CN110096166A - A kind of virtual input method - Google Patents

Abstract

The present invention relates to the field of input devices, in particular to a virtual input method, the method comprising: generating a 3D image of the input device to obtain a 3D input device; using a camera to capture an image of the user operating the 3D input device; The image of the 3D input device is operated to input the image recognition model, the image recognition model is used to recognize the input content of the user operating the 3D input device; and the input content is displayed on the display device. The invention provides a new input mode, which can realize input without a physical keyboard.

Description

A virtual input method

technical field

The invention relates to the field of input devices, in particular to a virtual input method.

Background technique

The keyboard is the most common and important computer input device. It is widely used in mobile terminals such as robots, mobile phones, computers, security access control, monitoring, and instruments and meters used in various industries. However, these input devices such as mice and various keyboards are currently Physical devices such as buttons and touch screens.

The gaps between the keys of some key keyboards are relatively large, which is easy to hide dirt and dirt, is difficult to clean, and easily leads to key failure. In addition, the keys of some keyboards are fixedly printed with codes and symbols of the required national language, and these marks cannot be changed and are easily worn.

Contents of the invention

In order to solve the above problems, the object of the present invention is to provide a virtual input method, which can realize input without a physical keyboard, and is a new input method.

Based on this, the present invention provides a virtual input method, said method comprising: a virtual input method, characterized in that comprising:

Generate a 3D image of the input device to obtain a 3D input device;

Use the camera to capture images of the user operating the 3D input device;

The image captured by the camera of the user operating the 3D input device is input into the image recognition model, and the image recognition model is used to identify the input content of the user operating the 3D input device;

Displaying the input content on a display device.

The method for generating a 3D image of an input device includes a persistence of vision technology or a holographic image technology.

The persistence of vision technology includes a fan, LEDs are evenly arranged on the blades of the fan, and the fan rotates at a preset speed to obtain a 3D image.

The holographic projection technology includes:

Using laser light as the illumination source, and dividing the light emitted by the illumination source into two beams, one beam is directed to the photosensitive sheet, and the other beam is directed to the photosensitive sheet after being reflected by the subject;

The two beams of light are superimposed on the photosensitive sheet to generate interference;

The hologram reproduced by the basic principle of digital image is further processed to remove digital interference and obtain a 3D image.

The image input image recognition model of the user operating the 3D input device captured by the camera includes:

adding a preset number of images to the training set corresponding to the initial image recognition model;

Using the initial image recognition model to train the image to obtain an image recognition model;

The image is recognized by the image recognition model to obtain a recognition result.

The recognition result includes the input content of the user operating the 3D input device, and the input content includes numbers, symbols, and letters.

Said using said initial image recognition model to train said image, and obtaining an image recognition model includes:

Select a sample image from the target training data set, input the current sample image to the initial image recognition model, and obtain a recognition result;

Determine whether the recognition result is consistent with the result of the user operating the 3D input device, iteratively execute the above training process until the recognition result is consistent with the result of the user operating the 3D input device, and obtain a trained image recognition model.

The target training set includes images of a user operating a 3D input device.

The invention provides a new input mode, which can realize input without a physical keyboard, and the input content can be changed according to the user's needs, and has flexibility.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. Those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is a flow chart of the virtual input method provided by the embodiment of the present invention;

Fig. 2 is a flow chart of acquiring an image recognition model provided by an embodiment of the present invention.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

Fig. 1 is a flowchart of a virtual input method provided by an embodiment of the present invention, the virtual input method comprising:

S101. Generate a 3D image of an input device.

The method for generating a 3D image of an input device includes a persistence of vision technology or a holographic image technology.

The persistence of vision technology is POV (Persistence of vision) technology. When an object is moving rapidly, after the image seen by the human eye disappears, the human eye can still retain the image for 0.1-0.4 seconds. This phenomenon is called Persistence of vision phenomenon. When the human eye looks at an object, the image is formed on the retina and input to the human brain by the optic nerve to feel the image of the object, but when the object is removed, the impression of the object by the optic nerve will not disappear immediately, but will last for 0.1-0.4 seconds , This property of the human eye is called "persistence of vision of the eye".

The persistence of vision technology includes a fan, LEDs are evenly arranged on the blades of the fan, and the fan rotates at a preset speed to obtain a 3D image. In this way, the rotation of the fan and the display of the LED light bar are combined, and the principle of persistence of the human eye is used to form an aerial phantom of graphics, animation and video. When imaging, all LED lights emit light, and the surrounding shell is dark and does not emit light. In this way, the human eye will only receive the stay of light, and the dark and dull light will not be recognized by the brain because it is too fast, so the display effect It will only stay on the light displayed by the LED light, so that the effect of aerial imaging is realized, and then a 3D floating image is presented. Specifically, the visual effect produced by light on the retina will remain for a period of time after the light stops acting. Although the recognition accuracy of the human eye is very high, the time required for the optic nerve to transmit the image to the brain is 1/24 second. Just using this factor, when the switching speed of multiple freeze-frame pictures reaches 24 frames, a continuous picture can be formed (the 24 frames referred to here is the frame rate directly displayed to the human eye, not the frame rate transmitted by the picture to some machines ). The frame rate of the holographic 3D advertising machine is generally about 30 frames per second, which means that the time for each frame to freeze is 1/30 second, so that the imaging effect is realized. At the same time, when the device is imaging, all LED lights emit light, and the surrounding shell is dark and does not emit light. Then when the device is operating, the human eye will only receive the stay of light, and the dark color will not be detected because it is too fast. The brain distinguishes, so that the display effect will only stay on the light displayed by the LED, thus realizing the effect of holographic 3D imaging.

The holographic image technology is a technology that uses the principles of interference and diffraction to record and reproduce a real three-dimensional image of an object. The holographic technology uses a laser as an illumination source, and divides the light emitted by the light source into two beams, one beam is directly directed to the photosensitive sheet, and the other beam is reflected by the subject and then directed to the photosensitive sheet. The two beams of light are superimposed on the photosensitive sheet to produce interference, and finally the hologram reproduced by the basic principle of digital image is used for further processing to remove the digital interference and obtain a clear holographic image. Users do not need to wear stereoscopic glasses or any other auxiliary equipment, they can watch images from different angles with naked eyes.

If the input device is relatively simple or has high cost requirements (in simple terms, the input device is relatively simple, which means that the function is relatively small, the type of input image is relatively small, and the image accuracy is low, such as the product in the security system. The keyboards of electronic devices such as plackets, mobile phones, computers, and input devices are relatively complex), and POV technology can be used to generate 3D graphics. If the input device is more complex, such as AI-based virtual games, large-scale security systems, building structure models, robots etc., or high-end equipment, can generate 3D graphics through holographic imaging technology.

S102. Use a camera to capture an image of the user operating the 3D input device.

When the user operates the 3D input device, for example, the user clicks on a certain part of the 3D image, the camera captures the operation process of the user.

S103. Input the image taken by the camera into the image recognition model of the user operating the 3D input device, where the image recognition model is used to recognize the input content of the user operating the 3D input device.

The camera captures the image of the user clicking on the 3D input device, and inputs the image to the image recognition module to identify the user's operation. For example, the user clicks the letter M of the 3D input device, and the camera captures the user's click. The 3D input device takes the image of the letter M, and inputs the image to the image recognition module, and the image recognition module can judge that the content input by the user in the image is the letter M.

The image input image recognition model of the user operating the 3D input device captured by the camera includes:

(1), adding a preset number of images to the training set corresponding to the initial image recognition model;

When the user operates the 3D input device, for example, when the user clicks on a certain part of the 3D image, the camera captures an image of the user's operation process, and the user's operation result, that is, the input content, can be obtained from the image. The images of the user's operation process captured by the camera may be used as training images for training the initial image recognition model and added to the training set corresponding to the initial image recognition model.

(2), using the initial image recognition model to train the image to obtain an image recognition model;

The training of the image by the initial image recognition model includes:

Select a sample image from the target training data set, input the current sample image to the initial image recognition model, and obtain a recognition result, and the target training set includes an image of a user operating a 3D input device;

Determine whether the recognition result is consistent with the result of the user operating the input device, iteratively execute the above training process until the recognition result is consistent with the result of the user operating the 3D input device, and obtain a trained image recognition model.

(3) Using the image recognition model to recognize the image, and obtain a recognition result.

The image of the user's operation process captured by the camera is input into the initial image recognition model to obtain a recognition result. The recognition result includes the input content of the input device operated by the user, and the input content includes numbers, symbols, and letters.

S104. Display the input content on a display device.

The image is recognized by the image recognition model, and the obtained recognition result is displayed by a display device.

The display device includes a mobile communication terminal with a display function such as a mobile phone, a computer, and an iPad.

The invention provides a new input mode, which can realize input without a physical keyboard, and the input content can be changed according to the user's needs, and has flexibility.

Fig. 2 is a flow chart of acquiring an image recognition model provided by an embodiment of the present invention, and the acquisition process of the image recognition model includes:

S201. Add a preset number of images to the training set corresponding to the initial image recognition model;

The image input image recognition model of the user operating the 3D input device captured by the camera includes:

When the user operates the 3D input device, for example, when the user clicks on a certain part of the 3D image, the camera captures an image of the user's operation process, and the user's operation result, that is, the input content, can be obtained from the image. The images of the user's operation process captured by the camera may be used as training images for training the initial image recognition model and added to the training set corresponding to the initial image recognition model.

S202. Using the initial image recognition model to train the image to obtain an image recognition model;

The training of the image by the initial image recognition model includes:

Select a sample image from the target training data set, input the current sample image to the initial image recognition model, and obtain a recognition result, and the target training set includes images of user-operated input devices;

Determine whether the recognition result is consistent with the result of the user operating the input device, iteratively execute the above training process until the recognition result is consistent with the result of the user operating the 3D input device, and obtain a trained image recognition model.

S203. Recognize the image by using the image recognition model, and obtain a recognition result.

The image of the user's operation process captured by the camera is input into the initial image recognition model to obtain a recognition result. The recognition result includes the input content of the input device operated by the user, and the input content includes numbers, symbols, and letters.

The above is only a preferred embodiment of the present invention, it should be pointed out that for those of ordinary skill in the art, without departing from the technical principle of the present invention, some improvements and replacements can also be made, these improvements and replacements It should also be regarded as the protection scope of the present invention.

Claims (8)

1. A virtual input method, characterized in that, comprising: Generate a 3D image of the input device to obtain a 3D input device; Use the camera to capture images of the user operating the 3D input device; The image captured by the camera of the user operating the 3D input device is input into the image recognition model, and the image recognition model is used to identify the input content of the user operating the 3D input device; Displaying the input content on a display device. 2. The virtual input method according to claim 1, wherein the method for generating the 3D image of the input device comprises a persistence of vision technology or a holographic image technology. 3. The virtual input method according to claim 2, wherein the persistence of vision technology includes a fan, LEDs are evenly arranged on the blades of the fan, and the fan rotates at a preset speed to obtain a 3D image. 4. The virtual input method according to claim 2, wherein the holographic projection technology comprises: Using laser light as the illumination source, and dividing the light emitted by the illumination source into two beams, one beam is directed to the photosensitive sheet, and the other beam is directed to the photosensitive sheet after being reflected by the subject; The two beams of light are superimposed on the photosensitive sheet to generate interference; The hologram reproduced by the basic principle of digital image is further processed to remove digital interference and obtain a 3D image. 5. The virtual input method according to claim 1, wherein the inputting the image of the user operating the 3D input device captured by the camera into the image recognition model comprises: adding a preset number of images to the training set corresponding to the initial image recognition model; Using the initial image recognition model to train the image to obtain an image recognition model; The image is recognized by the image recognition model to obtain a recognition result. 6 . The virtual input method according to claim 5 , wherein the recognition result includes input content of the user operating the 3D input device, and the input content includes numbers, symbols, and letters. 7. The virtual input method according to claim 5, wherein said utilizing said initial image recognition model to train said image, and obtaining an image recognition model comprises: Select a sample image from the target training data set, input the current sample image to the initial image recognition model, and obtain a recognition result; Determine whether the recognition result is consistent with the result of the user operating the 3D input device, iteratively execute the above training process until the recognition result is consistent with the result of the user operating the 3D input device, and obtain a trained image recognition model. 8. The virtual input method according to claim 7, wherein the target training set includes images of a user operating a 3D input device.