Multimodel Text Input by a Keyboard/Camera Text Input Module Replacing a Conventional Keyboard Text Input Module on a Mobile Device

Abstract

The present invention relates to a method and a module for a multimodal text input in a mobile device (1) via a keyboard or in a camera mode by holding the camera of the mobile device (1) over a written text, such that an image is taken of the written text and the written text is recognized, wherein the input text is output to an application requesting the input text, the method comprising the following steps: a) activating a keyboard mode; b) providing an A-Z-keyboard in a first field (4) for text input; c) activating the camera mode; d) capturing the image of the written text and displaying the captured image with the written text in a second field (5) of a display (2) of the mobile device (1); e) converting the captured image to character text by optical character recognition (OCR) and displaying the recognized character text on the display (2); outputting a selected character as the input text to the application upon a selection of the character on the A-Z-keyboard, or outputting a selected part of the recognized character text as the input text to the application upon a selection of the part of the recognized character text; wherein the respective selection takes place by a single keypress or control command, or by a single gesture.

Claims

1-15. (canceled)

16. A method for multimodal text input in a mobile device comprising a keyboard and a camera, the method comprising: requesting an input text; providing a keyboard and camera module comprising a keyboard mode for character selection and a camera mode, said keyboard mode for character selection comprising a frame for displaying input text in a display of a mobile device comprising a first field, and an A-Z keyboard for text input in said first field; and said camera mode comprising a second field and configured to capture an image of a written text; display said captured image in said second field; convert said captured image to character text by optical character recognition (OCR); and display the recognized character text in said frame, wherein said capture, display of said captured image, conversion, and display of said recognized character text is capable of being accomplished through a single operation; and outputting one or more of said selected characters and said recognized character text, or a combination of both, as input text to an application requesting said input text.

17. The method according to claim 16, wherein said frame of said display is a touch screen and said A-Z keyboard is a touch screen keyboard.

18. The method according to claim 16, wherein said recognized character text is displayed in one or more third fields of said frame or displayed as a text overlay of said captured image in said second field.

19. The method according to claim 16, wherein one or more suggestion candidates of said recognized character text are determined by an algorithm in connection with a database and are displayed in one or more third fields or as another overlay within said second field, and said one or more suggestion candidates are selectable by a keypress event on a key, wherein said key is one or more of a group consisting of a visible key; and a hidden key within one or more of a group consisting of one or more of the fields of said display, on the recognized character text, and on another text on said display.

20. The method according to claim 19, wherein said keypress event providing said selection is from the group consisting of a single mechanical keypress, a single touch keypress and a single swipe gesture on a touch screen.

21. The method according to claim 16, wherein frame size is the same frame size of an original standard keyboard module on said mobile device.

22. The method according to claim 16, wherein said second field is displayed adjacent to said A-Z keyboard in said first field.

23. The method according to claim 16, wherein the steps of capturing, displaying said captured image, converting, and displaying said recognized character text are executed repetitively until a certain keypress is detected to end said repetitive execution, wherein the respective latest recognized text in the part of the respective latest captured image is analyzed for new text in regards to a previously recognized character text that is output to said application requesting said input text, whereupon a control command is generated for the selection of the new text as the recognized character text, and said new text is output to said application requesting said input text.

24. The method according to claim 23, wherein said control command for the selection of said recognized character text is generated automatically via a detection algorithm, wherein said detection algorithm recognizes whether said recognized character text in a previously captured image and in the current captured image are the same.

25. The method according to claim 16, wherein: said keyboard mode for character selection is executed by a keyboard text input module; said camera mode is executed by a camera text input module, wherein the execution of said keyboard text input module is independent from the execution of said camera text input module, but wherein the execution of said camera text input module is dependent on the execution of said keyboard text input module; wherein said camera text input module is always active in the background to detect whether said keyboard text input module is active, wherein detection of execution of said keyboard text input module activates said camera mode, wherein at least the second field and the recognized character text are visible on the display; and wherein said camera mode is deactivated if said camera text input module detects that said keyboard text input module ceases to be active.

26. A method for multimodal text input in a mobile device comprising a keyboard and a camera, the method comprising: requesting an input text; providing a keyboard and camera module comprising a keyboard mode for character selection and a camera mode, said keyboard mode for character selection comprising a frame for displaying input text in a display of a mobile device comprising a first field; an A-Z keyboard for text input in said first field; and a first control key in the first field, wherein said first control key activates said camera mode and deactivates said keyboard mode; and said camera mode comprising a second field; and a second control key in the first field, wherein said second control key activates said keyboard mode and deactivates said camera mode, and said camera mode configured to capture an image of a written text; display said captured image in a second field; convert said captured image to character text by optical character recognition (OCR); and display the recognized character text in said frame, wherein said capture, display of said captured image, conversion, and display of said recognized character text is capable of being accomplished through a single operation; and outputting one or more of said selected characters and said recognized character text as input text to an application requesting said input text.

27. A mobile device configured to facilitate multimodal text input, the mobile device comprising a display configured to display text and image content in a frame having a first field, second field, and third field; an A-Z keyboard having a keyboard mode, wherein said A-Z keyboard is configured to receive input text; a camera having a camera mode; and a processor in communication with said display, said A-Z keyboard, and said camera, said processor being configured to receive input text via said A-Z keyboard, via said camera, or both; communicate with the A-Z keyboard to provide for selection of one or more characters on the A-Z keyboard; communicate with the camera to capture an image of said written text, display said captured image in said second field, convert said captured image to character text by optical character recognition (OCR), and display said recognized character text in said frame, wherein said capture, display of said captured image, conversion, and display of said recognized character text is capable of being accomplished through a single operation; and output said selected one or more characters and recognized character text to an application requesting input text.

28. The mobile device according to claim 27, wherein said processor is configured to generate said A-Z keyboard on said display using a touch screen interface, said A-Z keyboard being compatible with one or more applications requiring text input running on said mobile device, and said frame being maintained at a constant size on the display.

29. The mobile device according to claim 27, wherein said keyboard mode and said camera mode are both active simultaneously.

30. The mobile device according to claim 29, wherein said processor configures said A-Z keyboard as a touchscreen interface such that said touchscreen interface and said camera are active at the same time.

31. The mobile device according to claim 29, wherein said processor configures said A-Z keyboard as a touchscreen interface; the processor being further configured to display said captured image and said recognized character text in said second field; enable by one or more of a group consisting of a single keypress, a control command, and a single gesture the selection of said recognized character text as input text; and the selection of a character key from the A-Z keyboard as input text; and immediately output said input text to said application requesting said input text.

32. The mobile device according to claim 31, wherein said single keypress, said control command, and said single gesture include the selection of said recognized character text as input text; the selection of said character key from the A-Z keyboard as input text; and the immediate output of said input text to said processor.

33. The mobile device according to claim 27, wherein said A-Z keyboard comprises two sub-modules, wherein a first standard keyboard sub-module is activated by the processor in response to a request for input text by said application, said first standard keyboard sub-module being dependent on a second camera sub-module; and said second camera sub-module configured to be always active in the background detecting whether said first keyboard sub-module is activated.

34. The mobile device according to claim 33, wherein said processor is configured to activate said camera mode in response to the activation of said second camera sub-module, such that said second field is displayed adjacent to the A-Z keyboard; and shut down said camera mode when said second keyboard sub-module is no longer activated by said application.

35. The method according to claim 21, wherein said frame further comprises a third field.

36. The method of according to claim 22, wherein said frame comprises a third field displayed adjacent to said keyboard in said first field.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The foregoing will be apparent from the following more particular description of example embodiments of the invention, as illustrated in the accompanying drawings in which like reference characters refer to the same parts throughout the different views. The drawings are not necessarily to scale, emphasis instead being placed upon illustrating embodiments of the present invention.

[0023] FIG. 1a is a drawing of a mobile device with a display showing a text in an application window, an A-Z-keyboard with a camera mode key in a first field, a captured written text in the second field and a recognized text in a third field.

[0024] FIG. 1b is a drawing of the mobile device with the display showing the text in the application window and the A-Z-keyboard with the camera mode key in the first field.

[0025] FIG. 2a is a drawing of the mobile device with the display showing the text in the application window, a reduced keyboard with the camera mode key in the first field, the captured written text in the second field and the recognized text and suggestion candidates in three third fields.

[0026] FIG. 2b is a drawing of the mobile device with the display showing the text in the application window, the A-Z-keyboard with the CKK in the first field, the captured written text in the second field.

[0027] FIG. 3a is a drawing of the mobile device with the display showing the text in the application window, the reduced keyboard, the captured written text in the second field, the recognized text and suggestion candidates in three third fields, wherein the first field is overlapping the second and the third fields wherein a gesture crossing the first field is input.

[0028] FIG. 3b is a drawing of the mobile device with the display showing the text in the application window and the A-Z-keyboard, wherein another gesture is input crossing the first field.

[0029] FIG. 4 is a drawing of the mobile device with the display showing the text in the application window, an original A-Z-keyboard in the first field of an original keyboard application and the captured written text in the second field and the recognized text in the third field of a camera input module.

DETAILED DESCRIPTION

[0030] A description of example embodiments of the invention follows.

[0031] FIG. 1a shows a preferred embodiment of the present invention being a multimodal text input method and module on a mobile device 1. The multimodal text input method and module allow a user to input a text into an application on the mobile device 1 either via a keyboard or via a camera integrated in the mobile phone, wherein the user holds the mobile phone with the camera over a written text, the mobile phone captures an image of the written text and converts it via optical character recognition (OCR) into character text, such that parts of the recognized text can be selected as input text for the application. In other words, the user may input the text conventionally via the keyboard in what is called a keyboard mode or he may input the text via the camera in what is called a camera mode if a written text in front of him is available.

[0032] Preferably the multimodal text input method is implemented in a multimodal text input program module, or further simply called “module”, with the same interface to an application as an original keyboard module and under a respective operating system of the mobile device 1. If so, then said module can replace the original keyboard module on the mobile device 1, advantageously offering all applications multimodal text input features as described above. Thus the user can choose whether he wants to input the text conventionally via the keyboard or in the camera mode selecting the recognized text or a part thereof.

[0033] An alternative solution instead of replacing the original keyboard module by the complete integral multimodal text input program module is a provision of a separate camera text input module, wherein the camera text input module is like a complement or supplement to the original keyboard module and always active in the background to detect whether the original keyboard module is activated by the application. In case the keyboard module is detected as activated, the camera text input module activates the camera mode and preferably displays an additional field on the display 2 displaying the captured written text and preferably the recognized text which can be selected by a keypress, a gesture 9, by a voice command or the like. Thus the text can be input either via the keyboard or via the camera text input module in the camera mode and the input text will be sent to the application. In case the original keyboard module gets closed, by the application, a key, a timer or the like, the keyboard module is not detected anymore by the camera text input module, whereupon the camera text input module will deactivate the camera mode and delete any displayed second field 5 or third field 6 on the display 2.

[0034] For clarity, the “application” stands for instance for a translation application, a search application such as Google search or the like searching for a keyword in the internet, a phone application requesting for a phone number or a name or for any other application requesting text input.

[0035] The “text” stands preferably for any character text, a single or multiple character/s, a string of connected characters such as a word, a number or the like.

[0036] The “character” stands preferably for any character and for any number.

[0037] The “written text” stands preferably for any written or displayed text on a sheet of paper, in a book or also on a screen or the like which can be captured by the camera.

[0038] The “keyboard text input” is preferably understood as any text input on the keyboard.

[0039] The “keyboard” stands preferably for a touch screen keyboard, a mechanical conventional keyboard, a holographic keyboard or the like, wherein characters can be input manually or wherein a typing is registered.

[0040] The “A-Z-keyboard” stands preferably for a keyboard comprising a full set of keys from A to Z or from 0 to 1 and at least one control key such as ENTER, for instance.

[0041] The “original keyboard module” stands preferably for a standard keyboard module or sub-program, being already available for the respective mobile device 1 with its respective operating system and respective applications, but a new proprietary or separate keyboard module for a certain mobile device 1 is also conceivable.

[0042] A “Control key” stands preferably for a respective key on the keyboard executing a certain function or for a hidden touch key behind one of the displayed fields as the first 4, the second 5 and/or the third field 6, for instance.

[0043] A “keypress” or a respective “keypress event” stands preferably for a manual pressing of a certain key or hidden key of the keyboard or of the touch screen, for a certain gesture 9 on the touch screen or in front of a camera, and also for a swype gesture 9. A “single keypress” can also be a double click or double press or double touch on the same key or place. A certain signal strength or pattern from an acceleration sensor is preferably also usable as a keypress.

[0044] The “mobile device 1” stands preferably for any mobile-phone, smart-phone or the like.

[0045] The “display 2” stands preferably for any conventional non-touch-screen display as well as for any touch screen with touch key functionality.

[0046] FIG. 1a shows a preferred embodiment of an implementation of the multimodal text input method according to the present invention. FIG. 1 shows the mobile device 1 with the display 2 being the touch screen. On the touch screen there is displayed an application window 3 with a graphical user interface (GUI) of the application, the GUI comprising a text input field 3b and preferably a text output field below. As the text input field 3b has been activated by the application or by the user the application requested the text input and started the text input module being the multimodal text input module. The multimodal text input module starts displaying the conventional keyboard being the A-Z-keyboard in the first field 4. Preferably the first field 4 for the A-Z-keyboard has the same size as an original keyboard module which has been replaced on the mobile device 1 by the multimodal text input module. Thus the user may input text as usual via the displayed touch screen keyboard or he may press a camera mode key 8 to switch over to the camera mode.

[0047] FIG. 1b shows the same preferred embodiment of the implementation of the multimodal text input method as shown in FIG. 1a but after a keypress on the camera mode key 8. In the camera mode the first field 4 for the keyboard is preferably smaller compared to a size in the keyboard mode of FIG. 1a, such that the difference between the size of the first field 4 in the keyboard mode and the smaller size in the camera mode is used for displaying the second field 5 wherein a part of the captured written text is displayed and for displaying the third field 5 wherein the recognized text is displayed. In this preferred embodiment the size of any displayed fields in the keyboard mode as well as in the camera mode is always kept the same, such that the used size of the touch screen is kept the same as that of the original keyboard module, such that the original keyboard module can be replaced by the multimodal text input module without overlapping the application more than originally planned by a usage of the original keyboard module.

[0048] In the example shown in FIG. 1b, the user holds the camera of the mobile device 1 over the written text being “This text is on a sheet”, wherein the mobile device 1 captures preferably continuously images and displays a part of the respective image in the second field 5. Parallel to said process of capturing the respective images, the respective image gets analyzed by an optical character recognition (OCR) generating the recognized text and preferably displaying the recognized text in the third field 6. If the recognized and displayed text is correct and desired the user selects the recognized text as input text, whereupon the selected text is output to the application preferably in the same way as the text would have been input via the keyboard.

[0049] A selection of the recognized text in the third field can be made preferably by a second keypress on the camera key or a keypress on the third field or on the RETURN key or the like. Other preferred possibilities for the selection of the recognized text are for instance via voice or via a timer, such that if the camera is held for a time longer than a stored time limit over the same written text the selection is executed. Other kinds of selection are not excluded.

[0050] Another preferred embodiment of the camera mode includes word recognition within the recognized text, such that words are identified and such that upon a selection only the word in a focus or in the middle of the third field or behind cross hairs, respectively, is selected and output. Another preferred selection can be made by touching or pressing, respectively, on the word on the touch screen.

[0051] It is also imaginable that the recognized text gets displayed preferably as an overlaid recognized text over the captured text in the second field 5. In this case the captured text in the second field 5 gets preferably erased, such that the recognized text can be overlaid on the displayed image preferably in a similar size and width as the original text. This would reduce necessary space, such that the third field 6 is not needed anymore and therefore the first field 4 or the second field 5 can take over that space.

[0052] Another preferred method of selecting one of the words of the recognized text displayed adjacent to the displayed keyboard is via a keypress on a key in next proximity of the desired word as input text.

[0053] Another preferred method of selecting one of the words of the recognized text is by displaying next to each word a respective identifier such as a certain number, for instance, and by a keypress on that respective number on the keyboard.

[0054] Preferably by pressing a KEYBOARD key 7 the mode is switched back from the camera mode to the keyboard mode, wherein the keyboard as shown in FIG. 1a is bigger and easier to use than in the camera mode shown in FIG. 1b. It should be mentioned that in the camera mode according to the preferred embodiment shown in FIG. 1b the text input can also be undertaken via the shown keyboard, the only disadvantage would be that it is smaller. A switching between the keyboard and the camera mode can preferably be done in several different ways, by the keypress on the camera mode key 8 or the keyboard key 7, by a keypress on only the camera mode key 8 for changing the mode, by a gesture 9 as shown in FIG. 3a and FIG. 3b, and the like.

[0055] FIG. 2a shows another preferred embodiment of an implementation of the present invention. On the display 2 being the touch screen there are displayed three third fields 6, wherein preferably the first one displays the recognized text and the second and the third one beneath display suggestion candidates. The one or more suggestion candidates are preferably generated by an algorithm in connection to a database, wherein according the recognized character text, which might not be correctly recognized or spelled wrong, best fitting words with a high probability to be correct are generated as the suggestion candidates. The database is preferably a dictionary or a lookup-table. Thus the user can select whatever fits best as the input text. It is also imaginable that only one or more closest fitting words are generated as suggestion candidates for a recognized word in the recognized text which has not been found in the database. The keyboard in the first field 4 is in the camera mode preferably a reduced keyboard, wherein preferably only some necessary control keys or the bottom line of the A-Z-keyboard are displayed in order to provide more space for the camera mode input GUI being the second field 5 and one or more third fields 6 preferably in an area and of the size of the GUI of the original keyboard module.

[0056] A preferred embodiment of the present invention foresees also a sending of the captured image or preferably of the part of the displayed image in the second field 5 as image data to a remote server, where the image data are OCR processed and the recognized text and preferably the suggestion candidates are generated and returned to the mobile device 1. This preferred method of a kind of remote computing is advantageous as it keeps the calculation power and the memory requirement on the mobile device 1 small, especially as regards the database for the OCR and as regards different languages. Also the development effort is reduced drastically with regard to an implementation of image processing and the OCR on different mobile device 1 types and operating systems.

[0057] FIG. 2b shows another preferred embodiment of an implementation of the present invention, wherein the recognized text is displayed as the overlay over the text within the captured image in the second field 5. Preferably either the third field 6 with the recognized text is overlaid over the second field 5 with the captured image, or in the displayed part of the captured image the recognized characters are erased and only the recognized text in a similar size and width is overlaid.

[0058] FIG. 3a shows another preferred embodiment of an implementation of the present invention, wherein the mobile phone 1 is in the camera mode. In this preferred embodiment the GUI of the camera mode occupies a certain display area of the touch screen and none of the keyboard keys of the A-Z-keyboard is visible. The usage within the camera mode is as explained above, wherein a switch over or a change to the keyboard mode is preferably detected by a preferred gesture 9. The switch over to the keyboard mode is also imaginable by a keypress of any other defined gesture 9, of a hidden key, on the captured image in the second field 5, or the like. The full size of the text input GUI can be used in this way for the camera input mode, which is the certain display area. If the switch over to the keyboard mode is detected, preferably the image of FIG. 3b occurs on the mobile device 1.

[0059] FIG. 3b shows the same preferred embodiment of the implementation of the present invention as in FIG. 3a. Now the text input module is in the keyboard mode, wherein the A-Z-keyboard is displayed in the first field 4. Preferably the camera key 8 is displayed as well for another switch back to the camera mode as shown in FIG. 3a. The switch back from the keyboard mode to the camera mode is also possible by a preferred gesture 9, similar to the one described under the camera mode regarding FIG. 3a.

[0060] FIG. 4 shows another preferred embodiment of the implementation of the present invention, wherein the original keyboard module is kept on the mobile device 1 and wherein the multimodal keyboard input module is reduced to the camera text input module complementary to the original keyboard module. The preferred camera text input module is preferably adapted to enhance the original keyboard module on the mobile device 1 by the features of the camera mode. As the application requires the text input it starts the original keyboard module which is detected by the camera text input module being always active for detecting in the background. Upon the detection of the activated original keyboard module the camera text input module activates the camera and preferably opens the second field 5 and displays the part of the captured image with the written text.

[0061] Preferably the keyboard text input module is independent from the camera text input module, but the camera text input module is dependent of the state of the keyboard text input module, which is checked continuously by the camera text input module.

[0062] In the preferred embodiment shown the recognized text is displayed in the third field 6 below the second field 5. It is also imaginable to overlay the recognized text over the captured written text, such that the third field 6 is either overlaid over the second field 5, or such that the letters of the recognized text are overlaid over the captured written text within the part of the captured and displayed image.

[0063] FIG. 4 shows also hidden touch keys 10 within the second 5 and the third field 6. Such preferred hidden touch keys can also be seen as a reduced keyboard, wherein the hidden touch keys are displayed to the user by the frame or frame sides of the second 5 and/or the third field 6, for instance, or the like. A keypress on one of the hidden touch keys 10 within the preferred embodiment of FIG. 4 would select the recognized text, for instance.

[0064] Preferably the embodiment of FIG. 4 of the implementation of the present invention can also be used for another mobile device 1 with a conventional mechanical keyboard, wherein the second field 5 and/or the third field 6 are displayed and wherein the selection of the part or the whole of the recognized character text or of the suggestion candidate could be performed by the touch screen, by voice, by a rarely used key, by a certain pattern of an acceleration sensor, by an optical signal, or by another control signal. In the case of the conventional mechanical keyboard and if the display 2 is no touch screen, an additional key on the mechanical keyboard for the selection of a recognized text is rather difficult. Thus the selection could be controlled by another key event, such as for instance one generated by a voice command, by detection that a recognized text is in a repetitive sequence is for a longer time than a time limit the same one, by an acceleration sensor pattern or the like,

[0065] Where technical features mentioned in any claim are followed by reference signs, those reference signs have been included for the sole purpose of increasing intelligibility of the claims and accordingly, such reference signs do not have any limiting effect on the scope of each element identified by way of example by such reference signs.

REFERENCE NUMERALS

[0066] 1 mobile device [0067] 2 display [0068] 3 application window [0069] 3b text input field [0070] 4 first field [0071] 5 second field [0072] 6 third field [0073] 7 keyboard key [0074] 8 camera mode key [0075] 9 gesture [0076] 10 hidden touch key

[0077] The teachings of all patents, published applications and references cited herein are incorporated by reference in their entirety.

[0078] While this invention has been particularly shown and described with references to example embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the invention encompassed by the appended claims.