CN111182202B - A wearable device-based content recognition method and wearable device - Google Patents

Abstract

A content identification method based on a wearable device and the wearable device are provided, the wearable device comprises a smart host and a host support, the smart host can rotate at any angle within 360 degrees when standing up and being vertical to the host support, the method comprises the following steps: when the intelligent host is erected to be vertical to the host bracket, detecting whether a touch area of a user is positioned at the edge of a shooting visual angle of any shooting module of the intelligent host; if yes, controlling the intelligent host to rotate to the position where the shooting central point of any shooting module of the intelligent host is closest to the touch area when the intelligent host stands to be vertical to the host bracket; controlling any shooting module of the intelligent host with the shooting central point closest to the touch area to shoot the touch area to obtain a shot image; and performing content identification on the shot image. According to the embodiment of the application, the accuracy rate of content identification by using wearable equipment can be improved.

Description

A wearable device-based content recognition method and wearable device

technical field

The present application relates to the technical field of wearable devices, and in particular, to a content recognition method based on a wearable device and a wearable device.

Background technique

Currently, more and more wearable devices (such as phone watches) are equipped with cameras, which can realize rich functions. In practice, it is found that when the camera of a wearable device (such as a phone watch) is used to realize the content recognition function, if the user's touch area of the camera's shooting angle is located at the edge of the shooting angle of view, the captured image is easily distorted, causing the shooting The content in the image is difficult to be accurately recognized by wearable devices (such as phone watches), which reduces the accuracy of content recognition.

SUMMARY OF THE INVENTION

The embodiments of the present application disclose a wearable device-based content recognition method and a wearable device, which can improve the accuracy of content recognition using the wearable device.

A first aspect of the embodiments of the present application discloses a content recognition method based on a wearable device, the wearable device includes a smart host and a host stand, and the smart host can generate a 360° range when standing vertically with the host stand Rotation at any angle within, the method includes:

When the smart host is upright and perpendicular to the host support, detecting whether the user's reach area is at the edge of the shooting angle of view of any shooting module of the smart host;

If so, control the intelligent host to rotate to the point where the shooting center point of any shooting module of the intelligent host is closest to the touch area when it stands up and is perpendicular to the host bracket;

and, controlling any shooting module of the intelligent host whose shooting center point is closest to the touched area to photograph the touched area to obtain a photographed image;

Perform content recognition on the captured image.

As an optional implementation manner, in the first aspect of the embodiment of the present application, after the content identification of the captured image, the method further includes:

outputting the identification content of the captured image to be displayed on the screen of the smart host;

The intelligent host is controlled to rotate so that the screen faces the face of the wearer of the wearable device, so as to facilitate the wearer to view the identification content of the captured image.

As another optional implementation manner, in the first aspect of the embodiment of the present application, the control of the smart host to rotate so that the screen faces the face of the wearer of the wearable device, so as to facilitate the wearing After the user checks the identification content of the captured image, the method further includes:

Obtain the operation instruction issued by the wearer;

If the operation instruction indicates that the identification content is to be read, extract the text content in the identification content;

and, acquiring the click-through content selected by the wearer from the text content;

The point-to-point content is broadcast.

As another optional implementation manner, in the first aspect of the embodiment of the present application, the method further includes:

If the operation instruction indicates that the identified content is to be searched, extracting the graphic and text content in the identified content;

Search the question and answer information that matches the content of the picture and text from the online question bank;

Outputting the question and answer information to a learning device connected to the wearable device for display.

As another optional implementation manner, in the first aspect of the embodiment of the present application, after extracting the graphic and text content in the identified content, a question matching the graphic and text content is searched from an online question bank Before the question and answer information, the method further includes:

Identify the semantics of the graphic content, and determine whether the graphic content contains complete title information according to the semantics; if not, adjust the shooting angle of any shooting module of the intelligent host to include the complete title information , and perform the step of searching the question and answer information that matches the content of the graphic and text from the online question bank.

A second aspect of the embodiments of the present application discloses a wearable device, the wearable device includes a smart host and a host stand, wherein the smart host can occur within a 360° range when the smart host is vertical to the host stand Rotation at any angle, the intelligent host includes:

a touch detection unit, configured to detect whether the user's touch area is at the edge of the shooting angle of view of any shooting module of the smart host when the smart host is upright and perpendicular to the host bracket;

The first control unit is used to control the intelligent host to stand up and the host bracket when the contact detection unit detects that the contact area of the user is at the edge of the shooting angle of any shooting module of the intelligent host. When vertical, the shooting center point of any shooting module rotated to the intelligent host is closest to the touch area;

a second control unit, configured to control any photographing module of the intelligent host whose photographing center point is closest to the reach area to photograph the reach area to obtain a photographed image;

A content recognition unit, configured to perform content recognition on the captured image.

As an optional implementation manner, in the second aspect of the embodiment of the present application, the intelligent host further includes:

a first output unit, configured to output the identification content of the captured image to display on the screen of the intelligent host after the content identification unit performs content identification on the captured image;

A third control unit, configured to control the intelligent host to rotate so that the screen faces the face of the wearer of the wearable device, so as to facilitate the wearer to view the identification content of the captured image.

As another optional implementation manner, in the second aspect of the embodiment of the present application, the intelligent host further includes:

a first acquisition unit, configured to control the smart host to rotate to the face of the wearer whose screen faces the wearable device in the third control unit, so as to facilitate the wearer to view the identification of the captured image After the content, obtain the operation instruction issued by the wearer;

a first extracting unit, configured to extract the text content in the identification content when the operation instruction issued by the wearer obtained by the first obtaining unit indicates that the identification content is to be read;

a second acquiring unit, configured to acquire the click-through content selected by the wearer from the text content;

The broadcasting unit is used for broadcasting the point-to-point content.

As another optional implementation manner, in the second aspect of the embodiment of the present application, the intelligent host further includes:

a second extraction unit, configured to extract the graphic and text content in the identification content when the operation instruction issued by the wearer obtained by the first obtaining unit indicates that the identification content is to be searched;

An online search unit, used for searching the question and answer information matching the content of the graphic and text from the online question bank;

The second output unit is configured to output the question and answer information to a learning device connected to the wearable device for display.

As another optional implementation manner, in the second aspect of the embodiment of the present application, the intelligent host further includes:

The semantic recognition unit is configured to, after the second extraction unit extracts the graphic and text content in the identified content, before the online search unit searches for question and answer information matching the graphic and text content from the online question bank, identify the The semantics of the graphic content, according to the semantics to determine whether the graphic content contains complete title information;

An adjustment unit, configured to adjust the shooting angle of any shooting module of the intelligent host to include the complete title information when the semantic recognition unit determines that the graphic and text content does not contain complete title information, and trigger all The online search unit searches the online question bank for question and answer information that matches the graphic content including the complete question information.

A third aspect of the embodiments of the present application discloses another wearable device. The wearable device includes a smart host and a host support. When the smart host is vertical to the host support, any Angle of rotation, the intelligent host includes:

a memory in which executable program code is stored;

a processor coupled to the memory;

The processor invokes the executable program code stored in the memory to execute all or part of the steps in any of the wearable device-based content identification methods disclosed in the first aspect of the embodiments of this application.

A fourth aspect of the embodiments of the present application discloses a computer-readable storage medium, which stores a computer program, wherein the computer program enables a computer to execute any wearable device-based content recognition disclosed in the first aspect of the embodiments of the present application all or part of the steps in the method.

A fifth aspect of the embodiments of the present application discloses a computer program product, which, when the computer program product runs on a computer, enables the computer to execute any one of the wearable device-based content identification methods of the first aspect of the embodiments of the present application all or part of the steps.

Compared with the prior art, the embodiment of the present application has the following beneficial effects: in the embodiment of the present application, the wearable device may include a smart host and a host stand, wherein the smart host can have a 360° range when standing vertically with the host stand. Rotation at any angle within the smart host; when the smart host is upright and perpendicular to the host bracket, it can detect whether the user's reach area is at the edge of the shooting angle of any shooting module of the smart host; if so, it can control the smart host in When standing vertically with the host bracket, the shooting center point of any shooting module rotated to the intelligent host is closest to the above-mentioned reach area; and, any shooting module of the intelligent host whose shooting center point is closest to the above-mentioned reach area can be controlled The above-mentioned touched area is photographed to obtain a photographed image; on this basis, the content of the photographed image is recognized. It can be seen that by implementing the embodiment of the present application, the shooting angle of the shooting module of the wearable device can be changed by rotating the smart host of the wearable device, so that the user touch area originally at the edge of the shooting angle of view can be as close to the shooting center point as possible, reducing the number of shots taken. The distortion generated by the image, thereby reducing the adverse effect of the distortion on the content recognition of the captured image, can improve the accuracy of the content recognition using the wearable device.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following will refer to the drawings that are used in the embodiments to those of ordinary skill in the art. Without creative work, the drawings can also be obtained from these drawings. Additional drawings.

FIG. 1 is a schematic flowchart of a content recognition method based on a wearable device disclosed in an embodiment of the present application;

2 is a schematic flowchart of another wearable device-based content recognition method disclosed in an embodiment of the present application;

3 is a schematic flowchart of another wearable device-based content recognition method disclosed in an embodiment of the present application;

4 is a schematic structural diagram of a wearable device disclosed in an embodiment of the present application;

5 is a schematic structural diagram of another wearable device disclosed in an embodiment of the present application;

6 is a schematic structural diagram of another wearable device disclosed in an embodiment of the present application;

FIG. 7 is a schematic structural diagram of another wearable device disclosed in an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

It should be noted that the terms "comprising" and "having" and any modifications thereof in the embodiments of the present application are intended to cover non-exclusive inclusion, for example, a process, method, system, product or process including a series of steps or units. The apparatus is not necessarily limited to those steps or units expressly listed, but may include other steps or units not expressly listed or inherent to the process, method, product or apparatus.

Furthermore, the terms "installed", "arranged", "provided", "connected", "connected" should be construed broadly. For example, it may be a fixed connection, a detachable connection, or a unitary structure; it may be a mechanical connection, or an electrical connection; it may be directly connected, or indirectly connected through an intermediary, or between two devices, elements, or components. internal communication. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific situations.

The embodiments of the present application disclose a wearable device-based content recognition method and a wearable device, which can improve the accuracy of content recognition using the wearable device. A detailed description is given below in conjunction with the accompanying drawings.

In order to better understand the control method of the wearable device disclosed in the embodiment of the present application, a wearable device to which the method is applicable is first described below. A wearable device to which the content identification method based on a wearable device disclosed in the embodiment of the present application is applicable may include a smart host, a host bracket, and a side strap; wherein, the first end of the host bracket and the first end of the side strap are connected by plugging and unplugging. The second end of the host bracket and the second end of the side strap are also connected by plugging and unplugging; wherein, the first end (rotating end) of the intelligent host is movably connected to the first end of the host bracket through the rotating ball, and the intelligent host The second end (free end) is not connected. Under normal circumstances, the smart host can be stacked on the host bracket, that is, the bottom side of the smart host of the smart host is in contact with the upper surface of the host bracket; and when the smart host can be flipped at different angles relative to the host bracket by rotating the ball , the bottom side of the smart host of the smart host and the upper surface of the host bracket form a certain angle (the angle can be adjusted between 0 and 180°).

When the smart host is vertical to the host bracket (that is, the bottom side of the smart host of the smart host and the upper surface of the host bracket are at 90°), the smart host can also rotate at any angle within 360° through the rolling ball.

It should be noted that the wearable device described above is only one implementation of the wearable device to which the wearable device-based content identification method disclosed in the embodiments of the present application is applicable, and should not be construed as a limitation of the present application.

Please refer to FIG. 1 . FIG. 1 is a schematic flowchart of a content recognition method based on a wearable device disclosed in an embodiment of the present application. Wherein, the wearable device includes a smart host and a host bracket, and the smart host can rotate at any angle within a range of 360° when the smart host is vertical to the host bracket. As shown in Figure 1, the content identification method may include the following steps:

101. When the smart host of the wearable device is erected perpendicular to its host stand, the wearable device detects whether the user's reach area is at the edge of the shooting angle of any shooting module of the wearable device, and if so, perform steps 102 to 102 104; otherwise, end the process.

In the embodiment of the present application, when the smart host of the wearable device is erected perpendicular to its host bracket, any shooting module of the smart host can be in a pre-shooting state, and output its shooting angle to display on the screen of the smart host. For example, it can be that the rear camera module of the smart host is in a pre-shooting state, and outputs its shooting angle to display on the screen of the smart host; for another example, it can be the front camera module of the smart host It is in the pre-shooting state at the same time as the rear camera module. On this basis, if any camera module detects the face of the wearer of the wearable device, it outputs the camera angle of the other camera module to the smart host. displayed on the screen.

Further, the above-mentioned detection of whether the touched area of the user is at the edge of the shooting angle of view of any shooting module of its smart host may specifically include: detecting the user's touching operation on the screen of the smart host; when detecting the above-mentioned touching operation, Obtain the touch area of the touch operation; determine whether the distance between the touch area and the shooting center point of any shooting module of the above-mentioned smart host is greater than a preset threshold, if it is greater than, it can be judged that the touch area is in the above-mentioned smart host. the edge of the shooting angle of view of any shooting module, and then step 102 to step 104 are performed.

102. The wearable device controls its smart host to rotate when the smart host stands upright and its host stand is perpendicular to the host stand to rotate so that the shooting center point of any shooting module of the smart host is closest to the above-mentioned touch area.

In the embodiment of the present application, the wearable device can calculate the target rotation direction and target rotation angle of the smart host according to the coordinate information of the above-mentioned reach area and the current rotation angle (within a 360° range) of the smart host; and, according to the above target The rotation direction and the target rotation angle are used to control the intelligent host to perform corresponding rotation, so that the shooting center point of any shooting module of the intelligent host is closest to the above-mentioned touch area.

By implementing the above method, the user-touched area that is originally at the edge of the shooting angle of view can be made as close to the shooting center point as possible, reducing the distortion of the captured image, and further reducing the distortion caused by the following steps 103 to 104. Therefore, it is possible to improve the accuracy of content recognition using the wearable device.

103. The wearable device controls any photographing module of the smart host whose photographing center point is closest to the reach area to photograph the reach area to obtain a photographed image.

As an optional implementation manner, the wearable device may detect a voice keyword sent by the user, and perform a corresponding shooting operation according to the voice keyword. For example, when the voice keyword "photograph" is detected, any shooting module of the smart host whose shooting center point is closest to the above-mentioned touched area can be controlled to photograph the above-mentioned touched area to obtain a photographed image; When the voice keyword is "continuous shooting", any of the above shooting modules can be controlled to shoot continuously according to the preset shooting times and shooting time interval; for another example, when the voice keyword "after 2 seconds" and "photographing" are detected , you can control any of the above shooting modules to shoot after a delay of 2 seconds.

As another optional implementation manner, the wearable device may also perform a corresponding shooting operation according to the user's touching operation on the screen of the smart host. For example, when the above-mentioned touching operation is a single touch, any shooting module of the smart host whose shooting center point is closest to the above-mentioned touching area can be controlled to photograph the above-mentioned touching area to obtain a photographed image; for another example, when the above-mentioned touching operation is performed When it is a double-click, any one of the above-mentioned shooting modules can be controlled to shoot after a preset time delay; for another example, when the above-mentioned touch operation is a long press, one of the above-mentioned shooting modules can be controlled to automatically focus before shooting.

104. The wearable device performs content recognition on the captured image.

In the embodiment of the present application, the above-mentioned wearable device performing content recognition on the captured image may specifically include: sequentially passing the captured image through a preprocessing model and a content recognition model to obtain the recognition content of the captured image. The above preprocessing model may perform steps such as image and text segmentation, grayscale, binarization, noise reduction, deviation correction, character cutting, and text merging, which are not specifically limited in this embodiment of the present application.

It can be seen that implementing the content recognition method described in FIG. 1 can improve the accuracy of content recognition using the wearable device.

Please refer to FIG. 2. FIG. 2 is a schematic flowchart of another method for content recognition based on a wearable device disclosed in an embodiment of the present application. Wherein, the wearable device includes a smart host and a host bracket, and the smart host can rotate at any angle within a range of 360° when the smart host is vertical to the host bracket. As shown in Figure 2, the content identification method may include the following steps:

201. When the smart host of the wearable device is erected perpendicular to its host stand, the wearable device detects whether the user's reach area is at the edge of the shooting angle of any shooting module of the wearable device, and if so, perform steps 202 to 202. 210; otherwise, end the process.

202 , the wearable device controls its smart host to rotate when the smart host stands upright and its host stand is perpendicular to the host stand, so that the shooting center point of any shooting module of the smart host is closest to the above-mentioned touch area.

203. The wearable device controls any shooting module of the smart host whose shooting center point is closest to the touched area to photograph the touched area to obtain a photographed image.

204. The wearable device performs content recognition on the captured image.

205. The wearable device outputs the identification content of the captured image to display on the screen of its smart host.

As an optional implementation manner, the wearable device may also output the identification content of the captured image to be displayed on a learning device connected to the wearable device. Exemplarily, the above learning device may include various types of devices or systems having a display screen (eg, a learning tablet, a smart speaker with a screen, etc.), which is not specifically limited in the embodiment of the present application.

206. The wearable device controls its smart host to rotate so that the above-mentioned screen faces the face of the wearer of the wearable device, so as to facilitate the wearer to view the identification content of the above-mentioned captured image.

For example, the wearable device can determine whether any shooting module of the above-mentioned smart host detects the face of the wearer of the wearable device; if detected, obtain the direction of the wearer's face, and control the The smart host is rotated to this direction, so that the screen of the smart host faces the face of the wearer, so that the wearer can view the identification content of the captured image.

For another example, the wearable device can also detect the voice command issued by the user, and determine whether the voiceprint feature of the voice command matches the voiceprint feature of the wearer of the wearable device; if it matches, obtain and issue the voice. The direction of the sound source of the command, and control the above-mentioned smart host to rotate to this direction.

207. The wearable device acquires the operation instruction issued by the wearer.

Exemplarily, the wearable device can detect the voice command issued by the wearer, and obtain the operation command corresponding to the voice command through keyword detection technology or natural language processing technology; the wearable device can also detect the wearer's voice command. The touch operation on the screen of the smart host of the device, and the operation instruction corresponding to the touch area of the touch operation is obtained.

208. If the operation instruction indicates that the above-mentioned identification content is to be read, the wearable device extracts the text content in the above-mentioned identification content.

209. The wearable device acquires the click-through content selected by the wearer from the text content.

210. The wearable device broadcasts the read content.

By implementing the above method, the wearable device can realize the point-to-read function of the recognized content on the basis of performing content recognition on the captured image, thereby expanding the usage scenarios of the wearable device and improving its practicability.

As an optional implementation manner, after the wearable device performs steps 201 to 210 to broadcast the above-mentioned point-to-read content, the wearable device may also perform the following steps: when the above-mentioned point-to-read content is an English word, it can search for the English word on the Internet. Relevant information, wherein the relevant information can include the pronunciation, translation, and example sentences of the English word; output the above-mentioned relevant information to the screen of the smart host of the wearable device, and display it near the English word in the form of a floating window; broadcast in turn The relevant information; further, the wearable device can also detect the follow-up reading of the English word by the wearer, judge whether the follow-up reading is correct, and output a corresponding score. By implementing the above method, the wearer of the wearable device can be helped to conveniently consult and translate English learning materials, and at the same time, unfamiliar English words can be learned, which improves the practicability of the wearable device.

It can be seen that, implementing the content recognition method described in FIG. 2 can improve the accuracy of content recognition using the wearable device.

In addition, by implementing the content recognition method described in FIG. 2 , the wearable device can be used to realize the point-to-point reading function, thereby expanding the usage scenarios of the wearable device and improving its practicability.

Please refer to FIG. 3 . FIG. 3 is a schematic flowchart of another method for content recognition based on a wearable device disclosed in an embodiment of the present application. Wherein, the wearable device includes a smart host and a host bracket, and the smart host can rotate at any angle within a range of 360° when the smart host is vertical to the host bracket. As shown in Figure 3, the content identification method may include the following steps:

301. When the smart host of the wearable device is erected perpendicular to its host bracket, the wearable device detects whether the user's reach area is at the edge of the shooting angle of any shooting module of the wearable device, and if so, perform steps 302 to 301. 309; otherwise, end the process.

302 , the wearable device controls its smart host to rotate when the smart host stands upright to its host stand to rotate so that the shooting center point of any shooting module of the smart host is closest to the above-mentioned touch area.

303. The wearable device controls any shooting module of the smart host whose shooting center point is closest to the touched area to photograph the touched area to obtain a photographed image.

304. The wearable device performs content recognition on the captured image.

305. The wearable device outputs the identification content of the captured image to display on the screen of its smart host.

306. The wearable device controls its smart host to rotate so that the above-mentioned screen faces the face of the wearer of the wearable device, so as to facilitate the wearer to view the identification content of the above-mentioned captured image.

307. The wearable device acquires the operation instruction issued by the wearer.

308. If the operation instruction indicates that the above-mentioned identification content is to be searched, the wearable device extracts the graphic and text content in the above-mentioned identification content.

309. The wearable device recognizes the semantics of the graphic content, and judges whether the graphic content contains complete topic information according to the semantics. If not, perform steps 310 to 312; otherwise, perform steps 311 to 312.

Exemplarily, when the wearable device detects that the graphic content contains incomplete characters or sentences that do not meet the preset grammatical standard, it can be determined that the graphic content does not contain complete topic information.

310. The wearable device adjusts the shooting angle of any shooting module of its smart host to include the above-mentioned complete topic information.

Exemplarily, the wearable device can expand the shooting angle of any shooting module of its smart host to include the above-mentioned complete topic information; it can also rotate its smart host, and continuously monitor any shooting module during the rotation process. , until it contains the complete topic information above.

311. The wearable device searches the online question bank for question and answer information that matches the above-mentioned graphic and text content.

312. The wearable device outputs the question and answer information to the learning device connected to the wearable device for display.

In the embodiment of the present application, by performing the above steps 308 to 312, the wearable device can realize the function of taking pictures and searching for questions, thereby helping the wearer of the wearable device to perform autonomous learning and improving learning enthusiasm.

As an optional implementation manner, when the wearable device performs the above step 308, if the obtained operation instruction issued by the wearer of the wearable device indicates that the above identification content is to be searched and corrected, the following steps may also be performed : distinguish and extract the question information and answer information in the above-mentioned identification content respectively; search the question and answer information matching the above question information from the online question bank; according to the question and answer information, correct the above answer information to obtain the correction result; The correction result is output and displayed on the learning device connected with the wearable device.

Further, if the above correction result indicates that there is an error in the above answering information, the wrong answering information may also be highlighted on the above learning device to facilitate the user to view.

By implementing the above method, the wearer of the wearable device can be further helped to perform autonomous learning, and the ability and enthusiasm of the wearer to solve problems independently can be cultivated.

It can be seen that implementing the content recognition method described in FIG. 3 can improve the accuracy of content recognition using the wearable device.

In addition, by implementing the content recognition method described in FIG. 3 , the wearable device can be used to realize a question search function, thereby helping the wearer of the wearable device to perform autonomous learning and improving learning enthusiasm.

Please refer to FIG. 4 , which is a schematic structural diagram of a wearable device disclosed in an embodiment of the present application. Wherein, the wearable device includes a smart host and a host bracket, and the smart host can rotate at any angle within a range of 360° when the smart host is vertical to the host bracket. As shown in Figure 4, the smart host of the wearable device may include:

The touch detection unit 401 is used to detect whether the touch area of the user is at the edge of the shooting angle of view of any shooting module of the smart host when the smart host of the wearable device is erected perpendicular to its host bracket;

The first control unit 402 is used to control the intelligent host to stand up and be perpendicular to the above-mentioned host bracket when the above-mentioned touch detection unit 401 detects that the touched area of the user is at the edge of the shooting angle of any shooting module of the above-mentioned intelligent host. The shooting center point of any shooting module rotated to the intelligent host is closest to the above-mentioned touch area;

The second control unit 403 is configured to control any shooting module of the intelligent host whose shooting center point is closest to the touched area to photograph the touched area to obtain a photographed image;

The content recognition unit 404 is configured to perform content recognition on the captured image.

It can be seen that, implementing the wearable device described in FIG. 4 can make the user touch area that is originally at the edge of the shooting angle of view as close to the shooting center point as possible, reduce the distortion of the captured image, and then use the content recognition unit 404 for the captured image. When performing content recognition, the adverse effects caused by distortion are reduced, so that the accuracy of content recognition using wearable devices can be improved.

Please also refer to FIG. 5 , which is a schematic structural diagram of another wearable device disclosed in an embodiment of the present application. The wearable device shown in FIG. 5 is obtained by optimizing the wearable device shown in FIG. 4 . Compared with the wearable device shown in Figure 4, the smart host of the wearable device shown in Figure 5 further includes:

The first output unit 405 is configured to output the identification content of the captured image to display on the screen of the smart host of the wearable device after the content recognition unit 404 performs content recognition on the captured image;

The third control unit 406 is configured to control the above-mentioned intelligent host to rotate so that the above-mentioned screen faces the face of the wearer of the wearable device, so as to facilitate the wearer to view the identification content of the above-mentioned captured image.

The first obtaining unit 407 is configured to obtain the above-mentioned screen after the third control unit 406 controls the above-mentioned smart host to rotate to the face of the wearer whose screen is facing the wearable device, so as to facilitate the wearer to view the identification content of the above-mentioned captured image. the operating instructions issued by the wearer;

The first extracting unit 408 is configured to extract the text content in the identification content when the operation instruction issued by the wearer obtained by the first obtaining unit 407 indicates that the identification content is to be read;

The second obtaining unit 409 is configured to obtain the click-through content selected by the wearer from the text content;

The broadcasting unit 410 is used for broadcasting the point-to-point content.

As an optional implementation manner, after the broadcasting unit 410 broadcasts the above-mentioned point reading content, if the point reading content is an English word, the wearable device can also: search online for relevant information of the English word, wherein the relevant information Can include the pronunciation, translation, and example sentences of the English word; on this basis, the first output unit 405 outputs the above-mentioned relevant information to the screen of the smart host of the wearable device, and is displayed near the English word in the form of a floating window; The broadcast unit 410 broadcasts the relevant information in turn; further, the wearable device can also detect the follow-up reading of the English word by its wearer, and judge whether the follow-up reading is correct, and output the corresponding score through the first output unit 405 or the broadcast unit 410 . By implementing the above method, on the basis of using the wearable device to realize the point-to-read function, the wearer of the wearable device can be helped to conveniently consult and translate English learning materials, and at the same time, unfamiliar English words can be learned, which improves the wearability of the wearable device. Availability of equipment.

It can be seen that implementing the wearable device described in FIG. 5 can improve the accuracy of content recognition using the wearable device.

In addition, by implementing the wearable device described in Figure 5, the wearable device can be used to realize the point-and-read function, thereby expanding the usage scenarios of the wearable device and improving its practicability

Please refer to FIG. 6 together. FIG. 6 is a schematic structural diagram of another wearable device disclosed in an embodiment of the present application. The wearable device shown in FIG. 6 is obtained by optimizing the wearable device shown in FIG. 5 . Compared with the wearable device shown in Figure 5, the smart host of the wearable device shown in Figure 6 further includes:

The second extraction unit 411 is configured to extract the graphic and text content in the identification content when the operation instruction issued by the wearer obtained by the first obtaining unit 407 indicates that the identification content is to be searched;

The semantic recognition unit 412 is used for identifying the semantics of the graphic content after the above-mentioned second extraction unit 411 extracts the graphic content in the identified content, and judges whether the graphic content contains complete title information according to the semantics;

The adjustment unit 413 is used to adjust the shooting angle of any shooting module of the smart host of the wearable device to include the complete title information when the above-mentioned semantic recognition unit 412 judges that the graphic content does not contain the complete title information;

The online search unit 414 is used for when the above-mentioned semantic recognition unit 412 judges that the graphic content contains complete title information, and when the above-mentioned semantic recognition unit 412 judges that the graphic content does not contain complete title information, the above-mentioned adjustment unit 413 adjusts the wearable When the shooting angle of any shooting module of the smart host of the device contains the complete question information, search the question and answer information matching the graphic and text content from the online question bank;

The second output unit 415 is configured to output the question and answer information to a learning device connected to the wearable device for display.

In the embodiment of the present application, by implementing the above-mentioned wearable device, the function of taking pictures and searching for questions can be realized, thereby helping the wearer of the wearable device to perform autonomous learning and improving learning enthusiasm.

As an optional implementation manner, if the operation instruction issued by the wearer of the wearable device obtained by the first obtaining unit 407 indicates that the above identification content is to be searched and corrected, the second obtaining unit 411 can also determine and correct Extract the question information and answer information in the above-mentioned identification content respectively; Next, the online search unit 414 searches the question and answer information matching the above-mentioned question information from the online question bank; according to the question and answer information, the wearable device can search for the above-mentioned answer information Correction is performed to obtain a correction result; on this basis, the second output unit 415 outputs the correction result to a learning device connected to the wearable device for display.

Further, if the correction result indicates that there is an error in the answering information, the second output unit 415 may also highlight the answering information with an error on the learning device, so as to facilitate the user to check.

By implementing the wearable device, it can further help the wearer of the wearable device to conduct autonomous learning, and cultivate the wearer's ability and enthusiasm to solve problems independently.

It can be seen that implementing the wearable device described in FIG. 6 can improve the accuracy of content recognition using the wearable device.

In addition, by implementing the wearable device described in FIG. 6 , the wearable device can be used to realize the function of searching questions, thereby helping the wearer of the wearable device to perform autonomous learning and improving learning enthusiasm.

Please refer to FIG. 7 , which is a schematic structural diagram of another wearable device disclosed in an embodiment of the present application. Wherein, the wearable device includes a smart host and a host bracket, and the smart host can rotate at any angle within a range of 360° when the smart host is vertical to the host bracket. As shown in Figure 7, the smart host may include:

a memory 701 storing executable program code;

a processor 702 coupled to the memory 701;

The processor 702 invokes the executable program code stored in the memory 701 to execute all or part of the steps in any of the wearable device-based content identification methods shown in FIGS. 1 to 3 .

In addition, an embodiment of the present application further discloses a computer-readable storage medium, which stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute any of the wearable device-based content shown in FIGS. 1 to 3 . Identify all or some of the steps in the method.

In addition, the embodiment of the present application further discloses a computer program product, when the computer program product runs on a computer, all or part of the steps in any one of the wearable device-based content recognition methods shown in FIG. 1 to FIG. 3 on the computer.

Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium includes a read-only storage medium. Memory (Read-Only Memory, ROM), Random Access Memory (Random Access Memory, RAM), Programmable Read-only Memory (PROM), Erasable Programmable Read Only Memory (Erasable Programmable Read Only Memory, EPROM), One-time Programmable Read-Only Memory (OTPROM), Electronically-Erasable Programmable Read-Only Memory (EEPROM), CompactDisc Read -Only Memory, CD-ROM) or other optical disk storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

A wearable device-based content recognition method and a wearable device disclosed in the embodiments of the present application are described above in detail, and specific examples are used in this paper to illustrate the principles and implementations of the present invention. The descriptions of the above embodiments are only It is used to help understand the method of the present invention and its core idea; at the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific embodiments and application scope. The contents of the description should not be construed as limiting the present invention.

Claims (12)

1. A content recognition method based on a wearable device, wherein the wearable device comprises a smart host and a host bracket, the smart host can rotate within 360 degrees in any angle when standing up to be perpendicular to the host bracket, and the method comprises the following steps:

detecting touch operation of a user on a screen of the intelligent host when the intelligent host is erected to be vertical to the host bracket; when the touch operation is detected, acquiring a touch area of the touch operation;

when the distance between the touch area and the shooting center point of any shooting module of the intelligent host is larger than a preset threshold value, determining that the touch area is located at the edge of the shooting view angle of any shooting module of the intelligent host;

controlling the intelligent host to rotate to a shooting central point of any shooting module of the intelligent host to be closest to the touch area when the intelligent host is erected to be vertical to the host bracket;

any shooting module of the intelligent host with the shooting central point closest to the touch area is controlled to shoot the touch area, and a shot image is obtained;

and sequentially passing the shot image through a preprocessing model and a content identification model to obtain the identification content of the shot image.

2. The content recognition method according to claim 1, wherein after obtaining the recognition content of the captured image, the method further comprises:

outputting the identification content of the shot image to a screen of the intelligent host for display;

controlling the intelligent host to rotate to the face of the screen facing the wearer of the wearable device, so that the wearer can conveniently view the identification content of the shot image.

3. The content recognition method according to claim 2, wherein after controlling the smart host to rotate to face the screen toward the wearer of the wearable device so as to facilitate the wearer to view the recognition content of the captured image, the method further comprises:

acquiring an operation instruction sent by the wearer;

if the operation instruction indicates that the identification content is read, extracting text content in the identification content;

and acquiring click-to-read content selected by the wearer from the text content;

and broadcasting the point reading content.

4. The content recognition method of claim 3, further comprising:

if the operation instruction indicates that the topic searching is carried out on the identification content, extracting the image-text content in the identification content;

searching question and answer information matched with the image-text content from a networked question bank;

and outputting the question answering information to learning equipment connected with the wearable equipment for display.

5. The method of claim 4, wherein after extracting the teletext content of the identified content and before searching a networked question bank for question and answer information matching the teletext content, the method further comprises:

identifying the semantics of the image-text content, and judging whether the image-text content contains complete title information according to the semantics; if not, adjusting the shooting visual angle of any shooting module of the intelligent host to contain the complete question information, and executing the step of searching question and answer information matched with the image-text content from a networked question bank.

6. A wearable device comprising a smart host and a host stand, wherein the smart host is capable of any angle of rotation within a 360 ° range when standing up perpendicular to the host stand, the smart host comprising:

the touch detection unit is used for detecting touch operation of a user on a screen of the intelligent host when the intelligent host is erected to be vertical to the host bracket; when the touch operation is detected, acquiring a touch area of the touch operation; when the distance between the touch area and the shooting center point of any shooting module of the intelligent host is larger than a preset threshold value, determining that the touch area of the user is located at the edge of the shooting view angle of any shooting module of the intelligent host;

the first control unit is used for controlling the intelligent host to rotate to a position that a shooting central point of any shooting module of the intelligent host is closest to the touch area when the intelligent host is erected to be vertical to the host bracket when the touch detection unit detects that the touch area of a user is located at the edge of a shooting visual angle of any shooting module of the intelligent host;

the second control unit is used for controlling any shooting module of the intelligent host with the shooting central point closest to the touch area to shoot the touch area to obtain a shot image;

and the content identification unit is used for sequentially passing the shot image through the preprocessing model and the content identification model to obtain the identification content of the shot image.

7. The wearable device of claim 6, wherein the smart host further comprises:

the first output unit is used for outputting the identification content of the shot image to the screen of the intelligent host for display after the content identification unit obtains the identification content of the shot image;

and the third control unit is used for controlling the intelligent host to rotate to the face of the screen facing the wearer of the wearable device, so that the wearer can conveniently view the identification content of the shot image.

8. The wearable device of claim 7, wherein the smart host further comprises:

the first obtaining unit is used for obtaining an operation instruction sent by the wearer after the third control unit controls the intelligent host to rotate to the face of the wearer with the screen facing the wearable device so that the wearer can conveniently view the identification content of the shot image;

a first extracting unit, configured to extract text content in the identification content when the operation instruction sent by the wearer and acquired by the first acquiring unit indicates to perform point reading on the identification content;

a second acquisition unit configured to acquire click-to-read content selected by the wearer from the text content;

and the broadcasting unit is used for broadcasting the point-reading content.

9. The wearable device of claim 8, wherein the smart host further comprises:

the second extraction unit is used for extracting the image-text content in the identification content when the operation instruction sent by the wearer and acquired by the first acquisition unit indicates to search the identification content;

the network searching unit is used for searching question and answer information matched with the image-text content from a network question bank;

and the second output unit is used for outputting the question answering information to the learning equipment connected with the wearable equipment for display.

10. The wearable device of claim 9, wherein the smart host further comprises:

the semantic recognition unit is used for recognizing the semantics of the image-text content and judging whether the image-text content contains complete question information according to the semantics before the networked searching unit searches question and answer information matched with the image-text content from a networked question bank after the second extraction unit extracts the image-text content in the recognized content;

and the adjusting unit is used for adjusting the shooting visual angle of any shooting module of the intelligent host to contain the complete topic information when the semantic identification unit judges that the image-text content does not contain the complete topic information, and triggering the networking searching unit to search and contain the image-text content matched topic question-answer information of the complete topic information from the networking topic library.

11. A wearable device comprising a smart host and a host stand, wherein the smart host is capable of any angle of rotation within a 360 ° range when standing up perpendicular to the host stand, the smart host comprising:

a memory storing executable program code;

a processor coupled with the memory;

the processor calls the executable program code stored in the memory to execute the wearable device-based content identification method according to any one of claims 1 to 5.

12. A computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute the wearable-device-based content identification method according to any one of claims 1 to 5.

Images