CN113837986A

CN113837986A - Method, apparatus, electronic device, and medium for recognizing tongue picture

Info

Publication number: CN113837986A
Application number: CN202011473756.XA
Authority: CN
Inventors: 田洪宝; 狄帅; 何伟华; 谷超; 黄闻
Original assignee: Jingdong Technology Holding Co Ltd
Current assignee: Jingdong Technology Holding Co Ltd
Priority date: 2020-12-15
Filing date: 2020-12-15
Publication date: 2021-12-24

Abstract

Embodiments of the present disclosure disclose methods, apparatuses, electronic devices, and media for recognizing tongue manifestation. One embodiment of the method comprises: acquiring a tongue image to be identified; extracting a first image characteristic of the tongue image to be recognized, wherein the first image characteristic belongs to a traditional image characteristic; extracting a second image characteristic of the tongue image to be recognized by utilizing a pre-trained neural network; and generating health state information prompted by the tongue image to be recognized based on the matching of the first image feature, the second image feature and a preset tongue image database, wherein the tongue image database comprises a corresponding relation between the image features and description labels for indicating the health state. The embodiment makes full use of the advantages of the local features and the global features, and improves the matching degree of the generated health state information and the tongue image to be recognized.

Description

Method, apparatus, electronic device, and medium for recognizing tongue picture

Technical Field

Embodiments of the present disclosure relate to the field of computer technologies, and in particular, to a method, an apparatus, an electronic device, and a medium for recognizing a tongue picture.

Background

With the development of artificial intelligence technology, the application of computer vision technology to identification and analysis of medical images is increasing.

In the prior art, the tongue picture is recognized by analyzing the artificial design characteristics (such as color, texture, shape, etc.) of the tongue coating part mainly through computer vision, and then corresponding symptoms are obtained according to related experiences. However, the method of associating symptoms according to related experiences through simple feature extraction often results in inaccuracy due to the fact that the association relationship is not one-to-one. If the existing image classification method is directly applied, accurate classification is difficult to realize due to high tongue similarity.

Disclosure of Invention

Embodiments of the present disclosure propose a method, apparatus, electronic device, and medium for recognizing tongue picture.

In a first aspect, embodiments of the present disclosure provide a method for recognizing tongue images, the method including: acquiring a tongue image to be identified; extracting first image features of a tongue image to be recognized, wherein the first image features belong to traditional image features; extracting a second image characteristic of the tongue image to be recognized by utilizing a pre-trained neural network; and generating health state information prompted by the tongue image to be recognized based on the matching of the first image characteristic and the second image characteristic with a preset tongue image database, wherein the tongue image database comprises a corresponding relation between the image characteristics and a description label for indicating the health state.

In a second aspect, embodiments of the present disclosure provide a method for recognizing tongue images, the method including: acquiring a tongue image to be identified; the tongue image to be identified is sent to a target server; receiving health state information sent by a target server; and displaying the tongue picture recognition result based on the health state information.

In a third aspect, an embodiment of the present disclosure provides an apparatus for recognizing tongue manifestation, the apparatus including: a first acquisition unit configured to acquire a tongue image to be recognized; a first extraction unit configured to extract a first image feature of a tongue image to be recognized, wherein the first image feature belongs to a conventional image feature; a second extraction unit configured to extract a second image feature of the tongue image to be recognized by using a pre-trained neural network; the generating unit is configured to generate health status information prompted by a to-be-recognized tongue image based on matching of the first image feature and the second image feature with a preset tongue image database, wherein the tongue image database comprises a corresponding relation between the image features and description labels used for indicating health statuses.

In a fourth aspect, embodiments of the present disclosure provide an apparatus for recognizing tongue images, the apparatus including: a second acquisition unit configured to acquire a tongue image to be recognized; the third sending unit is configured to send the tongue image to be identified to the target server; the second receiving unit is configured to receive the health state information sent by the target server; a display unit configured to display a tongue picture recognition result based on the health status information.

In a fifth aspect, an embodiment of the present disclosure provides an electronic device, including: one or more processors; a storage device having one or more programs stored thereon; when executed by one or more processors, cause the one or more processors to implement a method as described in any one of the implementations of the first and second aspects.

In a sixth aspect, embodiments of the present disclosure provide a computer readable medium, on which a computer program is stored, which when executed by a processor, implements a method as described in any of the implementations of the first and second aspects.

According to the method, the device, the electronic equipment and the medium for identifying the tongue picture, the artificial design features of the image and the features extracted by the neural network are fused to serve as the basis for image matching, so that the advantages of the local features and the global features are fully utilized, and the matching degree of the generated health state information and the tongue picture to be identified is improved. And moreover, the accuracy of tongue picture identification is improved by carrying out corresponding processing from the aspects of image acquisition, result display and the like.

Drawings

Other features, objects and advantages of the disclosure will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:

FIG. 1 is an exemplary system architecture diagram in which one embodiment of the present disclosure may be applied;

FIG. 2 is a flow diagram of one embodiment of a method for identifying a tongue representation according to the present disclosure;

FIG. 3 is a schematic diagram of one application scenario of a method for recognizing tongue manifestation, according to an embodiment of the present disclosure;

FIG. 4 is a flow chart of yet another embodiment of a method for identifying a tongue representation according to the present disclosure;

FIG. 5 is a schematic diagram illustrating the structure of one embodiment of an apparatus for identifying tongue images according to the present disclosure;

FIG. 6 is a schematic diagram illustrating the structure of one embodiment of an apparatus for identifying tongue images according to the present disclosure;

FIG. 7 is a schematic structural diagram of an electronic device suitable for use in implementing embodiments of the present disclosure.

Detailed Description

The present disclosure is described in further detail below with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings.

It should be noted that, in the present disclosure, the embodiments and features of the embodiments may be combined with each other without conflict. The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

Fig. 1 illustrates an exemplary architecture 100 to which the method for identifying a tongue image or the apparatus for identifying a tongue image of the present disclosure may be applied.

As shown in fig. 1, the system architecture 100 may include

terminal devices

101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the

terminal devices

101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

The

terminal devices

101, 102, 103 interact with a server 105 via a network 104 to receive or send messages or the like. The

terminal devices

101, 102, 103 may have various communication client applications installed thereon, such as a web browser application, a shopping application, a search application, an instant messaging tool, a mailbox client, an image recognition application, and the like.

The

terminal apparatuses

101, 102, and 103 may be hardware or software. When the

terminal devices

101, 102, 103 are hardware, they may be various electronic devices having a display screen and supporting human-computer interaction, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like. When the

terminal apparatuses

101, 102, 103 are software, they can be installed in the electronic apparatuses listed above. It may be implemented as multiple pieces of software or software modules (e.g., software or software modules used to provide distributed services) or as a single piece of software or software module. And is not particularly limited herein.

The server 105 may be a server providing various services, such as a background server providing support for image recognition type applications on the

terminal devices

101, 102, 103. The background server may analyze the received tongue image to be recognized, generate a processing result (for example, health status information prompted by the tongue image to be recognized), and may also feed back the generated processing result to the terminal device.

The server may be hardware or software. When the server is hardware, it may be implemented as a distributed server cluster formed by multiple servers, or may be implemented as a single server. When the server is software, it may be implemented as multiple pieces of software or software modules (e.g., software or software modules used to provide distributed services), or as a single piece of software or software module. And is not particularly limited herein.

It should be noted that the method for recognizing the tongue image as described in the foregoing first aspect provided by the embodiment of the present disclosure is generally performed by the server 105, and accordingly, the apparatus for recognizing the tongue image is generally disposed in the server 105. The method for recognizing tongue images as described in the foregoing second aspect provided by the embodiments of the present disclosure is generally performed by the

terminal devices

101, 102, 103, and accordingly, the means for recognizing tongue images is generally provided in the

terminal devices

101, 102, 103. It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

With continued reference to FIG. 2, a flow 200 of one embodiment of a method for identifying a tongue manifestation in accordance with the present disclosure is shown. The method for recognizing the tongue picture comprises the following steps:

step 201, a tongue image to be recognized is obtained.

In the present embodiment, the executing body (such as the server 105 shown in fig. 1) of the method for recognizing the tongue image may acquire the tongue image to be recognized by a wired connection manner or a wireless connection manner. The tongue image to be recognized may be an original image captured by a user with respect to a tongue, or may be an image of a protruding tongue body obtained by processing the original image. As an example, the executing body may acquire the to-be-recognized tongue image which is stored locally in advance, or may acquire the to-be-recognized tongue image from an electronic device (for example,

terminal devices

101, 102, 103 shown in fig. 1) which is connected to the executing body in communication.

In some optional implementations of the present embodiment, the executing body may acquire the tongue image to be recognized by:

first, an acquired initial tongue image is acquired.

In these implementations, the performing agent may first acquire the acquired initial tongue image in various ways. The initial tongue image may be an unprocessed image captured for the tongue by a camera. In practice, the initial tongue image may be affected by the captured light to cause color distortion, and may also include non-tongue regions.

And secondly, preprocessing the initial tongue image to generate a tongue image to be identified.

In these implementations, the executing body may pre-process the initial tongue image acquired in the first step to generate a tongue image to be recognized. Wherein the pre-treatment may comprise at least one of: tongue color correction, tongue segmentation.

Based on the optional implementation mode, the scheme can reduce the deviation of illumination on colors through a color correction algorithm, and remove irrelevant areas through an image segmentation algorithm so as to reduce the adverse effect on subsequent matching caused by extracting the features of the non-tongue area.

Step 202, extracting a first image feature of the tongue image to be recognized.

In this embodiment, the executing body may extract the feature of the tongue image to be recognized acquired in step 201 as the first image feature in various ways. Wherein, the first image feature generally belongs to a traditional image feature. The conventional image features may include, but are not limited to, at least one of: color features, texture features, shape features.

And step 203, extracting a second image characteristic of the tongue image to be recognized by using a pre-trained neural network.

In this embodiment, the executing agent may extract the second image feature of the tongue image to be recognized by using a pre-trained neural network. The pre-trained neural network may include various convolutional neural networks for extracting image features. As an example, the convolutional neural network may be trained using a training sample composed of a sample tongue image and corresponding sample description information to adjust network parameters of the convolutional neural network. Thus, the second image feature may include an image feature extracted through a neural network to be distinguished from a conventional artificially constructed image feature.

And 204, generating health state information prompted by the tongue image to be recognized based on the matching of the first image characteristic and the second image characteristic with a preset tongue image database.

In this embodiment, based on the matching of the first image feature extracted in step 202 and the second image feature extracted in step 203 with the preset tongue image database, the executing body may generate the health status information prompted by the tongue image to be recognized in various ways. The tongue image database may include a correspondence between image features and descriptive tags indicating health status. The description labels may include, for example, body quality determination labels such as "damp-heat body" and "cold-intolerant body", disease labels such as "liver fire hyperactivity" and "heavy dampness", and health condition description labels such as "good body quality" and "health". As an example, the executing subject may first match the first image feature extracted in step 202 and the second image feature extracted in step 203 with image features in a preset tongue image database. Then, based on the description label corresponding to the matched feature, the executing body may generate the health status information in accordance with the corresponding description label.

In some optional implementations of this embodiment, the tongue image database may include a conventional image feature comparison sub-library and a network extracted feature comparison sub-library. The conventional image feature comparison sub-library may include a correspondence between the conventional image features and the description tags indicating the health status. The network extracted feature comparison sub-library may include a correspondence between the network extracted feature and a description tag indicating a health status. The above-described conventional image features and network extracted features may be in accordance with the forms of the aforementioned first image features and second image features, respectively. Based on this, the executing body can generate the health status information prompted by the tongue image to be recognized through the following steps:

the first step is to compare the first image characteristic with the traditional image characteristic comparison sub-library to generate a first matching result vector.

In these implementations, the executing entity may compare the first image feature extracted in step 202 with the conventional image feature comparison sub-library in various ways to generate a first matching result vector. Wherein, the elements in the first matching result vector may be used to characterize the similarity between the first image feature and the features included in the conventional image feature comparison sub-library.

As an example, the image feature may be in the form of a feature vector, and the execution subject may determine a cosine similarity between the extracted first image feature and a conventional image feature in the conventional image feature comparison sub-library. Then, the execution body may combine the determined cosine similarities into a first matching result vector. Wherein the first element in the first matching result vector is used to indicate the cosine similarity between the first image feature extracted in step 202 and the first conventional image feature compared in the conventional image feature comparison sub-library, and so on.

And secondly, comparing the second image features with the network extraction feature comparison sub-library to generate a second matching result vector.

In these implementations, the executing entity may compare the second image feature extracted in step 203 with the network-extracted feature comparison sub-library in various ways to generate a second matching result vector. Wherein, the elements in the second matching result vector may be used to characterize the similarity between the second image feature and the features included in the network extracted feature contrast sub-library. The above processes of performing the feature comparison and generating the second matching result vector may be similar to the above description of the first step, and are not repeated here.

And thirdly, selecting a first target image from the tongue image database based on the combination between the generated first matching result vector and the second matching result vector.

In these implementations, the executing agent may select the first target image from the tongue image database in various ways based on a combination between the generated first matching result vector and the second matching result vector.

Optionally, the executing body may select the first target image from the tongue image database by:

and S1, carrying out weighted average on the first matching result vector and the second matching result vector to generate a target matching result vector.

In these implementations, the executing entity may perform a weighted average on the first matching result vector generated in the first step and the second matching result vector generated in the second step to generate a target matching result vector. Specifically, the executing entity may multiply the first matching result vector and the second matching result vector by a preset weight, respectively, and then sum the multiplied vectors to generate the target matching result vector. Wherein the sum of the above multiplied preset weights is usually 1.

And S2, selecting an image corresponding to the element with the maximum similarity degree in the target matching result vector from the tongue image database as a first target image.

In these implementations, the executing agent may select, from the tongue image database, an image corresponding to an element of the target matching result vector generated in step S1, which represents the greatest degree of similarity, as the first target image. As an example, the dimension of the above target matching result vector may be 10,000. Wherein, the similarity degree indicated by the 128 th element (for example, cosine similarity is 0.89) in the target matching result vector is the maximum. The executing body may select an image corresponding to the 128 th element from the tongue image database as a first target image.

Based on the above optional implementation, the matching result of the extracted traditional image features and the matching result of the image features extracted by the network may be fused, so that the matched image is determined more comprehensively.

Optionally, the executing body may further select, from the tongue image database, an image corresponding to an element, of the generated first matching result vector and the second matching result vector, whose similarity degree is greater than a preset threshold, as the at least one first target image. As an example, the preset threshold may be, for example, 0.7. The executing agent may select an image corresponding to an element greater than 0.7 of the first matching result vector and the second matching result vector from the tongue image database as the first target image. As yet another example, the above preset thresholds may be set to 0.65 and 0.72, respectively. The executing agent may select an image corresponding to an element greater than 0.65 in the first matching result vector and an image corresponding to an element greater than 0.72 in the second matching result vector from the tongue image database as the first target image.

And fourthly, generating health state information according to the description label corresponding to the first target image.

In these implementations, the executing entity may generate the health status information corresponding to the description label in various ways according to the description label corresponding to the first target image selected in the third step. As an example, when the number of the first target images is 1, the executing entity may generate health status information (e.g., "health", "sub-health", etc.) matching the description tag corresponding to the first target image.

Optionally, based on the selected at least one first target image, the execution subject may generate the health status information in various ways according to the fusion of the description tags corresponding to the at least one first target image. As an example, the executing entity may generate the health status information according to whether the description tag corresponding to the at least one first target image is consistent. If the description tags are consistent, the execution subject may generate health status information according to the health status indicated by the corresponding description tag. If the description labels are not consistent, the executing body may generate health status information consistent with the description label corresponding to the image indicated by the element with the highest similarity degree in the first matching result vector and the second matching result vector. As another example, the executing entity may generate the health status information according to the number of health statuses indicated by the description labels corresponding to the at least one first target image. For example, when the health status indicated by the description label corresponding to the at least one first target image is 3 "healthy" and 1 "sub-healthy", the executing subject may generate health status information representing health.

Based on the optional implementation manner, optionally, according to the fusion of the description tags corresponding to the at least one first target image, the executing body may further generate the health status information by:

and S1, acquiring preset traditional image feature matching weights and network extraction feature matching weights in response to the fact that the health state indicated by the description label corresponding to the at least one first target image is inconsistent.

In these implementations, the preset traditional image feature matching weight and the network extracted feature matching weight may be used to adjust the weights of the traditional image feature and the network extracted feature. For example, when the quality of the conventional image features is considered to be high, the preset conventional image feature matching weight can be properly adjusted high; when the quality of the network used for feature extraction is considered to be high, the network extraction feature matching weight can be appropriately increased.

And S2, multiplying the element corresponding to at least one first target image by the acquired preset traditional image feature matching weight or network extraction feature matching weight to obtain an adjusted result.

In these implementations, as an example, the elements corresponding to the at least one first target image may include elements (e.g., 0.67, 0.75) selected from the first matching result vector and larger than a preset threshold (e.g., 0.6) and elements (e.g., 0.82, 0.78) selected from the second matching result vector and larger than a preset threshold (e.g., 0.7). Assuming that the acquired preset conventional image feature matching weight and the network extracted feature matching weight may be 0.9 and 0.8, respectively, the adjusted result may be 0.67 × 0.9 — 0.603 for element 0.67. For element 0.78, the adjusted result may be 0.78 × 0.8 — 0.624.

S3, determining an image corresponding to the adjusted result representing the maximum degree of similarity among the obtained adjusted results as a first reference image.

And S4, generating the health state information consistent with the description label corresponding to the first reference image.

Based on the optional implementation mode, the matching flexibility is improved by introducing the matching weights corresponding to different features.

and S1, in response to the fact that the health states indicated by the description labels corresponding to the at least one first target image are inconsistent, selecting matched additional questions from a preset additional question library.

In these implementations, the additional problem repository typically holds the problems associated with the descriptive tags. As an example, in response to determining that the at least one first target image corresponds to a descriptive label comprising "yang deficiency", "B", "C", "D", "A", "D", "A", "D", "A", "D", "B", "D", "B", "D", "B", "Yin deficiency body Quality of food", the execution subject can select the" yang deficiency constitution "and" from the preset additional question bank "Constitutions of yin deficiency"related questions as an additional question of matching.

S2, sending the matched additional question to the target device.

In these implementations, the target device may generally be a device that transmits the tongue image to be recognized.

And S3, receiving answer information corresponding to the matched additional questions.

And S4, generating the health status information according to the matching degree of the description label corresponding to the at least one first target image and the received at least one first target image.

Based on the optional implementation mode, the scheme can acquire related additional information by integrating the image recognition result and the interaction with the user side, and improves the accuracy of tongue picture recognition.

In some optional implementations of the present embodiment, the tongue image database may include a correspondence between the total image features and description tags for indicating health status. Based on this, the executing body can generate the health status information prompted by the tongue image to be recognized through the following steps:

and firstly, fusing the first image characteristic and the second image characteristic to generate a target image characteristic.

In these implementations, the executing entity may fuse the first image feature extracted in step 202 and the second image feature extracted in step 203 in various ways to generate the target image feature. The feature fusion can be performed in various ways. For example, for features with the same dimension, a weighted average may be used. For another example, a way of recompressing the dimensions after splicing may be employed. And will not be described in detail herein.

And secondly, comparing the target image features with the total image features in the tongue image database to generate a third matching result vector.

In these implementations, the execution principal may generate a third matching result vector in a manner similar to the aforementioned comparison of the first image feature against the conventional image feature against the sub-library.

And thirdly, selecting an image corresponding to the element with the maximum similarity degree with the representation in the third matching result vector from the tongue image database as a second target image.

And fourthly, generating health state information according to the description label corresponding to the second target image.

In these implementations, the above methods for selecting the second target image and generating the health status information may be consistent with the descriptions of the corresponding parts, and are not described herein again.

Based on the optional implementation mode, the images can be matched according to the fused traditional image features and the image features extracted by the network, and the health state information is generated based on the matched images, so that the accuracy of tongue picture identification is improved.

In some optional implementations of this embodiment, the executing body may further continue to perform the following steps:

step one, responding to the situation that the generated health state information prompts unhealthy, and selecting matched questions from a preset question bank according to the health state information.

In these implementations, in response to determining that the generated health status information suggests unhealthy, the executive selects a matching question from a preset question bank based on the health status information. As an example, the health status information may be "weak health". The execution subject may select "recently feel chilly? "," is feeble or not recently? "as a matter of match.

And secondly, sending the matching problem to the target equipment.

In these implementations, the executing agent may send the matching question selected in the first step to the target device. The target device may be a device that transmits the tongue image to be recognized.

And thirdly, receiving answer information corresponding to the matched questions.

In these implementations, the execution principal may receive answer information corresponding to the matched question sent by the target device. As an example, the above answer information may include "yes" or "no". Optionally, the answer information may further include content not included in the matched question. For example, the answer information may be "yes, and it is very easy to cool hands and feet recently".

And fourthly, selecting corresponding disease information from a preset disease information base as disease prompt information according to the answer information and the health state information.

In these implementations, according to the answer information received in the third step and the generated health status information, the execution subject may select corresponding disease information from a preset disease information base as disease indication information in various ways. The disease information base may include a corresponding relationship between disease description information and disease information. For example, the disease information base may include "headache, dry throat, rhinorrhea — typhoid fever".

And fifthly, sending disease prompt information to the target equipment.

Based on the optional implementation manner, the scheme can acquire the related additional information through interaction with the user side when the tongue picture indicates the unhealthy state, and can generate the disease prompting information for prompting the disease according to the acquired additional information and the health state information determined based on the image characteristics, so that the accuracy of tongue picture identification is further improved.

With continued reference to fig. 3, fig. 3 is a schematic diagram of an application scenario of a method for recognizing a tongue picture according to an embodiment of the present disclosure. In the application scenario of fig. 3, a user 301 uses a terminal device 302 to capture a tongue image 303. Thereafter, the user 301 clicks the "upload" button and the terminal device 302 sends the tongue image 303 to the backend server 304. The backend server 304 extracts the first image feature 305 and the second image feature 306 of the tongue image 303. Based on the matching of the extracted first image features 305 and second image features 306 with the preset tongue image database 307, the backend server 304 may generate health status information 308 (e.g., "health") prompted by the tongue image 303 described above. Optionally, the backend server 304 may further send the generated health status information 308 to the terminal device 302.

At present, one of the prior arts usually identifies tongue images only according to manual design features or existing image classification methods, resulting in inaccurate matching results. In the method provided by the embodiment of the disclosure, the artificial design features of the image and the features extracted by the neural network are fused to serve as the basis for image matching, so that the advantages of the local features and the global features are fully utilized, and the matching degree of the generated health state information and the tongue image to be recognized is improved.

With further reference to FIG. 4, a flow 400 of one embodiment of a method for identifying a tongue is shown. The process 400 of the method for identifying a tongue comprises the steps of:

step 401, a tongue image to be recognized is acquired.

In the present embodiment, the execution subject (e.g., the

terminal device

101, 102, 103 in fig. 1) of the method for recognizing a tongue image may acquire a tongue image to be recognized in various ways. The tongue image to be recognized may be an original image captured by a user with respect to a tongue, or may be an image of a protruding tongue body obtained by processing the original image.

in a first step, in response to detecting a tongue image capture operation, lighting condition information of an environment for capturing a tongue image is acquired.

In these implementations, in response to detecting the tongue image capture operation, the executing body described above may acquire lighting condition information of the environment for capturing the tongue image in various ways. As an example, the tongue image capture operation may be, for example, the user clicking a "take tongue picture" button. The execution body may acquire the lighting condition information from a light sensor installed therein.

And secondly, determining whether the ambient light indicated by the illumination condition information meets a preset tongue image acquisition condition.

In these implementations, the preset tongue image capture conditions may include, for example, that the illumination brightness belongs to a first preset interval, the illumination color temperature belongs to a second preset interval, and so on.

And thirdly, responding to the determination that the tongue image is not met, and displaying information for prompting acquisition of the tongue image in a proper lighting environment.

In these implementations, the information prompting acquisition of the tongue image in a suitable lighting environment may be, for example, "please acquire the tongue image in a suitable lighting condition".

And fourthly, in response to the determination of satisfaction, acquiring the tongue image as the tongue image to be recognized.

Based on the optional implementation mode, the quality of the acquired tongue image to be recognized can be improved from the image acquisition stage through light condition detection, and a basis is provided for accurate recognition of the tongue image.

And step 402, sending the tongue image to be identified to a target server.

In this embodiment, the executing body may send the tongue image to be recognized acquired in step 401 to the target server. The target server may include various servers capable of implementing tongue image recognition. Optionally, the target server may further include an execution body for executing the method for recognizing tongue manifestation described in the foregoing embodiment shown in fig. 2.

And step 403, receiving the health status information sent by the target server.

In this embodiment, the execution subject may receive the health status information sent by the target server. The health status information may be consistent with the corresponding descriptions in the foregoing embodiments, and is not described herein again.

In some optional implementations of the present embodiment, the health status information may be generated based on the method for recognizing tongue manifestation as described in the foregoing embodiment shown in fig. 2.

And step 404, displaying the tongue picture recognition result based on the health state information.

In this embodiment, based on the health status information received in step 403, the execution main body may display the tongue recognition result in various ways. As an example, the executing entity may directly display the health status information received in step 403 as the tongue recognition result. As another example, the executing entity may further send the health status information received in step 403 to a manual review terminal, and in response to receiving the confirmation information sent by the manual review terminal, the executing entity may display the health status information as the tongue recognition result.

In some optional implementations of this embodiment, based on the health status information, the executing body may display the tongue picture recognition result according to the following steps:

in the first step, a matching problem acquisition request is sent to a target server in response to the fact that the received health status information prompts unhealthy.

In these implementations, the execution principal may send a matching problem acquisition request to the target server in response to determining that the received health status information suggests unhealthy. Wherein the matching problem obtaining request is used for obtaining the problem associated with the received health status. Optionally, the above-mentioned question associated with the received health status may be the same as the question selected from the preset question library and matched as described in the foregoing embodiments, and will not be described herein again.

And secondly, responding to the received matching problem corresponding to the matching problem acquisition request, and displaying the matching problem.

And thirdly, receiving answer information which is input by the user and corresponds to the matched questions.

In these implementations, the execution body may receive answer information corresponding to the matching question, which is input by the user. The answer information may include various forms. For example, the answer information may include information that characterizes the acceptance of the question generated by the user clicking a "yes" button or the camera capturing the user's nodding behavior. For another example, the answer information may further include a voice answered by the user.

And fourthly, sending the answer information to the target server.

And fifthly, receiving disease prompt information sent by the target server.

In these implementations, the condition hint information may be determined based on the answer information and the health status information. As an example, the above-mentioned condition prompt information is generally consistent with at least one of the above-mentioned answer information and the above-mentioned health status information. Optionally, the above-mentioned disease indication information may also be consistent with the corresponding disease information selected from the preset disease information base described in the foregoing embodiment, and is not described herein again.

And sixthly, displaying disease prompt information.

Based on the optional implementation manner, the scheme can acquire the related additional information through interaction with the user when the tongue picture indicates the unhealthy state, and can generate the disease prompting information for prompting the disease according to the acquired additional information and the health state information determined based on the image characteristics, so that the accuracy of tongue picture identification is further improved.

As can be seen from fig. 4, the flow 400 of the method for recognizing tongue image in the present embodiment represents the steps of acquiring the tongue image to be recognized and displaying the tongue image recognition result based on the received health status information. Therefore, the scheme described in the embodiment can perform corresponding processing from the aspects of image acquisition, result display and the like so as to improve the accuracy of tongue picture identification.

With further reference to fig. 5, as an implementation of the method shown in the above figures, the present disclosure provides an embodiment of an apparatus for recognizing tongue images, which corresponds to the embodiment of the method shown in fig. 2, and which is particularly applicable in various electronic devices.

As shown in fig. 5, the apparatus 500 for recognizing tongue image provided by the present embodiment includes a first acquiring unit 501, a first extracting unit 502, a second extracting unit 503, and a generating unit 504. The first acquiring unit 501 is configured to acquire a tongue image to be recognized; a first extraction unit 502 configured to extract a first image feature of a tongue image to be recognized, wherein the first image feature belongs to a conventional image feature; a second extraction unit 503 configured to extract a second image feature of the tongue image to be recognized by using a pre-trained neural network; the generating unit 504 is configured to generate health status information prompted by the tongue image to be recognized based on matching of the first image feature and the second image feature with a preset tongue image database, where the tongue image database includes a correspondence between the image features and description tags indicating health statuses.

In the present embodiment, in the apparatus 500 for recognizing tongue image: the detailed processing of the first obtaining unit 501, the first extracting unit 502, the second extracting unit 503 and the generating unit 504 and the technical effects brought by the processing can refer to the related descriptions of step 201, step 202, step 203 and step 204 in the corresponding embodiment of fig. 2, which are not repeated herein.

In some optional implementations of this embodiment, the tongue image database may include a conventional image feature comparison sub-library and a network extracted feature comparison sub-library. The generating unit 504 may include a first comparing subunit (not shown), a second comparing subunit (not shown), a first selecting subunit (not shown), and a first generating subunit (not shown). Wherein the first comparison subunit may be configured to: and comparing the first image characteristic with the conventional image characteristic comparison sub-library to generate a first matching result vector. Wherein, the elements in the first matching result vector can be used to characterize the similarity between the first image feature and the features included in the conventional image feature contrastive sub-library. The second ratio subunit may be configured to: and comparing the second image features with the network extracted feature comparison sub-library to generate a second matching result vector. Wherein, the elements in the second matching result vector can be used to characterize the similarity between the second image feature and the features included in the network extracted feature contrast sub-library. The first selecting subunit may be configured to: based on the generated combination between the first matching result vector and the second matching result vector, a first target image is selected from the tongue image database. The first generating subunit may be configured to: and generating health state information according to the description label corresponding to the first target image.

In some optional implementations of this embodiment, the first selecting subunit may include a generating module (not shown in the figure) and a selecting module (not shown in the figure). The generating module may be configured to perform weighted average on the first matching result vector and the second matching result vector to generate a target matching result vector. The selecting module may be configured to select, from the tongue image database, an image corresponding to an element of the target matching result vector representing the greatest degree of similarity as the first target image.

In some optional implementations of this embodiment, the first selecting subunit may be further configured to: and selecting an image corresponding to an element with similarity degree larger than a preset threshold value in the generated first matching result vector and the second matching result vector from the tongue image database as at least one first target image. The first generating subunit may be further configured to: and generating health state information according to the fusion of the description labels corresponding to the at least one first target image.

In some optional implementations of this embodiment, the first generating subunit may be further configured to: in response to the fact that the health states indicated by the description labels corresponding to the at least one first target image are inconsistent, acquiring preset traditional image feature matching weights and network extraction feature matching weights; multiplying elements corresponding to at least one first target image by the acquired preset traditional image feature matching weight or network extraction feature matching weight to obtain an adjusted result; determining an image corresponding to an adjusted result which represents the maximum degree of similarity in the obtained adjusted result as a first reference image; health status information is generated consistent with the descriptive label corresponding to the first reference image. In some optional implementations of this embodiment, the first generating subunit may be further configured to: in response to the fact that the health states indicated by the description labels corresponding to the at least one first target image are inconsistent, selecting matched additional questions from a preset additional question library; sending the matched additional question to the target device; receiving answer information corresponding to the matched additional questions; and generating health status information according to the matching degree of the description label corresponding to the at least one first target image and the received at least one first target image.

In some optional implementations of the present embodiment, the tongue image database may include a correspondence between the total image features and description tags for indicating health status. The generating unit 504 may include a fusion subunit (not shown), a third ratio subunit (not shown), a second selection subunit (not shown), and a second generating subunit (not shown). Wherein the above-mentioned fusion subunit may be configured to: and fusing the first image characteristic and the second image characteristic to generate a target image characteristic. The third ratio subunit may be configured to: and comparing the target image features with the total image features in the tongue image database to generate a third matching result vector. The second selecting subunit may be configured to: and selecting an image corresponding to the element with the maximum similarity degree of the representation in the third matching result vector from the tongue image database as a second target image. The second generation subunit may be configured to: and generating health state information according to the description label corresponding to the second target image.

In some optional implementations of the present embodiment, the obtaining unit 501 may include a obtaining sub-unit (not shown in the figure) and a preprocessing sub-unit (not shown in the figure). Wherein the acquiring subunit may be configured to acquire the acquired initial tongue image. The preprocessing subunit may be configured to preprocess the initial tongue image to generate a tongue image to be recognized. Wherein the pre-treatment may comprise at least one of: tongue color correction, tongue segmentation.

In some optional implementations of the present embodiment, the apparatus 500 for recognizing a tongue image may further include: a first selecting unit (not shown), a first sending unit (not shown), a first receiving unit (not shown), a second selecting unit (not shown), and a second sending unit (not shown). Wherein the first selecting unit may be further configured to: and responding to the judgment that the generated health state information prompts unhealthy, and selecting matched questions from a preset question bank according to the health state information. The first transmitting unit may be further configured to: and sending the matched problem to the target device. The first receiving unit may be further configured to: answer information corresponding to the matched question is received. The second selecting unit may be further configured to: and selecting corresponding disease information from a preset disease information base as disease prompt information according to the answer information and the health state information. The second transmitting unit may be further configured to: and transmitting disease condition prompt information to the target equipment.

In the apparatus provided by the above embodiment of the present disclosure, the generation unit 504 fuses the artificially designed features extracted by the first extraction unit 502 and the features extracted by the neural network extracted by the second extraction unit 503 to serve as a basis for image matching, so that advantages of local features and global features are fully utilized, and a matching degree between the generated health state information and the tongue image to be recognized is improved.

With further reference to fig. 6, as an implementation of the method shown in the above figures, the present disclosure provides an embodiment of an apparatus for recognizing tongue images, which corresponds to the embodiment of the method shown in fig. 4, and which is particularly applicable in various electronic devices.

As shown in fig. 6, the apparatus 600 for recognizing tongue image provided by the present embodiment includes a second acquiring unit 601, a third transmitting unit 602, a second receiving unit 603, and a display unit 604. Wherein, the second acquiring unit 601 is configured to acquire a tongue image to be recognized; a third sending unit 602 configured to send the tongue image to be recognized to the target server; a second receiving unit 603 configured to receive the health status information sent by the target server; a display unit 604 configured to display the tongue recognition result based on the health status information.

In the present embodiment, in the apparatus 600 for recognizing tongue image: the specific processing of the second obtaining unit 601, the third sending unit 602, the second receiving unit 603, and the display unit 604 and the technical effects thereof can refer to the related descriptions of step 401, step 402, step 403, and step 404 in the corresponding embodiment of fig. 4, which are not described herein again.

In some optional implementations of the present embodiment, the second obtaining unit 601 may include an obtaining subunit (not shown in the figure), a determining subunit (not shown in the figure), a first displaying subunit (not shown in the figure), and a collecting subunit (not shown in the figure). Wherein the obtaining subunit may be configured to: in response to detecting the tongue image capture operation, lighting condition information of an environment for capturing a tongue image is acquired. The determining subunit may be configured to: and determining whether the ambient light indicated by the illumination condition information meets a preset tongue image acquisition condition. The first display subunit may be configured to: in response to determining that the tongue image is not satisfied, information is displayed prompting acquisition of the tongue image in a suitable lighting environment. The above-mentioned acquisition subunit may be configured to: in response to determining that the condition is met, a tongue image is captured as a tongue image to be recognized.

In some optional implementations of the present embodiment, the display unit 604 may include a first sending subunit (not shown in the figure), a second displaying subunit (not shown in the figure), a first receiving subunit (not shown in the figure), a second sending subunit (not shown in the figure), a second receiving subunit (not shown in the figure), and a third displaying subunit (not shown in the figure). Wherein the first transmitting subunit may be configured to: and responding to the received health state information prompt which is determined to be unhealthy, and sending a matching question acquisition request to the target server. The second display subunit may be configured to: and responding to the received matching question corresponding to the matching question acquisition request, and displaying the matching question. The first receiving subunit may be configured to: answer information corresponding to the matching question input by the user is received. The second transmitting subunit may be configured to: and sending the answer information to the target server. The second receiving subunit may be configured to: and receiving disease prompt information sent by the target server. Wherein, the disease prompt information can be determined according to the answer information and the health status information. The third display subunit may be configured to: and displaying the disease prompt information.

In some optional implementations of this embodiment, the health status information may be generated based on the method described in the foregoing embodiment of fig. 2.

The device provided by the above embodiment of the present disclosure performs corresponding processing from the aspects of image acquisition, result display, and the like through the second acquisition unit 601 and the display unit 604, so as to improve the accuracy of tongue recognition.

Referring now to FIG. 7, a block diagram of an electronic device (e.g., the server of FIG. 1) 700 suitable for use in implementing embodiments of the present disclosure is shown. The terminal device in the embodiments of the present disclosure may include, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, and the like, and a stationary terminal such as a digital TV, a desktop computer, and the like. The server shown in fig. 7 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.

As shown in fig. 7, electronic device 700 may include a processing means (e.g., central processing unit, graphics processor, etc.) 701 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)702 or a program loaded from storage 708 into a Random Access Memory (RAM) 703. In the RAM 703, various programs and data necessary for the operation of the electronic apparatus 700 are also stored. The processing device 701, the ROM 702, and the RAM 703 are connected to each other by a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.

Generally, the following devices may be connected to the I/O interface 705: input devices 706 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; an output device 707 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 708 including, for example, magnetic tape, hard disk, etc.; and a communication device 709. The communication means 709 may allow the electronic device 700 to communicate wirelessly or by wire with other devices to exchange data. While fig. 7 illustrates an electronic device 700 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided. Each block shown in fig. 7 may represent one device or may represent multiple devices as desired.

In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such embodiments, the computer program may be downloaded and installed from a network via the communication means 709, or may be installed from the storage means 708, or may be installed from the ROM 702. The computer program, when executed by the processing device 701, performs the above-described functions defined in the methods of embodiments of the present disclosure.

It should be noted that the computer readable medium described in the embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In embodiments of the disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In embodiments of the present disclosure, however, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the foregoing.

The computer-readable medium may be one contained in the electronic device (server or terminal device); or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs which, when executed by the electronic device, cause the server to: acquiring a tongue image to be identified; extracting first image features of a tongue image to be recognized, wherein the first image features belong to traditional image features; extracting a second image characteristic of the tongue image to be recognized by utilizing a pre-trained neural network; generating health state information prompted by a tongue image to be recognized based on matching of the first image feature, the second image feature and a preset tongue image database, wherein the tongue image database comprises a corresponding relation between the image features and description labels used for indicating health states; or cause the terminal device to: acquiring a tongue image to be identified; the tongue image to be identified is sent to a target server; receiving health state information sent by a target server; and displaying the tongue picture recognition result based on the health state information.

Computer program code for carrying out operations for embodiments of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units described in the embodiments of the present disclosure may be implemented by software or hardware. The described units may also be provided in a processor, and may be described as: a processor comprises a first acquisition unit, a first extraction unit, a second extraction unit and a generation unit; or the processor comprises a second acquisition unit, a third sending unit, a second receiving unit and a display unit. Here, the names of the units do not constitute a limitation to the unit itself in some cases, and for example, the first acquiring unit may also be described as a "unit that acquires a tongue image to be recognized".

The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention in the embodiments of the present disclosure is not limited to the specific combination of the above-mentioned features, but also encompasses other embodiments in which any combination of the above-mentioned features or their equivalents is made without departing from the inventive concept as defined above. For example, the above features and (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure are mutually replaced to form the technical solution.

Claims

1. A method for identifying a tongue comprising:

acquiring a tongue image to be identified;

extracting first image features of the tongue image to be recognized, wherein the first image features belong to traditional image features;

extracting a second image characteristic of the tongue image to be recognized by utilizing a pre-trained neural network;

and generating health state information prompted by the tongue image to be recognized based on the matching of the first image feature, the second image feature and a preset tongue image database, wherein the tongue image database comprises a corresponding relation between the image features and description labels used for indicating the health state.

2. The method of claim 1, wherein the tongue image database comprises a traditional image feature contrast sub-library and a network extracted feature contrast sub-library; and

the generating health state information prompted by the to-be-recognized tongue image based on the matching of the first image feature, the second image feature and a preset tongue image database comprises:

comparing the first image feature with the conventional image feature comparison sub-library to generate a first matching result vector, wherein elements in the first matching result vector are used for representing the similarity between the first image feature and features included in the conventional image feature comparison sub-library;

comparing the second image features with the network extracted feature comparison sub-library to generate a second matching result vector, wherein elements in the second matching result vector are used for representing the similarity between the second image features and the features in the network extracted feature comparison sub-library;

selecting a first target image from the tongue image database based on a combination between the generated first matching result vector and the second matching result vector;

and generating the health state information according to the description label corresponding to the first target image.

3. The method of claim 2, wherein said extracting a first target image from the tongue image database based on a combination between the generated first and second match result vectors comprises:

carrying out weighted average on the first matching result vector and the second matching result vector to generate a target matching result vector;

and selecting an image corresponding to an element with the maximum similarity degree in the target matching result vector from the tongue image database as the first target image.

4. The method of claim 2, wherein said extracting a first target image from the tongue image database based on a combination between the generated first and second match result vectors comprises:

selecting an image corresponding to an element with similarity degree larger than a preset threshold value in the generated first matching result vector and second matching result vector from the tongue image database as at least one first target image; and

and generating the health state information according to the fusion of the description labels corresponding to the at least one first target image.

5. The method according to claim 4, wherein the generating the health status information according to the fusion of the description labels corresponding to the at least one first target image comprises:

in response to the fact that the health states indicated by the description labels corresponding to the at least one first target image are inconsistent, acquiring preset traditional image feature matching weights and network extraction feature matching weights;

multiplying the element corresponding to the at least one first target image by the acquired preset traditional image feature matching weight or the network extraction feature matching weight to obtain an adjusted result;

determining an image corresponding to an adjusted result which represents the maximum degree of similarity in the obtained adjusted result as a first reference image;

generating health status information consistent with the description label corresponding to the first reference image.

6. The method according to claim 4, wherein the generating the health status information according to the fusion of the description labels corresponding to the at least one first target image comprises:

in response to determining that the health states indicated by the description labels corresponding to the at least one first target image are inconsistent, selecting matched additional questions from a preset additional question library;

sending the matched additional question to a target device;

receiving answer information corresponding to the matched additional questions;

and generating the health status information according to the matching degree corresponding to the received at least one first target image in the description label corresponding to the at least one first target image.

7. The method according to claim 1, wherein the tongue image database includes a correspondence between total image features and descriptive labels indicating health status; and

fusing the first image characteristic and the second image characteristic to generate a target image characteristic;

comparing the target image features with the total image features in the tongue image database to generate a third matching result vector;

selecting an image corresponding to an element with the maximum degree of similarity of the representation in the third matching result vector from the tongue image database as a second target image;

and generating the health state information according to the description label corresponding to the second target image.

8. The method of claim 1, wherein the acquiring a tongue image to be recognized comprises:

acquiring an acquired initial tongue image;

preprocessing the initial tongue image to generate the tongue image to be recognized, wherein the preprocessing comprises at least one of the following steps: tongue color correction, tongue segmentation.

9. The method according to one of claims 1 to 8, wherein the method further comprises:

responding to the generated health state information prompt which is determined to be unhealthy, and selecting matched questions from a preset question bank according to the health state information;

sending the matched question to a target device;

receiving answer information corresponding to the matched question;

selecting corresponding disease information from a preset disease information base as disease prompt information according to the answer information and the health state information;

and sending the disease prompt information to the target equipment.

10. A method for identifying a tongue comprising:

acquiring a tongue image to be identified;

sending the tongue image to be identified to a target server;

receiving health state information sent by the target server;

and displaying a tongue picture recognition result based on the health status information.

11. The method of claim 10, wherein the acquiring a tongue image to be recognized comprises:

acquiring illumination condition information of an environment for shooting a tongue image in response to detection of a tongue image acquisition operation;

determining whether the ambient light indicated by the illumination condition information meets a preset tongue image acquisition condition;

in response to determining that the tongue image is not satisfied, displaying information for prompting acquisition of the tongue image in a suitable lighting environment;

in response to determining satisfaction, capturing a tongue image as the tongue image to be recognized.

12. The method of claim 10, wherein said displaying a tongue recognition result based on said health status information comprises:

in response to determining that the received health status information prompts unhealthy, sending a matching problem acquisition request to the target server;

responding to the received matching question corresponding to the matching question acquisition request, and displaying the matching question;

receiving answer information corresponding to the matched questions and input by a user;

sending the answer information to the target server;

receiving disease condition prompt information sent by the target server, wherein the disease condition prompt information is determined according to the answer information and the health state information;

and displaying the disease prompt information.

13. The method according to one of claims 10 to 12, wherein the health status information is generated based on the method according to one of claims 1 to 9.

14. An apparatus for recognizing tongue picture, comprising:

a first acquisition unit configured to acquire a tongue image to be recognized;

a first extraction unit configured to extract a first image feature of the tongue image to be recognized, wherein the first image feature belongs to a traditional image feature;

a second extraction unit configured to extract a second image feature of the tongue image to be recognized by using a pre-trained neural network;

the generating unit is configured to generate health status information prompted by the tongue image to be recognized based on matching of the first image feature and the second image feature with a preset tongue image database, wherein the tongue image database comprises a corresponding relation between image features and description labels for indicating health statuses.

15. An apparatus for recognizing tongue picture, comprising:

a second acquisition unit configured to acquire a tongue image to be recognized;

a third sending unit configured to send the tongue image to be identified to a target server;

the second receiving unit is configured to receive the health state information sent by the target server;

a display unit configured to display a tongue picture recognition result based on the health status information.

16. An electronic device, comprising:

one or more processors;

a storage device having one or more programs stored thereon;

when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-13.

17. A computer-readable medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1-13.