponding to finger movement is determined. Similar to a normal keyboard, the input device determines input data different from the case where each key is pressed separately when detection data corresponding to the case where a plurality of specific keys are pressed simultaneously is acquired. . For example, when the Shift key and the a key are pressed at the same time, the input device determines a capital letter “A” as input data.

JP-T-2006-503350

  However, the conventional input device has a problem that the operation state of the virtual key cannot be accurately grasped based on the detection data. More specifically, in the conventional input device, an operation of pressing a virtual key (hereinafter referred to as “normal operation”) and an operation of continuously pressing the virtual key (hereinafter referred to as “continuation operation”). ").” Was sometimes not accurately distinguished. For this reason, the conventional input device has a problem that appropriate input data according to the virtual key operation state cannot be determined.

  The present invention has been made to solve the above-described problems, and an object thereof is to provide an input device, an input method, and an input program that accurately determine input data according to the operation state of a virtual operation member. And

  In order to solve the above-described problem, an input device according to a first aspect is attached to a hand and generates a sensor that generates an electrical signal corresponding to the movement of the hand, and detects data including the electrical signal generated by the sensor. Output means for outputting as data, and an input device for determining input data corresponding to the movement of the hand operating one or more virtual operation members based on the detection data output by the output means A start operation for starting a normal operation that is an operation in which the operation of the operation member is continued for less than a threshold and an operation in which the operation is continued for a longer time than the threshold And an input timing that is an input timing that is a timing at which the hand operates the operation member in any one of the operation states for each operation state including the release operation for stopping the continuous operation. Storage control means for associating the detection data within the acquisition period of the first predetermined length with assignment data, which is data assigned to the operation member, in the storage means in association with each other. , Comparing the detection data output from the output means within the acquisition period and the learning data for each operation state stored in the storage means, and assigning the allocation data corresponding to the detection data, Determining means for determining the input data.

  The change pattern of the detection data when the user operates the operation member differs depending on the operation state. For example, when a virtual keyboard key is used as the operation member and an acceleration sensor is used as the sensor, even if the same key is operated, the change pattern of the detection data in the normal operation, the start operation, and the release operation Is different. The input device of the first aspect can accurately distinguish the normal operation, the start operation, and the release operation based on the detection data. That is, the input device can avoid that the continuation operation is not distinguished from the normal operation, or that the continuation operation start operation and the normal operation are erroneously recognized as two different normal operations. Therefore, even when an input operation in which a plurality of operation states are combined is executed, the input device can accurately determine input data corresponding to the operation state. In addition, the input device accurately distinguishes between normal operation and continuous operation. For example, even when the same key is operated, learning data can be set so that different input data is input for each operation state. It is. For example, the input device determines “a” as input data when the key “a” is operated in the normal operation, and the continuous operation is executed when the key “a” is operated in the continuous operation. It is also possible to determine the number of “a” corresponding to a given period as input data. As described above, the input device according to the first aspect can improve the operability by appropriately setting the learning data according to the operation state.

  In the input device according to the first aspect, the detection data output from the output means within the acquisition period is compared with the learning data stored in the storage means, and the start operation is detected before the detection is performed. A period detecting unit that detects a period until the release operation is detected as the continuous operation period, and the storage unit is configured to execute the normal operation during the continuous operation period detected by the period detecting unit. The learning data corresponding to the combination operation that is a case operation is stored as combination learning data, and the determination unit outputs the detection data output from the output unit during the continuous operation period detected by the period detection unit. And the combination learning data stored in the storage means, and the combination learning data corresponding to the detected data is stored in the storage means. If they are, it may determine the input data corresponding to the detected data. In this case, the input device can accurately determine the input data when the combination operation is executed based on the detection data and the combination learning data.

  In the input device according to the first aspect, the first predetermined length may be different between the normal operation, the start operation, and the release operation. In the input device in this case, it is possible to improve the determination accuracy of the input data by appropriately determining the first predetermined length for each operation state.

  In the input device according to the first aspect, the input device includes timing detection means for detecting the input timing based on a change in the detection data within a period of the second predetermined length output by the output means, and the determination means includes the determination means, When the input timing is detected by the timing detection means, a process for determining the input data is executed, a period from the start of the acquisition period to the input timing is set as L1, and the acquisition period ends from the input timing. L2 is a minimum time interval when the normal operation is continuously executed, and T1 is a minimum time interval from when the start operation of the combination operation is detected to when the release operation is detected. When T2 is set and A is the second predetermined length, the first predetermined length of the start operation is L1 <T1 and L2 <T2 and <L1 + L2 is satisfied, the first predetermined length of the release operation satisfies L1 <T2, L2 <T1, and A <L1 + L2, and the first predetermined length of the normal operation is L1 <T1 and L2 <T1 And A <L1 + L2. In this case, the input device performs processing executed by the input device as compared with a case where processing for determining input data (hereinafter referred to as determination processing) is performed regardless of whether or not input timing is detected. The load can be reduced. Since the first predetermined length is accurately set according to the operation state, the input device can improve the accuracy of determining the input data according to the operation state.

  The input method according to the second aspect includes an output step of outputting data including the electrical signal generated by a sensor attached to a hand and generating an electrical signal corresponding to the movement of the hand as detection data, and the output step An input method for determining input data corresponding to the movement of the hand that operates one or more virtual operation members based on the detection data output in step 1, wherein the operation member is operated continuously An operation including a normal operation that is an operation that is less than or equal to a threshold, a start operation that starts a continuous operation in which the operation of the operation member is continued longer than the threshold, and a release operation that stops the continuous operation An acquisition period including an input timing that is a timing at which the hand operates the operation member in any one of the operation states for each state, and is within the acquisition period of a first predetermined length A storage control step of associating the detection data with allocation data, which is data allocated to the operation member, and storing it in the storage means as learning data, and the detection data within the acquisition period output from the output step And a determination step of comparing the learning data for each operation state stored in the storage means and determining the allocation data corresponding to the detection data as the input data. The input method of the second aspect can obtain the same operational effects as the input device of the first aspect.

  The input program of the third aspect is an input program for causing a computer to function as various processing means of the input device of the first aspect. The input program according to the third aspect can exhibit the operational effects as various processing means of the input device according to the first aspect by being executed by a computer.

2 is a schematic diagram showing a physical configuration of the input device 1. FIG. 3 is a block diagram showing an electrical configuration of a data collection device 2. FIG. 3 is a block diagram showing an electrical configuration of a data determination device 4. FIG. It is explanatory drawing of the window information memorize | stored in the window information storage area 383. FIG. It is explanatory drawing of the identification window 300 set to the detection data pattern 220, and the input determination window 400. FIG. It is explanatory drawing of the input judgment window 400 set to the detection data pattern 220. FIG. It is explanatory drawing which illustrates the detection data pattern 231 containing the detection data 232 and detection data 236 when a normal operation is performed continuously. It is explanatory drawing which illustrates the detection data pattern 240 containing the detection data 243 when the continuation operation of combination operation is performed. It is explanatory drawing of the learning data storage area 384. FIG. It is explanatory drawing of the combination learning data storage area 385. FIG. 5 is a flowchart of detection processing executed by the data collection device 2 and the data collection device 3. 4 is a flowchart of main processing executed by the data determination device 4. It is a flowchart of the input mode process performed by the main process of FIG. Detection data pattern 251 (left hand, little finger, Z axis), detection data pattern 252 (right hand, middle finger, Z axis) and detection in specific example 2 in which lowercase A and capital letters K and L are sequentially input It is explanatory drawing which illustrates the data pattern 253 (right hand, ring finger, Z-axis). It is a flowchart of the composite process performed in the input mode process of FIG.

  DESCRIPTION OF EMBODIMENTS Hereinafter, an input device 1 according to an embodiment embodying the present invention will be described in order with reference to the drawings. These drawings are used for explaining the technical features that can be adopted by the present invention, and the configuration of the apparatus and the flowcharts of various processes described are not intended to be limited to the drawings. This is just an illustrative example.

  First, the physical configuration of the input device 1 will be described with reference to FIG. The input device 1 determines the input data by detecting the movement of the user's hand operating the virtual operation member. In particular, the input device 1 according to the present embodiment detects an action of a hand (finger) operating a virtual keyboard key, and determines input data based on the key corresponding to the operation. As shown in FIG. 1, the input device 1 includes a data collection device 2, a data collection device 3, and a data determination device 4. The data collection device 2 is attached to the user's right hand. The data collection device 3 is attached to the left hand of the user. Each of the data collection device 2 and the data collection device 3 transmits detection data to the data determination device 4. The data determination device 4 is arranged at a position that can be confirmed by the user, and determines input data based on detection data transmitted from the data collection device 2 and the data collection device 3.

  A physical configuration of the data collection device 2 and the data collection device 3 will be described. The physical configurations of the data collection device 2 and the data collection device 3 are basically the same. Taking the data collection device 2 as an example, the physical configuration of the data collection device 2 and the data collection device 3 will be described. As shown in FIG. 1, the data collection device 2 includes a glove unit 23 (a glove unit 33 in the data collection device 3), a sensor 21, a housing 22, and a harness 24. Hereinafter, each member with which the data collection device 2 is provided is demonstrated.

  The glove part 23 is a glove for the right hand worn on the user's right hand. Five sensors 21 are provided corresponding to each finger of the hand to which the data collection device 2 is attached, and detects the movement of the finger to which the sensor 21 is attached. Each sensor 21 is attached to the fingertip portion of the surface on the back side of the glove part 23. The sensor 21 only needs to have a function of detecting movement of a hand, particularly a finger, and for example, either an acceleration sensor or a gyro sensor is used. The sensor 21 of this embodiment is an acceleration sensor. When the power of the data collection device 2 is turned on, the sensor 21 always outputs an electrical signal indicating the acceleration (G) of the finger to which the sensor 21 is attached for each detection direction (axis), as will be described later. (See FIG. 2). In the present embodiment, three directions (three axes) including an x-axis direction (x-axis), a y-axis direction (y-axis), and a z-axis direction (z-axis) in the orthogonal coordinate system are set as detection directions (axes). Has been.

  The housing 22 has a rectangular parallelepiped shape, and is attached to the wrist portion on the back surface of the glove part 23. The housing | casing 22 accommodates the control part 25 (refer FIG. 2). The harness 24 electrically connects the sensor 21 and the control unit 25. A signal output from the sensor 21 is transmitted to the control unit 25 via the harness 24.

  Next, the physical configuration of the data determination device 4 will be described. As shown in FIG. 1, the data determination device 4 includes a housing 36, a display unit 39, and an input unit 40. The housing 36 has a rectangular parallelepiped shape and houses a control unit 37 (see FIG. 3) described later. The display unit 39 is a liquid crystal display having a rectangular shape in plan view. The display unit 39 displays various information including the operation mode of the input device 1, input data, and instructions. The input unit 40 is a button used when the user sets the operation mode of the input device 1.

  Next, the electrical configuration of the input device 1 will be described. The electrical configurations of the data collection device 2 and the data collection device 3 are the same. Taking the data collection device 2 as an example, the electrical configuration of the data collection device 2 and the data collection device 3 will be described.

  As shown in FIG. 2, the data collection device 2 includes a control unit 25 and a sensor 21. The control unit 25 includes a CPU 41, a ROM 42, a RAM 43, an A / D converter 45, an RF module 46, an antenna 47, and a bus 44. The CPU 41 is connected to each of the ROM 42 and the RAM 43 via the bus 44. Hereinafter, each structure with which the control part 25 is provided is demonstrated.

  The CPU 41 is responsible for the main control of the data collection device 2 and executes various calculations and processes related to determination of input data according to various programs stored in a program storage area (not shown) of the ROM 42. The program may be stored in an external storage device such as a flexible disk.

  Although not shown, the ROM 42 includes a plurality of storage areas including a program storage area and a setting storage area. Various programs including a detection program are stored in the program storage area. The detection program is a program for executing detection processing described later. In the setting storage area, initial settings and parameters of the data collection device 2 are stored. The RAM 43 is an arbitrarily readable / writable storage device, and stores the calculation result of the CPU 41 and the like.

  The A / D converter 45 converts the electrical signal (analog signal) output from the sensor 21 into a digital signal. The A / D converter 45 is connected to the sensor 21 and the CPU 41. The antenna 47 transmits and receives data between the data collection device 2 and the external device. In this embodiment, the antenna 47 transmits detection data to the data determination apparatus 4 by radio | wireless, for example. The RF module 46 is connected between the antenna 47 and the CPU 41 and controls modulation.

  Next, the electrical configuration of the data determination device 4 will be described. As shown in FIG. 3, the control unit 37 of the data determination device 4 includes a CPU 101, a ROM 102, a RAM 103, a flash memory 38, an RF module 105, an antenna 106, a display unit 39, and an input unit 40. . The CPU 101 is connected to the ROM 102, the RAM 103, the flash memory 38, and the bus 104. Hereinafter, each structure with which the data determination apparatus 4 is provided is demonstrated.

  The CPU 101 manages the main control of the data determination device 4 and executes various operations and processes related to determination of input data according to various programs stored in a program storage area (not shown) of the ROM 102. The program may be stored in an external storage device such as a flexible disk.

  Although not shown, the ROM 102 includes a plurality of storage areas including a program storage area. Various programs including the main program are stored in the program storage area. The main program is a program for executing main processing for determining input data based on the detection data output from the data collection device 2 and the data collection device 3. The acquisition program is a program for executing an acquisition process described later. The RAM 103 stores calculation results and the like calculated by the CPU 101.

  The flash memory 38 includes a detection data storage area 381, an operation mode storage area 382, a window information storage area 383, a learning data storage area 384, a combination learning data storage area 385, and other information storage areas 386. A plurality of storage areas are provided. In the detection data storage area 381, the electrical signal output from the sensor 21 and the time stamp given to the electrical signal are stored as detection data. The time stamp included in the detection data indicates a time stamp clock value when the electric signal output from the sensor 21 is acquired by the control unit 25. The detection data is stored for each detection condition in which the type of the hand and finger to which the sensor 21 is attached and the detection direction (axis) are combined. The operation mode storage area 382 stores the operation mode of the input device 1. In the present embodiment, either the learning mode or the input mode can be set as the operation mode. When the learning mode is set as the operation mode, processing for updating the learning data storage area 384 is executed. When the input mode is set as the operation mode, processing for determining the input data is executed based on the detection data. The window information storage area 383 stores a condition for setting a window when a main process described later is executed. Details of the window information storage area 383 will be described later. The learning data storage area 384 stores learning data. In the combination learning data storage area 385, combination learning data is stored. Details of the learning data and the combination learning data will be described later. The other information storage area 386 stores other information such as conditions that are referred to when the main process is executed.

  The antenna 106 transmits and receives data wirelessly between the data determination device 4 and the external device. The external devices include a data collection device 2, a data collection device 3, and an external device such as a PC that is an output destination of input data. The RF module 105 is connected to the CPU 101 and the antenna 106 and manages modulation / demodulation control. The display unit 39 is connected to the CPU 101 and is controlled by the CPU 101. The input unit 40 is connected to the CPU 101.

  Next, the window information storage area 383 will be described with reference to FIGS. As shown in FIG. 4, window setting conditions are stored in the window information storage area 383 for each operation state. The window defines a detection data acquisition period set in a process of determining input data based on detection data acquired by the data determination device 4 over time in a main process described later. In the present embodiment, two types of windows, an input determination window 400 illustrated in FIG. 5 and an identification window 300, are set in the main process. The input determination window 400 defines a detection data acquisition period used for processing for detecting the input timing 221. As shown in FIG. 6, in this embodiment, when the difference between the maximum value and the minimum value of detection data in the input determination window 400 (shown by the arrow 226) is a predetermined value (for example, 1000 mg) or more, the input determination window The latest data acquisition time in 400 (the right end of the input determination window 400) is set as the input timing. The predetermined value is preferably determined in consideration of the change over time of the detection data when the input operation is executed and the change over time of the detection data when the input operation is not executed. The identification window 300 defines a detection data acquisition period (acquisition range) used for processing for determining input data. As shown in FIG. 5, the input timing 221 and the input determination window 400 are included in the range defined by the identification window 300.

  In the present embodiment, a normal operation, a continuation operation, and a combination operation are set as operation states. The normal operation refers to an operation in which the time for which a virtual key is continuously pressed is equal to or less than a threshold value (for example, 0.5 msec). The continuous operation is an operation in which the time for which the virtual key is continuously pressed is longer than a threshold value. The continuation operation includes a start operation and a release operation. The start operation refers to an operation for starting a continuation operation. The release operation is an operation for stopping the continuous operation. The combination operation is an operation in which a normal operation is executed during a period in which a continuation operation is being executed (hereinafter referred to as “continuation operation period”). The continuous operation period is a period from when the start operation is detected to when the release operation is detected. In the present embodiment, the pattern of detection data when the continuation operation is executed hardly changes during the period excluding before and after the start operation and before and after the release operation. For this reason, the input device 1 specifies the duration by detecting the start operation and the release operation. In addition, the said threshold value should just set suitably the value which distinguishes normal operation and continuation operation. When a virtual keyboard operation is detected as in the input device 1, the threshold value that defines the normal operation (keystroke) and the continuous operation is preferably set for each user because there are individual differences. Further, in the present embodiment, it is assumed that the key operation is performed with different hands for the continuation operation of the combination operation and the normal operation. In the present embodiment, the above operations are collectively referred to as an input operation.

  In general, if the width of the identification window is within a certain range, the longer the width, that is, the greater the number of detection data used for the process of determining input data, the higher the accuracy of determining input data. On the other hand, if the width of the identification window exceeds a certain range, detection data corresponding to a plurality of input operations may be included in the same identification window. In such a case, the input data Determining accuracy deteriorates. On the other hand, as shown in FIGS. 7 and 8, the detection data 232 when the input operation is executed, the detection data 241 when the start operation is executed, and the detection data 242 when the release operation is executed. And the change patterns are different from each other. In particular, in the input timing 233 when the input operation is performed, the input timing 245 when the start operation is performed, and the input timing 246 when the release operation is performed, detection data before and after each input timing is detected. The time required to converge differs for each operation state.

  For this reason, in this embodiment, the setting conditions L1 and L2 of the identification window 300 are defined for the normal operation, the start operation, and the release operation, respectively. As illustrated in FIG. 5, L1 represents a period from the start 222 of the identification window 300 to the input timing 221, and L2 represents a period from the input timing 221 to the end 223 of the identification window 300. L1 and L2 are preferably determined as follows. As shown in FIG. 7, the minimum time interval when the normal operation is continuously executed, that is, the minimum time interval between the input timing 233 and the input timing 234 detected when the normal operation is continuously executed. Is T1. As shown in FIG. 8, in the combination operation detection data 243, the minimum time interval from when the start operation input timing 245 is detected until the release operation input timing 246 is detected is T2. Let A be the width of the input determination window 400.

  As shown in FIG. 5, it is preferable that L1 and L2 of the start operation satisfy L1 <T1, L2 <T2, and A <L1 + L2. Similarly, L1 and L2 of the release operation satisfy L1 <T2, L2 <T1, and A <L1 + L2, and L1 and L2 of the normal operation are L1 <T1, L2 <T1, and A <L1 + L2. It is preferable to satisfy. In this way, by setting L1, L2, and A, it is possible to prevent noise from being included in the identification window while securing the number of detection data necessary for determining input data. . More preferably, L1 and L2 do not include the input timing 224 outside the identification window 300 or fluctuations before and after the input timing 225. For example, in FIG. 7, it is more preferable that T1 ′ is used instead of T1 in the above relational expression so that the window used when detecting the input timing 234 does not include the fluctuation of the period 235. . If L1, L2, and A are set in this way, it is possible to more reliably avoid noise from being included in the identification window, so that the accuracy of determining the input timing can be further improved. it can. In the present embodiment, A is set to the same value of 25 msec regardless of the operation state. T1 and T2 are preferably set for each user because of individual differences, but may be set values common to a plurality of users. In this embodiment, T1 is 100 msec and T2 is 70 msec. As shown in FIG. 5, L1 and L2 in each operation state are values that satisfy the above relational expression.

  Next, the learning data storage area 384 and the combination learning data storage area 385 will be described with reference to FIG. 9 and FIG. The learning data is data in which the allocation data is associated with the feature amount of the detection data when the key corresponding to the allocation data is operated. The feature amount of the detection data is five average values calculated by the following procedure based on the detection data in the identification window 300. First, as shown in FIG. 5, the range of the identification window 300 is equally divided into five ranges having a length indicated by an arrow 301. Next, an average value of the detection data is calculated for each divided range. As shown in FIG. 9, the learning data of this embodiment is stored for each operation state and for each detection condition. The detection condition is a condition in which the hand type, the finger type, and the detection direction are combined. The hand types include a right hand and a left hand. The finger types include thumb, index finger, middle finger, ring finger, and little finger.

  The input device 1 of this embodiment can specify the operation state and the allocation data based on the detection data by storing learning data for each operation state. As will be described later, the input device 1 determines the input data by combining the operation state specified based on the detection data and the allocation data. When the operation state is a normal operation, allocation data included in the learning data is determined as input data.

  On the other hand, when the operation state is a combination operation, allocation data is specified based on the learning data, and allocation data included in the combination learning data is determined as input data based on the combination of the specified allocation data. . For this reason, the allocation data when the combination operation is executed is stored in a combination learning data storage area 385 provided separately from the learning data storage area 384. As shown in FIG. 10, in the combination learning data storage area 385, allocation data is associated with each combination operation. The key type when the combination operation is performed is specified based on the learning data stored in the learning data storage area 384 and the detection data.

  Next, processing executed by the input device 1 will be described. First, processing executed by the data collection device 2 and the data collection device 3 will be described. The processes executed by the data collection device 2 and the data collection device 3 are basically the same. In the data collection device 2 and the data collection device 3, when the power is turned on, a start instruction is transmitted to the data determination device 4, and a plurality of processes including an output process and a detection process are executed. . The data collection device 2 and the data collection device 3 transmit an end instruction to the data determination device 4 when the power is turned off. The output process and the detection process are each executed by the CPU 41 in accordance with a program stored in the ROM 42 of FIG.

  In the output process, an electrical signal indicating acceleration (mg) in three types of detection directions (x-axis direction, y-axis direction, z-axis direction) is controlled from the sensor 21 during the period when the power of the data collection device 2 is turned on. Always output to the unit 25. In the detection process, the electrical signal output in the output process is detected at predetermined intervals by the CPU 41 (see FIG. 2) included in the control unit 25. The detection process will be described with reference to FIG. As shown in FIG. 11, in the detection process, first, a clock (not shown) is referred to and it is determined whether or not it is a detection time (S5). In S5, it is determined that it is the detection time when a predetermined period (for example, 1 msec) has elapsed since the previous detection. The predetermined period is stored in the other information storage area 386.

  When it is not the detection time (S5: No), it waits until the detection time is reached. When it is the detection time (S5: Yes), the electrical signal output from the sensor 21 is acquired, and a time stamp is given to the acquired electrical signal (S10). The electrical signals acquired in S10 are 15 types of electrical signals obtained by combining five types of fingers and three types of detection directions. The time stamp given to each electric signal is, for example, a time stamp clock value when the electric signal is acquired. Next, the electrical signal acquired in S10 and the time stamp given to the electrical signal are output to the data determination device 4 as detection data (S15). The detection data transmitted in S15 includes information indicating detection conditions. Next, when the power of the data collection device 2 is ON (S20: No), the process returns to S5, and when the power is OFF (S20: Yes), the detection process ends.

  Next, processing executed by the data determination device 4 will be described. The data determination device 4 is activated when the start instruction transmitted from the data collection device 2 or the data collection device 3 is acquired. The data determination device 4 starts a plurality of processes including an acquisition process and a main process after activation. The data determination device 4 performs various processes when an end instruction is transmitted from either the data collection device 2 or the data collection device 3 or when the data determination device 4 is powered off. The termination process is executed separately from the above process. The acquisition process and the main process are each executed by the CPU 101 in accordance with a program stored in the ROM 102 of FIG. In the acquisition process, the detection data output from the data collection device 2 or the data collection device 3 is received, and the received detection data is stored in the detection data storage area 381. The data determination device 4 stores detection data for each detection condition.

  In the main process, a process according to the operation mode of the input device 1 is executed. The main process will be described with reference to FIGS. First, an outline of the main process will be described. The input device 1 of this embodiment determines input data based on the detection data transmitted from the data collection device 2 and the data collection device 3 in the main process. The input data of the present embodiment is a character or symbol associated with each key of the virtual keyboard in advance.

  As shown in FIG. 12, in the main process, an initialization process is first executed (S105). In S <b> 105, for example, various setting values for executing the main process are read from the ROM 102 or the flash memory 38 in FIG. 3. The read setting values are stored in the RAM 103. Next, the input mode is set as the operation mode, and the set operation mode is stored in the operation mode storage area 382 (S110).

  Next, it is determined whether or not the operation mode of the input device 1 is a learning mode (S115). In the learning mode, processing for updating the feature amount included in the learning data in the learning data storage area 384 is executed. When the user operates the input unit 40 to input a predetermined instruction, the learning mode is set as the operation mode. The process when the operation mode is not the learning mode (S115: No) will be described later. When the operation mode is the learning mode (S115: Yes), processing in the learning mode is executed.

  In the process in the learning mode, first, an operation state is set, and the set operation state is stored in the RAM 103 of FIG. 3 (S120). S120 is a process for updating learning data for each operation state. In S120, either the normal operation or the continuous operation is set as the operation state. Next, the allocation data is displayed on the display unit 39 (S125). In S125, in this embodiment, the key type (for example, “a”) of the virtual keyboard is displayed as the allocation data. After confirming the display unit 39, the user performs an operation when pressing the key “a” shown on the display unit 39. Next, window conditions corresponding to the operation state set in S120 are set, and the set window conditions are stored in the RAM 103 (S130). A specific example 1 is a case where the normal operation is set as the operation state in S120. In the first specific example, the window information storage area 383 in FIG. 4 is referred to, 80 msec is set as L1, 80 msec is set as L2, and 25 msec is set as A. When a continuous operation is set as the operation state, a window condition for a start operation is first set in S130, and a window condition for a release operation is set in S130 executed after the process of S190.

  Next, a predetermined number of pieces of detection data are acquired in order from the newest acquisition order, and the acquired detection data is stored in the RAM 103 of FIG. 3 (S135). S135 is processing for setting an identification window. The number of detection data acquisitions in S135 is determined according to the width (L1 + L2) of the identification window set in S125 and the detection data acquisition frequency J in the acquisition process. In specific example 1, when the acquisition interval J is 1 msec, the number of detection data acquisitions is 160. Next, an input determination window is set in the detection data acquired in S135 (S140).

  Next, based on the detection data in the input determination window set in S140, it is determined whether an input operation has been executed (S145). In S145, first, the input determination window set in S140 is equally divided into a predetermined number of periods, and an average value of detection data (hereinafter referred to as “period average value”) is calculated for each divided period. Next, the input operation is executed when the difference between the maximum value and the minimum value of the period average value is greater than or equal to the threshold for the detection data of any detection condition, that is, when the input timing is detected. It is judged that it was done. Comparing the difference between the maximum value and the minimum value of the average period value when the input operation is performed, generally, the difference between the maximum value and the minimum value of the average period value in the z-axis direction of the finger that has performed the input operation Is the largest. Therefore, in this embodiment, the input timing is detected based on the detection data of the finger that has performed the input operation. The threshold value is appropriately determined in consideration of the change over time of the detected data when the input operation is executed and the change of the detected data when the input operation is not executed, and is stored in the flash memory 38. In this embodiment, 1000 mg is set as the threshold value. If the input operation has not been executed (S145: No), the process returns to S135. As shown in FIG. 6, the input determination window 400 is sequentially shifted from the left side to the right side by repeatedly executing S135 and S140.

  When an input operation is executed in S145 (S145: Yes), the detection data (80 + 80 = 160 msec data) included in the identification window set in S130 is read from the RAM 103 (S150) and read. A feature amount is calculated based on the detected data (S155). In S155, the period of the identification window 300 is equally divided into five periods, and an average value is calculated as the feature amount of the detected data for each divided period. Next, the feature amount calculated in S155 and the allocation data displayed in S125 are associated with each other and stored as learning data in the learning data storage area 384 (S160).

  Next, the operation mode storage area 382 is referred to, and it is determined whether or not the operation mode is the input mode (S165). When the user operates the input unit 40 to input a predetermined instruction, the operation mode is changed from the learning mode to the input mode. In S165, when the operation mode is the input mode (S165: Yes), when the power of the data determination device 4 is not turned off (S215: No), the process returns to S115. When the power of the data determination device 4 is turned off (S215: Yes), the main process ends. When the operation mode is not the input mode (S165: No), the RAM 103 is referred to, and it is determined whether or not the operation state set in S120 is a continuous operation (S180). If it is a continuation operation (S180: Yes), it is determined whether learning data for the release operation has been acquired (S185). When the operation state is set to the continuous operation, first, the learning data of the key start operation displayed in S125 is updated, and then the learning data of the key release operation is updated. If learning data for the release operation has not been acquired (S185: No), an instruction for prompting the release operation is displayed on the display unit 39 (S190), and the process returns to S130. In S130 after S190, a window condition for the release operation is set.

  When the operation state set in S120 is not a continuous operation (S180: No), or when learning data of a release operation has already been acquired (S185: Yes), all key types of the virtual keyboard are S125. It is determined whether or not it is displayed at (S195). If there is a key type not displayed in S125 (S195: No), the process returns to S125, and the next key type is displayed (S125). When all the key types are displayed (S195: Yes), it is determined whether all the operation states are set in S120 (S200). If there is an unset operation state (S200: No), the process returns to S120, and the next operation state is set. In S165, when the operation mode is the input mode (S165: Yes) or when all operation states are set (S200: Yes), the input mode is set as the operation mode (S205). Next, when the power of the data determination device 4 is not turned off (S215: No), the process returns to S115. When the power of the data determination device 4 is turned off (S215: Yes), the main process ends.

  In S115, when the operation mode is not the learning mode (S115: No), the input mode process is executed (S210). The input mode process will be described with reference to FIGS. As an example, assume a specific example 2 in which a lowercase letter A and uppercase letters K and L are input in this order as shown in FIG. As shown in FIG. 14, in the specific example 2, the A key is normally operated at the timing t1, and the Shift key is continuously operated during the period from the timing t2 to the timing t5. Furthermore, in the specific example 2, the K key is normally operated at the timing t3, and the L key is normally operated at the timing t4.

As shown in FIG. 13, in the input mode process, first, the window information storage area 383 is referred to and the window information is read (S305). Next, a predetermined number of detection data items are acquired in the order from the newest acquisition order, and the acquired detection data items are stored in the RAM 103 (S310).
The number of detection data acquired in S310 is defined by the width A of the input determination window and the detection data acquisition interval J in the acquisition process (S310). When the length A of the input determination window is 25 msec and the acquisition interval J is 1 msec, 25 detection data are acquired. The detection data in the input determination window is acquired by the process of S310.

  Next, based on the detection data acquired in S310, it is determined whether or not an input operation has been executed (S315). The process of S315 is the same as the process of S145. If the input operation has not been executed (S315: No), the process returns to S310. When the input operation is executed (S315: Yes), the variable C for counting the elapsed time is incremented, and the incremented variable C is stored in the RAM 103 (S320). The initial value of variable C is zero.

  Next, it is determined whether the elapsed time from the input timing detected in S315 is equal to or longer than the start operation L2 (hereinafter simply referred to as “S_L2”) (S330). In S330, whether or not the elapsed time from the input timing detected in S315 is equal to or longer than S_L2 is determined based on the variable C stored in the RAM 103. When the elapsed time is not equal to or longer than S_L2 (S330: No), the process returns to S320. When the elapsed time is equal to or longer than S_L2 (S330: Yes), a start operation identification window (hereinafter referred to as “window P”) is set, and the set window P is stored in the RAM 103 (S335). Next, a predetermined number of detection data items are acquired in the order from the newest acquisition order, and the acquired detection data items are stored in the RAM 103 (S340). It is defined by the width (L1 + L2) of the window P and the detection data acquisition interval J in the acquisition process. As shown in FIG. 4, when the width of the window P is 140 msec and the acquisition interval J is 1 msec, 140 detection data are acquired in S340.

  Next, the detection data acquired in S340 is processed, and the processed detection data is stored in the RAM 103 (S345). The process of S345 is the same as S155, and based on the detection data acquired in S340, five average values are obtained for each 28 msec period obtained by dividing the period 140 msec of the window P into five equal parts. Next, the learning data storage area 384 is referred to, and the detection data processed in S345 is compared with the learning data of the start operation (S350). Next, based on the comparison result in S350, it is determined whether or not the input operation detected in S315 is a start operation (S355). When the difference between the learning data of any start operation and the detection data processed in S345 is less than a predetermined value, it is determined that the start operation has been executed. The predetermined value in S355 is appropriately determined in consideration of conditions including variations in detected data when the same operation is executed and differences between the detected data when other operations are executed, and the flash memory 38 Is remembered.

  When the input operation detected in S315 is a start operation (S355: Yes), allocation data corresponding to the input operation detected in S315 is stored in the RAM 103 (S360). The allocation data corresponding to the input operation detected in S315 is allocation data included in the learning data of the start operation whose difference from the detection data processed in S345 is less than a predetermined value. Next, a composite process described later is executed (S365).

  When the input operation detected in S315 does not correspond to the learning data of any start operation (S355: No), a process for determining whether or not the input operation detected in S315 is a normal operation is executed. The Specifically, first, the variable C is incremented, and the incremented variable C is stored in the RAM 103 (S370). In the present embodiment, since L2 (hereinafter simply referred to as “N_L2”) for normal operation is set to a value larger than S_L2, the processes of S370 and S375 are executed. Next, it is determined whether the elapsed time from the input timing detected in S315 is N_L2 or more (S375). In S375, the determination is made based on the variable C stored in the RAM 103.

  If the elapsed time is not N_L2 or more (S375: No), the process returns to S370. When the elapsed time is N_L2 or more (S375: Yes), a normal operation identification window (hereinafter referred to as “window Q”) is set, and the set window Q is stored in the RAM 103 (S385). Next, a predetermined number of detection data items are acquired in the order from the newest acquisition order, and the acquired detection data items are stored in the RAM 103 (S390). The predetermined number in S390 is defined by the width A of the window Q and the detection data acquisition interval J in the acquisition process. When the width A of the window Q is 160 msec and the acquisition interval is 1 msec, 160 pieces of detection data are acquired in the order from the new acquisition order in S390.

  Next, the detection data acquired in S390 is processed, and the processed detection data is stored in the RAM 103 (S395). The process of S395 is the same as that of S155, and five average values are calculated based on the detection data acquired in S390. Next, the learning data storage area 384 is referred to, and the detection data processed in S395 is compared with the learning data for normal operation (S400). Next, based on the comparison result in S400, it is determined whether or not the input operation detected in S315 is a normal operation (S405). When the difference between any of the normal operation learning data and the detection data processed in S395 is less than a predetermined value, it is determined that the normal operation has been executed.

  When the input operation detected in S315 is a normal operation (S405: Yes), input data corresponding to the input operation detected in S315 is determined, and the determined input data is stored in the RAM 103 (S410). The input data determined in S410 is allocation data included in the learning data whose difference from the detection data processed in S395 is less than a predetermined value. In the second specific example, the input timing when the A key is normally operated at the timing t1 in FIG. 14 is detected (S315: Yes), and the feature amount (average value) of the detected data is compared with the learning data of the normal operation. Based on the result, “a” is determined as input data (S410). Next, the input data determined in S410 is transmitted via the antenna 106 to an external device (not shown) such as a PC (S415).

  When the input operation detected in S315 does not correspond to any normal operation learning data (S405: No), or after S415, the power of the data determination device 4 is not turned off (S420: No), the process returns to S305. When the power of the data determination device 4 is turned off (S420: Yes), the input mode process (main process) ends.

  In concrete example 2, the input timing of the start operation of the “Shift key” is detected at the timing t2 in FIG. 14 (S315: Yes). In this case, it is determined in S355 that the operation is a start operation (S355: Yes), and the allocation data “Shift” corresponding to the processed detection data is stored in the RAM 103 (S360). In S360, allocation data included in the learning data whose difference from the detection data processed in S345 is less than a predetermined value is stored in the RAM 103.

  Next, a composite process is executed (S365). The composite process is executed during the continuous operation period. In the composite process, a process for determining input data corresponding to the combination operation is executed. The composite process will be described with reference to FIG. As shown in FIG. 15, first, a predetermined number of detection data is acquired in the order of new acquisition order, and the acquired detection data is stored in the RAM 103 (S500). The predetermined number in S500 is the same as S390. Next, when the power of the data determination device 4 is turned off (S505: Yes), the composite process (main process) ends. If the power of the data determination device 4 is not turned off (S505: No), an input determination window is set in the detection data acquired in S500, and the set input determination window is stored in the RAM 103 (S510). ). The process of S510 is the same as S140.

  Next, it is determined whether an input operation has been executed based on the detection data in the input determination window (S515). The process of S515 is the same as that of S145. If the input operation has not been executed (S515: No), the process returns to S500. When the input operation is executed (S515: Yes), the detection data acquired in S500 is read from the RAM 103 of FIG. 3 (S520), and the read detection data is processed (S525). The process of S525 is the same as that of S155, and five average values are calculated based on the detection data acquired in S520.

  Next, the learning data storage area 384 is referred to, and the detection data processed in S525 is compared with the learning data of the normal operation and the release operation (S530). Next, as in S405, it is determined whether or not the input operation detected in S515 is a normal operation (S535). When the input operation detected in S515 is a normal operation (S535: Yes), a combination operation in which the continuous operation started from S315 and the normal operation detected in S515 are combined is combined learning data storage area 384. It is determined whether or not it is stored in (S550). The combination operation stored in the combination learning data storage area 385 is referred to as “effective combination”. When the combination operation is not an effective combination (S550: No), the process returns to S500. In this case, input data corresponding to the combination operation is not determined. In the second specific example, as shown in FIG. 14, the input timing of the normal operation of the “L key” is detected at the timing t3 during the continuous operation period of the “Shift key” (S535: Yes), and it is an effective combination. It is judged (S550: Yes). In this case, capital L, which is the allocation data included in the combination learning data that matches the combination operation “Shift + L”, is determined as input data (S555). The determined input data is stored in the RAM 103. Next, the input data determined in S555 is transmitted via the antenna 106 to an external device (not shown) such as a PC (S560). Next, the process returns to S500.

  In S535, when the input operation detected in S515 does not correspond to any normal operation learning data (S535: No), it is determined whether or not the input operation detected in S515 is a release operation (S540). ). When the difference between the learning data of the release operation and the detection data processed in S525 is less than a predetermined value, it is determined that the release operation has been executed. The release operation detected in S540 is an operation corresponding to the start operation detected in S315 of FIG. When the input operation detected in S515 does not correspond to the release operation corresponding to the start operation detected in S315 (S540: No), the process returns to S500. When the input operation detected in S515 is a release operation (S540: Yes), the composite process ends, and the process returns to the input mode process of FIG. In the second specific example, the input timing of the release operation of the “Shift key” is detected at timing t5 in FIG. 14 (S515: Yes), and it is determined that the release operation has been executed in S540 (S535: No, S540: Yes). ).

  As described above, the input device 1 determines the input data by detecting the movement of the user's hand. S15 in FIG. 11 corresponds to the output process of the present invention. In S15, the CPU 41 that outputs the detection data to the data determining device 4 functions as the output means of the present invention. S160 in FIG. 12 corresponds to the storage control step of the present invention, and the CPU 101 that executes the processing of S160 functions as the storage control means of the present invention. The flash memory 38 that stores learning data including combination learning data corresponds to the storage means of the present invention. S410 in FIG. 13 and S555 in FIG. 15 correspond to the determination step of the present invention, and the CPU 101 that executes the processes of S410 and S555 functions as the determination unit of the present invention. The CPU 101 that detects the input timing in S315 and S515 functions as the timing detection means of the present invention.

  According to the input device 1, the normal operation, the start operation, and the release operation can be accurately distinguished based on the detection data. That is, the input device 1 can avoid that the continuous operation is not distinguished from the normal operation, or that the start operation and the release operation of the continuous operation are erroneously recognized as two different normal operations. Therefore, the input device 1 can accurately determine the input data according to the operation state even when an input operation in which a plurality of operation states are combined is executed. That is, the input device 1 can determine which virtual key is operated in which operation state based on the detection data. Furthermore, the input device 1 can accurately determine the input data when the combination operation is executed based on the detection data and the combination learning data. For example, even when the same normal operation (for example, “A” keystroke) is performed as in the above-described specific example 2, different input data (for example, lowercase letters) is used depending on whether or not it is performed during the continuous operation period. "A" and capital letter "A") can be determined.

  Further, the input device 1 can improve the accuracy of determining input data by appropriately determining the width of the identification window for each operation state. The input device 1 executes the determination process when the input timing is detected in S315 and S515 of the main process. For this reason, the input apparatus 1 can reduce the load of the process performed by the input apparatus 1 compared with the case where the determination process is performed regardless of whether the input timing is detected. Since the input device 1 accurately sets the width of the identification window and the correspondence between the identification window and the input determination window according to the operation state, compared to the case where the window condition is set regardless of the operation state, The accuracy of determining input data can be improved.

  In addition, this invention is not limited to said embodiment, A various change is possible. For example, the following modifications (a) to (e) may be added as appropriate.

  (A) The configuration of the input device 1 can be changed as appropriate. For example, the type of sensor provided in the input device 1, the mounting position of the sensor, and the number of sensors can be changed as appropriate. More specifically, the input device may install an acceleration sensor as a sensor at the fingertip of each finger and the third joint of each finger. For example, the input device 1 is separate from the data collection device 2, the data collection device 3, and the data determination device 4, but is not limited thereto. For example, at least one of the data collection device 2 and the data collection device 3 may have a function as the data determination device 4. Further, for example, the data collection device 2 and the data collection device 3 and the data determination device 4 are connected wirelessly, but may be connected by wire.

  (B) In the above embodiment, a virtual keyboard key is assumed as a virtual operation member, but other virtual operation members may be assumed. For example, virtual operation members exemplified in the following (b-1) to (b-3) may be assumed.

  (B-1) As a virtual operation member, a button and a mouse body included in the mouse may be used. Examples of the operation state in this case include a button click operation, a double click operation, and a continuation operation. An operation state includes a movement operation of the mouse body. As the combination operation, for example, a drag for moving the mouse body while continuously operating the buttons can be cited.

  (B-2) Sticks and buttons included in the joystick controller may be used as virtual operation members. Examples of the operation state in this case include a stick tilting operation in addition to a normal button operation and a continuous operation. Examples of the combination operation include an operation of tilting the stick while pressing a button.

  (B-3) A touch panel may be used as a virtual operation member. Examples of the operation in this case include a normal operation of the touch panel and a continuous operation.

  (C) Various processes executed by the input device can be changed as appropriate. For example, the following modifications (c-1) to (c-5) may be added.

  (C-1) For example, in the input mode process of FIG. 13, the input data may be determined by the following procedure. In S340, a predetermined number (eg, 100) of detection data is acquired. In S345, the feature amount of the detection data acquired in S340 is calculated. Examples of the feature amount include a variance value and an amplitude. In S350, the feature amount calculated in S345 is compared with the learning data, and input data corresponding to the detection data is determined. The learning data in this case is a feature amount previously associated with the allocation data. Further, for example, the input system of the above embodiment has determined the input data by comparing each of the detection patterns with the comparison value for the detection data of each detection condition, but the detection data of a specific detection condition, The input data may be determined by comparing with the learning data. In addition, for example, the detection data acquired in S340 and the learning data may be compared to determine input data corresponding to the detection data.

  (C-2) The learning data and combination learning data referred to in the input mode of the main process may be acquired from another storage medium. Examples of other storage media include a CD-ROM and a memory card.

  (C-3) The detection data used for the process of detecting the input timing may be any data as long as the input operation is detected, and can be changed as appropriate. For example, the input timing may be detected based on detection data of the hand on the side including the finger that has performed the input operation. In this case, for example, the finger that has performed the input operation may be specified based on the temporal change of the detection data. For example, the input timing may be detected based on the detection data of all detection conditions. In general, detection data of a finger that has performed an input operation changes most. Therefore, detection data that includes a finger that has performed an input operation is preferably used as detection data that is used for processing for detecting input timing. For example, when a sensor other than an acceleration sensor is used as the sensor, the input timing may be detected using acceleration calculated based on data obtained from the sensor.

  (C-4) The input timing detection criteria can be changed as appropriate. For example, the detection reference may be whether or not the detection data has reached a predetermined value (for example, 1500 mg) or more. The predetermined value in this case is preferably a value that takes into account the value of the detection data when the input operation is executed and the value of the detection data when the input operation is not executed. Further, for example, whether or not there is a maximum value may be used as a detection reference. The maximum value may be detected by a known method.

  (C-5) Regardless of whether or not the input timing is detected, the detection data (feature value) may be compared with the learning data to determine the input data corresponding to the detection data.

  (E) Various setting values referred to in the main process can be changed as appropriate. For example, the same value may be set for L1 and L2 that define the width of the identification window regardless of the operation state. The learning data and the combination learning data of the above embodiment can be appropriately changed according to the virtual operation member, the assumed operation state, and the type of sensor. For example, in the above embodiment, it is assumed that the continuation operation and the normal operation are executed with different hands, but a case where both the operations are executed with the same hand may be assumed. For example, a case where a plurality of continuous operations are executed with the same hand, such as “Ctrl + Alt + Delete” on the keyboard, may be assumed. In this case, if learning data is stored for each operation, the type of input operation can be specified based on the detection data as in the above embodiment. Specifically, an operation in which two or more specific keys are simultaneously pressed may be learned and identified as one operation. In the above-described example, an operation in which “Ctrl + Alt” is simultaneously pressed is learned as one continuous operation, and Delete is learned as a normal operation. In this way, input data corresponding to “Ctrl + Alt + Delete” can be determined by the same processing as in the above embodiment. Specifically, an operation in which two or more specific keys are simultaneously pressed may be learned and identified as one operation. In the above example, an operation in which “Ctrl + Alt” is keyed in the same period is learned as one continuous operation, and an operation in which Delete is keyed is learned as a normal operation. In this way, input data corresponding to “Ctrl + Alt + Delete” can be determined by the same processing as in the above embodiment.

DESCRIPTION OF SYMBOLS 1 Input device 2, 3 Data collection device 4 Data determination device 21 Sensor 22 Control part 38 Flash memory 41, 101 CPU
42,102 ROM
43,103 RAM
45 A / D converters 47 and 106 Antenna 383 Window information storage area 384 Learning data storage area 385 Combination learning data storage area

Claims (6)

A sensor that is attached to the hand and generates an electrical signal corresponding to the movement of the hand; and an output unit that outputs data including the electrical signal generated by the sensor as detection data, and is output by the output unit. An input device that determines input data according to the movement of the hand that operates one or more virtual operation members based on the detection data,
A normal operation that is an operation in which the operation of the operation member is continued below a threshold; a start operation that starts a continuous operation that is an operation in which the operation of the operation member is continued longer than the threshold; and For each operation state including a release operation for stopping a continuous operation, the acquisition period includes an input timing that is a timing at which the hand operates the operation member in any of the operation states, and has a first predetermined length. Storage control means for associating the detection data within the acquisition period with assignment data, which is data assigned to the operation member, and storing it in a storage means as learning data;
The detection data output from the output means within the acquisition period is compared with the learning data for each operation state stored in the storage means, and the allocation data corresponding to the detection data is An input device comprising: determining means for determining input data. The detection data output from the output means within the acquisition period is compared with the learning data stored in the storage means until the release operation is detected after the start operation is detected. Period detecting means for detecting a period as the continuous operation period;
The storage means stores the learning data corresponding to a combination operation that is an operation when the normal operation is performed during the continuous operation period detected by the period detection means as combination learning data.
The determination unit compares the detection data output from the output unit with the combination learning data stored in the storage unit during the continuous operation period detected by the period detection unit, and performs the detection. The input device according to claim 1, wherein when the combination learning data corresponding to data is stored in the storage unit, the input data corresponding to the detection data is determined.   The input device according to claim 1, wherein the first predetermined length is different between the normal operation, the start operation, and the release operation. Timing detection means for detecting the input timing based on a change in the detection data within a period of the second predetermined length output by the output means;
The determining unit executes a process of determining the input data when the input timing is detected by the timing detecting unit;
The period from the start of the acquisition period to the input timing is L1, the period from the input timing to the end of the acquisition period is L2, and the minimum time interval when the normal operation is continuously executed is T1, When the minimum time interval from the detection of the start operation of the combination operation to the detection of the release operation is T2, and the second predetermined length is A,
The first predetermined length of the start operation satisfies L1 <T1 and L2 <T2 and A <L1 + L2.
The first predetermined length of the release operation satisfies L1 <T2 and L2 <T1 and A <L1 + L2.
The input device according to claim 3, wherein the first predetermined length of the normal operation satisfies L1 <T1 and L2 <T1 and A <L1 + L2. An output step of outputting, as detection data, data including the electrical signal generated by a sensor attached to the hand and generating an electrical signal corresponding to the movement of the hand, the detection data output in the output step; An input method for determining input data corresponding to the movement of the hand operating one or more virtual operation members,
A normal operation that is an operation in which the operation of the operation member is continued below a threshold; a start operation that starts a continuous operation that is an operation in which the operation of the operation member is continued longer than the threshold; and An acquisition period including an input timing that is a timing at which the hand operates the operation member in any of the operation states for each operation state including a release operation for stopping the continuous operation, and the first predetermined length of the acquisition period A storage control step of associating the detection data within the acquisition period with allocation data that is data allocated to the operation member and storing it in the storage means as learning data;
The allocation data corresponding to the detection data by comparing the detection data output from the output step within the acquisition period and the learning data stored for each operation state stored in the storage means And a determination step of determining the input data as the input data.   An input program for causing a computer to function as various processing means of the input device according to claim 1.

Images