Method and device for obtaining a control signal from an operating gesture

Abstract

Method for obtaining at least one control signal intended for setting an operating parameter by operating at least one selectable object (103) located in an object area (105) from a recognition and evaluation of an operating gesture of a hand (101) of a user, wherein in the operating gesture a directional axis (104) of a selected finger (102) of the hand (101) is directed towards a selected one of the at least one object (103) and at least predominantly at the same time a rotation (115; 116) of the hand (101) is carried out; for selecting the object (103) comprising the method steps - taking at least one image of the hand (101) when performing the operating gesture in a recognition area (111), - recognizing the selected finger (102) in at least one image of the hand (101), - Detecting the direction axis (104) of the selected finger (102), - Calculating an intersection area of the direction axis (104) with the object area (105) and - Determining the object (103) closest to this intersection area; for operating the object (103) comprising the further method steps - recognizing at least three fingertips (117, 118, 119, 120, 121) in at least one image of the hand (101), - determining the positions of at least three fingertips (117, 118, 119, 120, 121) on an image plane (112), - calculating a best-fit line (125) from the positions of at least three fingertips (117, 118, 119, 120, 121) on the imaging plane (112), - determining an inclination of the best-fit line (125) relative to a reference direction on the imaging plane (112), - determining an initial position (122) of the regression line (125), - Selecting the initial position (122) of the regression line (125) as the reference line (123), - determining the inclination of the reference line (123) relative to the reference direction on the imaging plane (112), - determining the amount and direction of an angle forming a relative rotation (124) between the regression line (125) and the reference line (123), and - Formation of the control signal from the amount and direction of the relative rotation.

Description

Prior ArtThe document EP 2 441 635 A1 discloses a vehicle-user interface system and a method for equipping a vehicle electronics system with a user interface. The vehicle-user interface comprises a projector for projecting an image onto a projection surface of the passenger compartment of the vehicle and a camera arranged in the passenger compartment, which monitors a spatial region that can be reached by a hand of a user and that can extend in front of the projection surface. The camera acquires image data of the spatial area and transmits this image data to a data processing unit, from which the image data are analyzed with the aid of an image analysis method. The data processing unit recognizes the hand of the user in the spatial domain and carries out image-assisted gesture recognition, for example by decomposing the captured image into a set of features taking into account captured individual images from an image sequence. Based on these features, different positions of the vehicle occupant's hand can be detected. The data processing unit is configured with a device for recognizing gestures as a sequence of positions of the hand and signals the gestures as inputs of the user in order to operate functions of the vehicle electronics system.Operating elements can be represented by the projection image. The data processing unit recognizes the gesture of a finger of the user pointing to an operating element or touching it. If the vehicle electronic system is a navigation system, the projection image can comprise a map section to be displayed, and the data processing unit recognizes gestures for controlling the map display, such as a movement of a finger over the displayed map section, in order to track the map in the direction in question. By spreading two fingers on the projected map section, the map can be displayed on an enlarged scale and the scale is reduced with the opposite finger movement. When performing the gesture, the user's hand does not have to touch the projection surface, but the gesture may also be performed and recognized at some distance from the projection surface, e.g., where the user's hand is usually on driving, thereby making the driver less distracted.The document US 2012 / 0105613 A1 discloses a method for insensitive, video-assisted recognition of handwriting and gestures for use in motor vehicles. To receive input from a user, a first travel time curve of a center of gravity of a hand of the user is scanned while making a gesture. A second travel time curve of a fingertip of the user's hand is also scanned when performing the gesture. An alphanumeric character represented by the gesture of the hand is determined as a function of the two travel time curves.From the document U.S. Pat. No. 7,812,826 B2 a portable communication device with multiple touch input is known, which detects one or more multiple touches and movements and carries out one or more operating operations on an object displayed on a touch screen in accordance with the one or more multiple touches and / or movements. The graphical object can be a rotary knob, and the user can place one or more fingers on the circumference of the rotary knob and execute a rotating or rotating movement in or counter-clockwise in order to set an operating parameter on the device, for example, as an operating operation.DE 10 2009 032 069 A1 discloses a method and a device for providing a user interface in a vehicle.DE 10 2009 057 739 A1 discloses an operating device for a vehicle with pressure or gesture detection.DE 10 2006 037 156 A1 discloses an interactive operating device and method for operating the interactive operating device.DE 10 2010 007 455 A1 discloses a system and a method for contactless detection and recognition of gestures in a three-dimensional space.Disclosure of the InventionThe invention has the object of creating a simple and reliable method for obtaining an actuating signal by detecting and evaluating an easily and intuitively trainable and executable operating gesture and a device for carrying out the method.This object is achieved by a method for obtaining at least one actuating signal determined for setting an operating parameter by operating at least one selectable object located in an object region from a detection and evaluation of an operating gesture of a hand of a user, wherein in the operating gesture a direction axis of a selected finger of the hand is directed at a selected one of the at least one object and a rotation of the hand is carried out at least predominantly simultaneously. In order to obtain the at least one actuating signal, a two-part hand gesture is thus recognized and evaluated, which is largely similar to an operation, in particular of a rotatable actuating element, and can therefore be easily and intuitively learned and executed. In this operating gesture, the user points to the selected object with the selected finger, preferably an index finger, and performs an actuating movement for operating the object by rotating the hand. The rotation of the hand takes place about an axis of rotation which preferably, but not necessarily, at least substantially corresponds to the direction axis of the selected finger.For selecting the object, the method according to the invention comprises the following method steps: recording at least one image of the hand in a recognition region when carrying out the operating gesture, recognizing the selected finger in the at least one image of the hand, recognizing the direction axis of the selected finger, calculating a section of the direction axis with the object region and determining the object closest to this section region.For operating the object, the method according to the invention comprises the further method steps: recognizing at least three fingertips in the at least one image of the hand, determining the positions of the at least three fingertips on an imaging plane, calculating a compensation straight line from the positions of the at least three fingertips on the imaging plane, determining an inclination of the compensation straight line with respect to a reference direction on the imaging plane, determining a starting position of the compensation straight lines, selecting the starting position of the compensation straight lines as a reference straight line, determining the inclination of the reference straight lines with respect to the reference direction on the imaging plane, determining the amount and direction of an angle forming a relative rotation between the compensation straight line and the reference straight line, and forming the actuating signal from the amount and direction of the relative rotation.In the context of the present invention, the object described above is generally understood to mean a spatial area to which the selected finger is directed, i.e. to which the selected finger is shown, during the execution of the operating gesture. This can be a real object, such as a real, mechanical operating element, but preferably a virtual object. The object represents in particular an operating element. The spatial area forming the object can also be empty; the object does not have to be represented therein. The object serves only for assigning the location of the operating parameter to the operating gesture. The operation of the object then means setting of a-preferably analog-operating parameter by the associated operating element for which the object is a symbol. The object area is then a spatial area to which objects for selection or operation by the operating gesture are assigned or in which they are arranged. The object region can be substantially one-, two- or three-dimensional depending on the arrangement of the objects. The section area is understood to mean the section of the direction axis that falls within the object area. Depending on the configuration of the object region, this is a point or a route.The detection area is the spatial area in which the user's hand is located for detecting and evaluating the operating gesture by a sensor, in which the hand can thus be detected by the sensor. The imaging plane is understood to mean the sensor plane, i.e. the plane on which the image of the hand is recorded. If the sensor is preferably formed by a camera, the imaging plane is understood to mean the image plane of the camera.The at least one image of the hand is preferably recorded three-dimensionally, for example using a time-of-flight camera, also referred to as a TOF camera. Furthermore, a sequence of images, e.g. a video sequence, is preferably recorded in the recognition area when carrying out the operating gesture of the hand. In a simplified embodiment, two-dimensional images are recorded. For recognizing the selected finger and its direction axis, image recognition methods of a known type are used, which provide data on the position and direction of the direction axis. From these data and the known arrangement of the object region, the intersection region can be calculated easily. Since, furthermore, the position of the object or objects in the object region is known, the object which is closest to the intersection region can likewise easily be determined. In the selection of terms of the present invention, the position, for example of a straight line, a finger, etc., comprises the position thereof, also referred to as the location and described by spatial location coordinates, and the direction thereof, described by spatial direction coordinates.For operating the object, the already present images of the hand are preferably evaluated, so that renewed recording is dispensed with. However, a two-dimensional recording is advantageous for this purpose. If three-dimensional images are present, these can be evaluated in a simplified manner. When recognizing the at least three fingertips, one of them is preferably, but not necessarily, that of the selected finger. This simplifies the execution and the recognition and evaluation of the operating gesture. This applies, for example, to calculating the best fit line from the positions of the at least three fingertips on the imaging plane. The best fit line is a line between the positions of the fingertips on the imaging plane, which is calculated according to a predetermined algorithm, for example in such a way that the sum of the perpendicular distances of the positions of the fingertips from the best fit line assumes a minimum. The reference straight line is defined as the best fit line at the beginning of the operating gesture, i.e. the best fit line in its initial position. This can be done, for example, on the basis of the first images of the hand recorded by the sensor when the latter is brought into the detection region. Preferably, the reference straight line corresponds to a relaxed and approximately horizontal position of the hand. After defining the reference straight lines, relative rotations of the hand, i.e. of the compensation straight lines, can be calculated. The slopes are determined both once from the reference straight line and then continuously from the respectively currently calculated compensation straight line. For this purpose, a coordinate system assumed for the imaging plane of the sensor forms a reference direction. As soon as the reference straight line is determined, the relative rotation can be determined for each image recorded by the sensor, on which the execution of the operating gesture is represented and the hand is correctly detected.When determining the relative rotation between the compensation straight line and the reference straight line, which is calculated on the basis of the profile of the gradient of the compensation straight line over a sequence of images recorded by the sensor, preferably a camera, only their slopes or slopes are decisive. The positions, i.e. the locations at which the mentioned straight lines are located, are insignificant for the evaluation. This facilitates the execution and recognition of the operating gesture, in particular the rotation of the hand, since the recognition of the rotation is stable with respect to possible displacements of the hand within the sequence of the images. From the relative rotation, the magnitude thereof, i.e. the absolute value of the angle between the mentioned straight lines, and the sign, i.e. the direction of rotation of the compensation straight line with respect to the reference straight line, are used for evaluating the operating gesture, i.e. for forming the actuating signal for operating the object, i.e. for setting the operating parameter for which the object is a symbol.The invention enables a simple and intuitively manageable operation which does not require any mechanical operating elements or touch screens. The operating gesture can be executed in free space with a secure visibility and at the same time enables an operating parameter of a technical device to be selected and set, preferably an operating parameter for operating electrical or electronic devices of a vehicle and / or in a vehicle.Advantageous embodiments of the method according to the invention are characterized in the dependent claims which refer back thereto.According to a preferred development of the method according to the invention, the detection and evaluation of the operating gesture for forming the actuating signal is carried out as long as the hand of the user is located in the detection region and the direction axis of the selected finger is directed at one of the objects, i.e. as long as a hand posture is detected as an operating gesture in the above sense. Conversely, the generation of the control signal is ended as soon as the relative rotation is zero, the hand is moved out of the detection area or the hand posture is changed such that the operating gesture is clearly no longer executed. Preferably, as long as the fingers are extended and slightly spread and the positions of the fingertips on the imaging plane at least approximately form a line, it is assumed that the operating gesture is still being carried out. In a preferred development, the spreading of the fingers can be detected, for example, from distances between the individual fingertips or from angles between direction axes of the fingers. The start of the operating gesture is recognized accordingly. Alternatively, the recognition of the operating gesture is initiated or ended by a further, preferably deviating, operating gesture.In a further preferred embodiment of the method according to the invention, the selected object or its position is displayed and / or marked and / or visually highlighted in a display area in order to feed back the recognition of the operating gesture to the user. The display area is the spatial area in which the objects are displayed for perception by the user and which is arranged at any desired position, which is preferably easily visible by the user. This is in particular a planar or curved display surface, but also an arbitrarily contoured display space; this is used in particular when displaying the objects via smart glasses. The display area is formed by a display device such as a screen, a visual field display, smart glasses or the like. The invention enables object area and display area to differ from one another. In an advantageous embodiment, however, the object area is at least almost identical to the display area. In particular, the objects are displayed directly at the position to which the direction axis of the selected finger is directed; there, the selected object is advantageously displayed in a manner emphasized with respect to non-selected objects.In a further preferred embodiment of the method according to the invention, the fingertips of the thumb, of the index finger and of the little finger are recognized in the at least one image of the user's hand. Preferably, the positions of middle and / or ring fingers are not evaluated for operating the object, because these do not provide any additional information that is essential for detecting the rotation of the hand when the hand is held relaxed.In an advantageous development of the method according to the invention, an increment is assigned to the determined amount of relative rotation according to a predefined function in order to form the actuating signal, said increment being statically and / or repeatedly added to or subtracted from an initial value of the operating parameter to be set according to the direction of relative rotation for each predefined time period. Increment is understood to mean an amount of change by which the operating parameter is adjusted, i.e. changed, when the object is operated. The relative rotation can thus be used directly for setting the operating parameter by linking the angle forming it via a linear or non-linear function to values which are added in the correct sign to the value of the operating parameter which was present at the beginning of the execution of the operating gesture. In order to achieve a higher sensitivity or accuracy of the setting, instead a continuous, i.e. temporally continuous, change of the operating parameter is also carried out, the increment of which, i.e. extent per predefined time period, depends on the relative rotation. The change in the operating parameter is preferably made visible to the user, in particular in the display area, so that the user can adapt the rotation of the hand to a desired speed of the change in the operating parameter.According to a further preferred embodiment of the method according to the invention, a change in direction of the direction axis during the operation of the selected object is evaluated for a continuous change in the selection made. If the user changes the position or the direction axis of the selected finger during the setting of the operating parameter by the operating gesture, this change is ignored depending on the operating parameter for stabilizing the selection made or is taken into account for a dynamic change of the selection made. If, for example, a zoom level of a road map representation is set by the operating gesture, the change in direction is taken into account, for example, for continuously changing a zoom center during zooming. If, for example, a value of a-preferably virtual-slider in a setting menu is changed by the operating gesture, the change in direction is advantageously ignored in order to avoid unintentional selection of another slider which would then likewise be adjusted.The above-mentioned object is furthermore achieved according to the invention by an apparatus for carrying out the above method. For this purpose, the device has: an object area with at least one selectable object located therein, a display area for displaying and / or marking and / or visually emphasized display of the at least one selected object, a sensor for capturing at least one image of a hand of a user when carrying out an operating gesture, a recognition area in which the hand of the user can be captured by the sensor when carrying out the operating gesture, an imaging plane of the sensor, preferably an image plane of a camera, for imaging the image of the hand thereon, and a computing unit. The latter is configured to recognize a selected finger in the at least one image of the hand, recognize a direction axis of the selected finger, calculate an intersection region of the direction axis with the object region, determine the object closest to this intersection region, recognize at least three fingertips in the at least one image of the hand, ascertain the positions of the at least three fingertips on the imaging plane, calculate a compensation straight line from the positions of the at least three fingertips on the imaging plane, ascertain an inclination of the compensation straight line with respect to a reference direction on the imaging plane, determine a starting position of the compensation straight line, choose the starting position of the compensation straight line as a reference straight line, determine the inclination of the reference straight line with respect to the reference direction on the imaging plane, determining the magnitude and direction of an angle forming a relative rotation between the compensation straight line and the reference straight line, and for forming a control signal from the magnitude and direction of the relative rotation.For recognizing the operating gesture, an arrangement of the sensor is advantageous in such a way that, in the case of a customary, pleasant or intuitively assumed hand posture, the fingers point at least approximately toward the sensor or roughly in the direction of the sensor, preferably the camera. As a result, rotations of the hand and thus of the fingertips can be detected well since they are effected at least substantially about the optical axis of the sensor, that is to say the movement can be recorded two-dimensionally since its detection does not require a depth determination.The object mentioned above is also achieved according to the invention by a computer program product having program parts for executing the method according to the invention, a data structure generated by the method or the computer program product, and a data carrier having such a computer program product or such a data structure.Exemplary embodiments of the invention are illustrated in the drawing and are described in more detail below, corresponding elements being provided with the same reference numerals in all figures and a repeated description of these elements being omitted. The following are shown: FIG. 1 shows a roughly schematic representation of an example of a device with which the method according to the invention can be carried out, FIGS. 2, 3 and 4 each show a roughly schematic representation of a hand posture when carrying out an operating gesture, FIG. 5 shows a roughly schematic representation of positions of the fingertips of a hand in a neutral posture with associated compensation lines, FIG. 6 shows the representation according to FIG. 5 with a rotated hand posture, and FIG. 7 shows a comparison of different slopes of the best fit line.FIG. 1 shows, roughly in a block diagram, an exemplary embodiment of a device 100 for obtaining at least one actuating signal, here for setting an operating parameter of a vehicle, from a detection and evaluation of an operating gesture of a hand 101 of a user, as shown schematically in FIGS. 2 to 4. During the operating gesture, a selected finger 102, e.g. index finger, is directed onto an object 103 selectable thereby. All fingers of the hand 101 are extended and spread slightly. For placement, the hand 101 is rotated substantially about a direction axis 104 of the index finger 102 out of a neutral posture 114, e.g., to the right 115 for enlarging or to the left 116 for reducing the operating parameter determined by the selected object 103.The device 100 has an object region 105, which is formed here by a virtual, rectangular surface which, although drawn as such in FIG. 1, does not have to be visible to the user, but can also be designed as a virtual or actual touch surface. In the object region 105, here on the surface, there is the selected object 103, here a section of the virtual surface, which likewise does not have to be visible. A display area 106, here e.g. a screen in a dashboard 107 of a motor vehicle indicated with steering wheel 108 and gear lever 109, serves for displaying or marking or visually, e.g. colored, highlighted display of the selected object 103 and-for the sake of better orientation of the user-at least a part of the hand 101 in its posture relative to the object 103. A sensor 110, here a camera arranged on the instrument panel 107, preferably a camera recording three-dimensional images, e.g. a time of flight (TOF) camera, serves for recording an image, preferably an image sequence, of the hand 101 when carrying out the operating gesture, for which purpose the hand 101 is brought into a recognition region 111, here the recording region of the camera 110. During the recording, the hand 101 is imaged on an imaging plane 112 of the camera 110.A computing unit 113 connected to the camera 110 and the display area 106 serves for recognizing the selected finger 102 and its direction axis 104 in the image of the hand 101, calculating an intersection point of the direction axis 104 with the object area 105 and determining the object 103 closest to the intersection point. The arithmetic unit 113 is also used to recognize at least three fingertips in the image of the hand 101 and to determine the positions of the fingertips on the imaging plane 112. In FIG. 5, the positions of the fingertips of the thumb 117, the index finger 118, the middle finger 119, the ring finger 120, and the little finger 121 of the hand 101 are roughly schematically shown in the neutral posture 114. The arithmetic unit 113 calculates an associated compensation straight line 122 from the positions of the fingertips 117 to 121 and the inclination thereof, for example with respect to a reference direction on the imaging plane 112. The starting position of the best fit line 122 is advantageously that in the neutral position 114; it is also selected as a reference straight line 123 which corresponds to the best fit line 122 in FIG. 5. When carrying out the operating gesture, the inclination of the compensation straight line changes by rotation 115 or 116 of the hand 101, which, as shown in FIG. 6, assumes the position 125 by an angle referred to as relative rotation 124, by which the reference straight line 123 must be rotated here about an intersection point 126 with the compensation straight line in order to bring both into line. The arithmetic unit 113 finally serves for ascertaining the amount and direction of the relative rotation and for forming the actuating signal therefrom for setting the operating parameter.FIG. 7 shows a comparison of different slopes, i.e. different values of the relative rotation 124 of the compensation straight line 125 with respect to the reference straight line 123, which are used for obtaining different control signals for setting the operating parameter. In the first diagram from the left, the best fit line 125 is strongly rotated to the left. From this, an actuating signal is advantageously derived, by means of which the value of the operating parameter is rapidly reduced. In the second diagram, the best fit line 125 is only slightly rotated to the left for deriving a control signal for slowly reducing the value of the operating parameter. In the middle diagram, no change of the operating parameter takes place in the neutral position. Correspondingly, according to the fourth diagram, the value of the operating parameter is increased slowly and, after the fifth diagram, the value of the operating parameter is increased rapidly. As an example, with a relative rotation of 2° to 5°, the value of the operating parameter per second is reduced by 0.5 units and with a relative rotation of 5° to 10° by 2 units per second. An increase in the value of the operating parameter takes place accordingly upon rotation in the opposite direction. Further ranges of values of the relative rotation can be defined. The operating parameter is reduced or increased the faster the hand is inclined.In summary, according to the invention, a method for obtaining an actuating signal by detecting and evaluating an operating gesture is provided, in which a direction axis of a selected finger of a hand is directed onto an object and the hand is rotated. For this purpose, the method steps of capturing an image of the hand, recognizing the selected finger, recognizing the direction axis, calculating a section of the direction axis with the object region, determining the object closest to this section and recognizing at least three fingertips, ascertaining the positions of the fingertips on an imaging plane, calculating a compensation straight line from the positions of the fingertips, ascertaining an inclination of the compensation straight lines, determining a starting position of the compensation straight lines and selecting the starting position as a reference straight line, determining the inclination of the reference straight lines, ascertaining the amount and direction of an angle between the compensation straight line and the reference straight lines and forming the actuating signal from the amount and direction of the angle are proposed. Furthermore, according to the invention, a device for carrying out the method is proposed. This makes it possible to dispense with mechanical operating elements or touch screens.

Claims

Method for obtaining at least one actuating signal determined for setting an operating parameter by operating at least one selectable object (103) located in an object region (105) from a detection and evaluation of an operating gesture of a hand (101) of a user, wherein in the operating gesture a direction axis (104) of a selected finger (102) of the hand (101) is directed at a selected one of the at least one object (103) and at least predominantly simultaneously a rotation (115; 116) of the hand (101) is carried out; for selecting the object (103) comprising the method steps of - recording at least one image of the hand (101) when carrying out the operating gesture in a recognition area (111), - recognizing the selected finger (102) in the at least one image of the hand (101), - recognizing the direction axis (104) of the selected finger (102), - calculating an intersection area of the direction axis (104) with the object area (105), and - determining the object (103) closest to this intersection area; for operating the object (103) comprising the further method steps of - recognizing at least three fingertips (117, 118, 119, 120, 121) in the at least one image of the hand (101), - determining the positions of the at least three fingertips (117, 118, 119, 120, 121) on an imaging plane (112), calculating a compensation straight line (125) from the positions of the at least three fingertips (117, 118, 119, 120, 121) on the imaging plane (112), - determining an inclination of the compensation straight line (125) with respect to a reference direction on the imaging plane (112), - determining a starting position (122) of the compensation straight line (125), - selecting the starting position (122) of the compensation straight line (125) as a reference straight line (123), - determining the inclination of the reference straight line (123) with respect to the reference direction on the imaging plane (112), - determining the amount and direction of an angle forming a relative rotation (124) between the compensation straight line (125) and the reference straight line (123), and - forming the actuating signal from the amount and direction of the relative rotation.Method according to Claim 1, characterized in that the detection and evaluation of the operating gesture for forming the actuating signal is carried out as long as the hand (101) of the user is located in the detection region (111) and the direction axis (104) of the selected finger (102) is directed at one of the objects.Method according to Claim 1 or 2, characterized in that, in order to feed back the recognition of the operating gesture to the user, the selected object (103) or its position is displayed and / or marked and / or visually highlighted in a display area (106).Method according to one or more of the preceding claims, characterized in that the fingertips of the thumb (117), of the index finger (118) and of the little finger (121) are detected in the at least one image of the user's hand (101).Method according to one or more of the preceding claims, characterized in that, in order to form the actuating signal, an increment is assigned to the determined amount of relative rotation according to a predefined function, said increment being added statically and / or repeatedly per predefined time period or being subtracted therefrom according to the direction of relative rotation to an initial value of the operating parameter to be set.Method according to one or more of the preceding claims, characterized in that a change in direction of the direction axis (104) during the operation of the selected object (103) is evaluated for a continuous change in the selection made.Device (100) for carrying out the method according to one or more of Claims 1 to 6, having - an object region (105) having at least one selectable object (103) located therein, - a display region (106) for displaying and / or marking and / or displaying the at least one selected object (103) in an optically emphasized manner, - a sensor (110) for recording at least one image of a hand (101) of a user when carrying out an operating gesture, - a detection region (111) in which the hand (101) of the user can be recorded by the sensor (110) when carrying out the operating gesture, - an imaging plane (112) of the sensor (110) for imaging the image of the hand (101) thereon, and - a computing unit (113) for ◯ detection of a selected finger (102) in the at least one image of the hand (101), ◯ Recognizing a direction axis (104) of the selected finger (102), ◯ Calculating an intersection region of the direction axis (104) with the object region (105), ◯ Determining the object (103) closest to this intersection region, ◯ Recognizing at least three fingertips (117, 118, 119, 120, 121) in the at least one image of the hand (101), ◯ Determining the positions of the at least three fingertips (117, 118, 119, 120, 121) on the imaging plane (112), ◯ Calculating a compensation straight line (125) from the positions of the at least three fingertips (117, 118, 119, 120, 121) on the imaging plane (112), ◯ Determining an inclination of the compensation straight line (125) with respect to a reference direction on the imaging plane (112), ◯ determining a starting position (122) of the compensation straight line (125), ◯ selecting the starting position (122) of the compensation straight line (125) as a reference straight line (123), ◯ determining the inclination of the reference straight line (123) with respect to the reference direction on the imaging plane (112), ◯ determining the magnitude and direction of an angle forming a relative rotation (124) between the compensation straight line (125) and the reference straight line (123), and for ◯ forming a control signal from the magnitude and direction of the relative rotation.Computer program product, comprising program parts for executing a method according to at least one of the preceding claims 1 to 6.Machine-readable, in particular computer-readable, data structure, generated by a method according to at least one of Claims 1 to 6 and / or by at least one computer program product according to Claim 8.Machine-readable, in particular computer-readable, data carrier on which at least one computer program product according to Claim 8 is recorded and / or stored and / or on which at least one data structure according to Claim 9 is kept ready for retrieval.

Images