CN106777882A - A kind of diet reminding method and device - Google Patents
A kind of diet reminding method and device Download PDFInfo
- Publication number
- CN106777882A CN106777882A CN201611035685.9A CN201611035685A CN106777882A CN 106777882 A CN106777882 A CN 106777882A CN 201611035685 A CN201611035685 A CN 201611035685A CN 106777882 A CN106777882 A CN 106777882A
- Authority
- CN
- China
- Prior art keywords
- data
- factor
- influence
- food image
- food
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G06F19/3475—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
Abstract
The present invention provides a kind of diet reminding method and device, wherein the diet reminding method includes:Obtain food image and health data;The food image is identified to obtain modified data;Prompt message is provided a user with according to the modified data and the health data.
Description
Technical field
The present invention relates to Image Information Processing field, and in particular to a kind of diet reminding method and device.
Background technology
According to ministry of Health of China data display, end Chinese chronic in 2015 and surpassed 2.6 hundred million, and blood for example high
Pressure, high fat of blood, diabetic also increasing year by year, therefore the health that is careful in one's diet for similar chronic, is conducive to
Alleviate disease development, while be careful in one's diet middle element and the content of each element, also outstanding for pregnant woman with the people for being devoted to losing weight
Its is important.
People rule of thumb can artificially judge whether diet meets the health status of oneself with knowledge, or be beneficial to
The health problem of oneself concern, but because dietetic variety is various and the limitation of experience knowledge, people are difficult objectively
Judge whether diet is reasonable.When problems are run into, people can also be sentenced by inquiring about existing health guidance data come objective
Disconnected this problem, but the sign of individual is often more complicated, and artificial inquiry mode is obviously less efficient.
The content of the invention
Therefore, the present invention is to solve it is existing determine diet whether be suitable to itself sign situation scheme efficiency it is low
Problem.
In view of this, the present invention provides a kind of diet reminding method, including:
Obtain food image and health data;
The food image is identified to obtain modified data;
Prompt message is provided a user with according to the modified data and the health data.
Preferably, prompt message is provided a user with according to the modified data and the health data, including:
According to the health data and the modified data, determine influence of the modified data to the health data because
Son;
Prompt message is provided a user with according to the factor of influence.
Preferably, it is described prompt message is provided a user with according to the factor of influence to include:
The factor of influence is compared with least one default factor of influence threshold value;
Prompt message is provided a user with according to comparison result.
Preferably, the default factor of influence threshold value is multiple, multiple default factor of influence threshold values can determine to
A few factor of influence scope, the factor of influence is fallen into when in the range of the different factors of influence, and the prompt message is not
It is identical.
Preferably, the health data includes:At least one illness data and at least one user's sign data;The basis
The health data and the modified data, determine factor of influence of the modified data to the health data, including:
The illness data and user's sign data are obtained respectively to the weight shared by the factor of influence of the modified data;
Factor of influence of the health data to the modified data according to the weight calculation.
Preferably, it is described that prompt message is provided a user with according to the modified data and the health data, including:
The health data that prestores corresponding with the health data is inquired about in database;
Inquire about in the database with prestore health data and the health factor of every kind of food, wherein the data
There is the health factor of multiple prestore health data and every kind of foods corresponding with the health data that prestored each Suo Shu in storehouse.
Preferably, the acquisition food image, including:
Single or multiple single frames food images are obtained in units of frame;
The multiple single frames food image is processed respectively, identifies the food image.
Preferably, the acquisition food image, including:
Multiple single frames food images are obtained in units of frame;
Split is carried out to the multiple single frames food image;
The food image after to split is processed, and identifies the food image.
Preferably, it is described the multiple single frames food image is processed respectively, the food image is identified, wrap
Include:
Sampling selection is carried out to the multiple single frames food image;
The single frames food image that sampling is obtained is identified respectively, to be identified result;
Each described recognition result is weighted average calculating operation;
The food image is determined according to the operation result.
Preferably, it is described the multiple single frames food image is processed respectively, the food image is identified, wrap
Include:
The multiple single frames food image is identified, to be identified result;
Each described recognition result is weighted average calculating operation;
The food image is determined according to the operation result.
Correspondingly, the present invention provides a kind of diet suggestion device, including:
Acquiring unit, for obtaining food image and health data;
Recognition unit, for being identified obtaining food name data and modified data to the food image;
Factor of influence determining unit, for according to the health data and the modified data, determining the modified data
To the factor of influence of the health data;
Tip element, for providing a user with prompt message according to the factor of influence.
The diet reminding method and device for providing according to embodiments of the present invention, by obtaining food image, and to food figure
As being identified to obtain food name data and modified data, it is possible thereby in determining image according to the image that user provides
Comprising which food or composition;The health data of interest by obtaining user, and pair food for determining or composition with it is strong
Health data are analyzed, it may be determined that food or composition indicated by image pay close attention to it influence of health data, Jin Ergen
User is accordingly pointed out according to analysis result so that user can according to the food indicated by the clear and definite image of prompt message whether
Health problem on itself concern has influenceed.This programme is given objectively to be given and carried using the data and image of user input
Show information, it is not necessary to which user inquires about great mass of data and artificially judges, with efficiency higher.
Brief description of the drawings
In order that present disclosure is more likely to be clearly understood, below according to specific embodiment of the invention and combine
Accompanying drawing, the present invention is further detailed explanation, wherein
Fig. 1 is the flow chart of the diet reminding method that the embodiment of the present invention 1 is provided;
Fig. 2 is the structural representation of the diet suggestion device that the embodiment of the present invention 2 is provided.
Specific embodiment
The embodiment of the present invention provides a kind of diet reminding method, as shown in figure 1, including:
S11, obtains food image and health data.Food image can by take pictures or video record form obtain,
Wherein take pictures or the food image that obtains of form of video record can be taken pictures or record in advance, or using taking the photograph
As being automatically obtained after the direct alignment target of head.Shoot video or carry out take pictures obtain food image shooting process in, eventually
The mode for only shooting is in addition to user is automatically stopped shooting, it is also possible to using equipment such as gyroscopes, by recognizing that gyroscope is believed
Cease to terminate to shoot.
Above-mentioned health data can include the illness data and user's sign data of user, wherein, illness data can be
The illness data of user itself, or user compares the illness data of concern, and user's sign packet is included:Whether other are suffered from
Chronic disease, in the recent period whether have any discomfort symptom with the sex of user, age, occupation etc. individual relevant information.
Health data can be interacted by voice or word and obtained, it is also possible to while being obtained using the mode of voice and word
, the selection of interactive mode can be to be input into by word or known by voice by carrying out selection in interactive mode list
Other mode provides input information.For the nonstandard data of user input, the embodiment of the present invention can carry out appropriate conversion
Treatment, for example, can be converted into standard input data and obtain according to the trained synonym model for obtaining by input information, its
Middle synonym model is obtained by the method for natural language processing, and by similar word unification for standard is input into, for example user is defeated
Enter diabetes and high blood glucose, can unify to be described for the language of standard.
S12, is identified to food image to obtain modified data, the data can include food title and food into
Point.Recognize that the mode of content of interest has various from image, the embodiment of the present invention preferably uses machine learning or depth
The Intelligent Recognition modes such as habit recognize food or composition from image.
Specifically, the identification to food image can be obtained by the good data classification model identification of training in advance, its
Middle data classification model is trained as training set by the use of the database comprising abundance of food image and correspondence modified data
Arrive, data classification model is classified to the food image of user input by deep learning model, may thereby determine that figure
Food or composition as included in.
In addition, in this step, food title and COF can also be shown for pointing out user simultaneously.
S13, prompt message is provided a user with according to the modified data and the health data.According to modified data and strong
Health data both data determine that the mode of a prompt message has various, for example can be corresponding with prompt message with pre-stored data
Relation, confirms prompt message by inquiring about by way of database.Certainly, the present invention is not limited only to this, the embodiment of the present invention
Additionally provide multiple preferred embodiment, will specifically introduce in greater detail below.
The content of prompt message can have various, and the presentation mode of prompt message can also have various.Can for example lead to
The form of voice message or ejection prompting frame is crossed, points out user to forbid the edible food image corresponding food or a small amount of edible
The corresponding food of food image or the corresponding food of the edible food image and recommend related other beneficial foods
Thing data etc..
The diet reminding method and device for providing according to embodiments of the present invention, by obtaining food image, and to food figure
As being identified to obtain modified data, it is possible thereby to during image is determined according to the image that user provides comprising which food or
Person's composition;The health data of interest by obtaining user, and pair determine food or composition be analyzed with health data,
Can determine that food or composition indicated by image pay close attention to it influence of health data, and then according to analysis result to user
Accordingly pointed out so that user can according to the food indicated by the clear and definite image of prompt message whether to itself pay close attention to health
Problem has influenceed.This programme is given using the data and image of user input and objectively provides prompt message, it is not necessary to used
Family is inquired about great mass of data and is artificially judged, with efficiency higher.
On above-mentioned steps S12, the machine learning scheme that the embodiment of the present invention is used is with existing based on deep learning
Food identifying schemes unlike, the identification of existing food uses standard depth learning framework, the kind of object in identification image.
Although last layer of grader of deep learning framework is softmax multi classifiers, it can not judge current input figure
Be as in multiple foods as input, or single diet is used as input.For example, when we shoot food A, depth
Practise in network returning result, as a result the probability of A is substantially far above other predicted values, at this moment it is considered that input identification image is A.
And work as A in returning result, when tri- probability of outcome values of B, C are all higher, it is impossible to determine it is have three input targets in present image,
Still only have a target but network to cannot be distinguished by it is A, B, which in C.
The method of conventional solution multiple target input classification is learnt e-learning to solve multiple target by multi-tag
Problem, but its accuracy rate cannot be guaranteed always, while also needing to preferable multi-tag mark.Another method is by many mesh
Mark is not done as multi-target detection (detection), can be by the position mark of different target out.It is right that do so needs
Target in training data carries out position mark, brings more workloads.This problem, mark are recognized simultaneously for food
If food is all placed in plate in note data, and real scene data are all placed in various irregular containers, use inspection
Surveying model may also bring along problem.
Based on above mentioned problem, we employ one kind by traceback depth neutral net, calculate current layer neuron to spy
Determine the mode of recognition result contribution margin to determine to recognize the Position Approximate of target in image, and be confirmed whether it is multiple target
Method.
If recognition result is A, the probability of B, C is all higher, is found in original image by recalling, to A, B, C result
Contribution highest pixel is all reunited and is concentrated in together, then show it is currently single object input, A, B, and only one of which is correctly tied in C
Really.If it is different positions to reunite, multiple target input is shown to be.So as to can disposably be obtained by follow-up post processing
Uncertain number target input recognition result.
Meanwhile, the identification model of the embodiment of the present invention is not limited to the deep neural network model of deep learning.May be used also
To use color, the shape of different colours, food overall profile, Texture eigenvalue extracts characteristics of image, used as grader point
Category feature.
As one preferred embodiment, above-mentioned steps S13 may include steps of:
S131a, according to health data and modified data, determines factor of influence of the modified data to health data.For example, logical
Crossing data classification model and identify the food title, and then obtain food includes:Tri- kinds of COF of A, B, C, according to user
Health data, tri- kinds of factors of influence of COF of A, B, C are determined respectively, weighted array obtains health data relative to food
Overall factor of influence, that is, food is integrally to the factor of influence of user health.
S132a, prompt message, i.e. this programme are provided a user with by modified data to the shadow of health data according to factor of influence
Sound has carried out quantification treatment, and relatively different prompt messages are can determine that according to the numeric ratio for quantifying.
Further, above-mentioned steps S132a, including:
S1321a, factor of influence is compared with least one default factor of influence threshold value;
S1322a, prompt message is provided a user with according to comparison result.
Preferably, default factor of influence threshold value therein can be multiple, and the default factor of influence threshold value of multiple can determine
At least one factor of influence scope, factor of influence is fallen into when in the range of the different factors of influence, and prompt message is differed.
Specifically, judge the magnitude relationship of the factor of influence and default factor of influence threshold value, draw factor of influence to
The influence degree of family health, dietary recommendation is further provided for by influence degree, pre-sets multiple default factor of influence threshold values
And the multiple different factor of influence scopes that multiple default factor of influence threshold values are divided, by judging factor of influence and presetting
The relation of factor of influence, provides a user with prompt message, and the wherein interval of factor of influence can be normalized to 0-1;For example work as bag
When including four default factors of influence, when factor of influence is more than default first factor of influence threshold value, less than default second factor of influence
During threshold value, user is pointed out to forbid the corresponding food of edible food image;When factor of influence is more than default second factor of influence threshold value,
During less than default 3rd factor of influence threshold value, user is pointed out to eat the corresponding food of food image on a small quantity;When factor of influence is more than
Default 3rd factor of influence threshold value, during less than default 4th factor of influence threshold value, points out user edible food image correspondence
Food and recommend related other beneficial modified datas.
The present invention provides two kinds of implementation methods of determination factor of influence.As the first implementation method, above-mentioned health data
At least one illness data and at least one user's sign data can be specifically included, wherein illness data can be user input
Title or the title of concern illness for having suffered from the disease etc.;Sign data can be the characteristics such as age, body weight, the occupation of user
According to.Further, above-mentioned steps S131a can include:
S1311a, obtains the illness data and user's sign data to shared by the factor of influence of the modified data respectively
Weight, namely various foods are to the weighing factor of various illnesss, and various foods are to the weight of various signs.For example, on
User health data are stated for X, X is according to multi-dimensional data (various illness data and various signs using algorithm set in advance
Data) numerical value that calculates, X can include illness x1, illness x2, age x3, body weight x4Deng.User health data X and food into
The various weights divided between k are θk1, θk2, θk3, θk4.
S1312a, factor of influence of the modified data to the health data according to the weight calculation.For example it is final
The factor of influence Y=Σ Y for calculatingk, Yk=θkX。
The weight of above-mentioned factor of influence can be trained by mass data and obtained.Such as Yk=θkIn X, for different people
Group's health data, each dimension of X may be not quite similar.Health data dimension can in advance be sieved by way of statistics
Choosing, determines fixed data dimension.For different user provide health data, by synonymicon by illness describe into
The unified normalized of row, the data dimension of missing is supplemented by way of asking in reply user or directly filling numerical value, filling
Numerical value can be the assembly average of 0 or current dimension.
It is trained by a large amount of training datas, so as to obtain the health of different health data X and different foods
Respective weights θ between factor Y.
Used as second implementation method of determination factor of influence, above-mentioned steps S131a can include:
S1311b, inquires about the health data that prestores corresponding with the health data in database;
S1312b, inquire about in the database with prestore health data and the health factor of every kind of food, wherein
There is the health of multiple prestore health data and every kind of foods corresponding with the health data that prestored each Suo Shu in the database
The factor.
Detection operation on image, the present invention provides numerous embodiments.As one of them preferred embodiment,
This method can obtain single frames or multiple single frames food images in the step of obtaining food image in units of frame, then distinguish
Multiple single frames food images are processed, food image is identified.
Alternatively, it is above-mentioned the multiple single frames food image is processed respectively, the food image is identified, specifically
Can be:Sampling selection is carried out to the multiple single frames food image;The single frames food image that sampling is obtained is entered respectively
Row identification, to be identified result;Each described recognition result is weighted average calculating operation;It is true according to the operation result
The fixed food image.
Alternatively, it is above-mentioned the multiple single frames food image is processed respectively, the food image is identified, specifically
Can be:The multiple single frames food image is identified, to be identified result;Each described recognition result is carried out
Weighted mean operation;The food image is determined according to the operation result.
Specifically, when the food image for obtaining is shot by video mode, video image can be divided into multiple single frames simultaneously
Be respectively processed with identification, can be by being weighted average calculating operation, according to computing to multiple recognition results for obtaining of identification
Result determines food image or in advance sampling selected part recognition result, and the recognition result that sampling is obtained is weighted averagely
Computing, food image is determined according to operation result.
Used as another preferred embodiment, this method can be obtained in the step of obtaining food image in units of frame
Take multiple single frames food images;Split is carried out to the multiple single frames food image;The food image after to split is carried out
Treatment, identifies the food image, after the video image that single frames is input into is carried out into split, the video figure after processing split
As identifying food image.
Processed by obtaining food image to above-mentioned video record, improve the various of food image acquisition pattern
Property.
An alternative embodiment of the invention is provided and additionally provides a kind of diet suggestion device, as shown in Fig. 2 including:Obtain
Unit 21, recognition unit 22, Tip element 23, wherein,
Acquiring unit 21 is used to obtain food image and health data;
Recognition unit 22 is used to that the food image to be identified to obtain food name data and modified data;
Tip element 23 is used to provide a user with prompt message according to the modified data and the health data.
Obviously, above-described embodiment is only intended to clearly illustrate example, and not to the restriction of implementation method.It is right
For those of ordinary skill in the art, can also make on the basis of the above description other multi-forms change or
Change.There is no need and unable to be exhaustive to all of implementation method.And the obvious change thus extended out or
Among changing still in the protection domain of the invention.
Claims (11)
1. a kind of diet reminding method, it is characterised in that including:
Obtain food image and health data;
The food image is identified to obtain modified data;
Prompt message is provided a user with according to the modified data and the health data.
2. method according to claim 1, it is characterised in that it is described according to the modified data and the health data to
User provides prompt message, including:
According to the health data and the modified data, factor of influence of the health data to the modified data is determined;
Prompt message is provided a user with according to the factor of influence.
3. method according to claim 2, it is characterised in that described that prompting letter is provided a user with according to the factor of influence
Breath includes:
The factor of influence is compared with least one default factor of influence threshold value;
Prompt message is provided a user with according to comparison result.
4. method according to claim 3, it is characterised in that the default factor of influence threshold value is multiple, it is multiple described
Default factor of influence threshold value can determine at least one factor of influence scope, the factor of influence fall into the different influences because
When in subrange, the prompt message is differed.
5. method according to claim 2, it is characterised in that the health data includes:At least one illness data and extremely
Few 1 user's sign data;It is described according to the health data and the modified data, determine the modified data to described strong
The factor of influence of health data, including:
The illness data and user's sign data are obtained respectively to the weight shared by the factor of influence of the modified data;
Factor of influence of the health data to the modified data according to the weight calculation.
6. method according to claim 1, it is characterised in that according to the modified data and the health data to user
Prompt message is provided, including:
The health data that prestores corresponding with the health data is inquired about in database;
Inquire about in the database with prestore health data and the health factor of every kind of food, wherein in the database
There is the health factor of multiple prestore health data and every kind of foods corresponding with the health data that prestored each Suo Shu.
7. method according to claim 1, it is characterised in that the acquisition food image, including:
Multiple single frames food images are obtained in units of frame;
The multiple single frames food image is processed respectively, identifies the food image.
8. method according to claim 1, it is characterised in that the acquisition food image, including:
Multiple single frames food images are obtained in units of frame;
Split is carried out to the multiple single frames food image;
The food image after to split is processed, and identifies the food image.
9. method according to claim 7, it is characterised in that it is described respectively to the multiple single frames food image at
Reason, identifies the food image, including:
Sampling selection is carried out to the multiple single frames food image;
The single frames food image that sampling is obtained is identified respectively, to be identified result;
Each described recognition result is weighted average calculating operation;
The food image is determined according to the operation result.
10. method according to claim 7, it is characterised in that described to be carried out to the multiple single frames food image respectively
Treatment, identifies the food image, including:
The multiple single frames food image is identified, to be identified result;
Each described recognition result is weighted average calculating operation;
The food image is determined according to the operation result.
A kind of 11. diet suggestion devices, it is characterised in that including:
Acquiring unit, for obtaining food image and health data;
Recognition unit, for being identified obtaining food name data and modified data to the food image;
Factor of influence determining unit, for according to the health data and the modified data, determining the modified data to institute
State the factor of influence of health data;
Tip element, for providing a user with prompt message according to the factor of influence.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611035685.9A CN106777882A (en) | 2016-11-18 | 2016-11-18 | A kind of diet reminding method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611035685.9A CN106777882A (en) | 2016-11-18 | 2016-11-18 | A kind of diet reminding method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106777882A true CN106777882A (en) | 2017-05-31 |
Family
ID=58970309
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611035685.9A Pending CN106777882A (en) | 2016-11-18 | 2016-11-18 | A kind of diet reminding method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106777882A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108877896A (en) * | 2018-06-01 | 2018-11-23 | 四川黑石曼吉健康科技有限公司 | A kind of artificial intelligence generation method for weight management |
CN109300526A (en) * | 2018-11-06 | 2019-02-01 | 维沃移动通信有限公司 | A kind of recommended method and mobile terminal |
CN109727658A (en) * | 2019-02-22 | 2019-05-07 | 上海鹰瞳医疗科技有限公司 | Dietary recommendations continued method, apparatus and system |
CN109887574A (en) * | 2019-01-31 | 2019-06-14 | 广州市格利网络技术有限公司 | Health preserving with food and dietetic therapy guidance method and device based on food materials identification |
CN110718282A (en) * | 2019-10-14 | 2020-01-21 | 杭州睿琪软件有限公司 | Packaging food identification method and device |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104408553A (en) * | 2014-11-13 | 2015-03-11 | 曾强 | Health weight-reducing management system and management method thereof |
CN104899814A (en) * | 2015-05-08 | 2015-09-09 | 努比亚技术有限公司 | Method for intelligently reminding healthy diet and terminal |
CN105286788A (en) * | 2015-09-19 | 2016-02-03 | 深圳市前海安测信息技术有限公司 | System and method for controlling diet in patient of chronic disease based on human characteristic data |
CN105468913A (en) * | 2015-11-25 | 2016-04-06 | 哈尔滨商业大学 | Intelligent dining table system and healthy diet control system based on Internet of Things |
CN105512501A (en) * | 2016-01-05 | 2016-04-20 | 京东方科技集团股份有限公司 | Intelligent diet analysis system |
KR20160071013A (en) * | 2014-12-11 | 2016-06-21 | 엘지전자 주식회사 | Glass type mobile terminal and method for controlling the same |
CN105809598A (en) * | 2016-03-10 | 2016-07-27 | 青岛海尔智能家电科技有限公司 | Dietary intake management method, device and cloud platform |
-
2016
- 2016-11-18 CN CN201611035685.9A patent/CN106777882A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104408553A (en) * | 2014-11-13 | 2015-03-11 | 曾强 | Health weight-reducing management system and management method thereof |
KR20160071013A (en) * | 2014-12-11 | 2016-06-21 | 엘지전자 주식회사 | Glass type mobile terminal and method for controlling the same |
CN104899814A (en) * | 2015-05-08 | 2015-09-09 | 努比亚技术有限公司 | Method for intelligently reminding healthy diet and terminal |
CN105286788A (en) * | 2015-09-19 | 2016-02-03 | 深圳市前海安测信息技术有限公司 | System and method for controlling diet in patient of chronic disease based on human characteristic data |
CN105468913A (en) * | 2015-11-25 | 2016-04-06 | 哈尔滨商业大学 | Intelligent dining table system and healthy diet control system based on Internet of Things |
CN105512501A (en) * | 2016-01-05 | 2016-04-20 | 京东方科技集团股份有限公司 | Intelligent diet analysis system |
CN105809598A (en) * | 2016-03-10 | 2016-07-27 | 青岛海尔智能家电科技有限公司 | Dietary intake management method, device and cloud platform |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108877896A (en) * | 2018-06-01 | 2018-11-23 | 四川黑石曼吉健康科技有限公司 | A kind of artificial intelligence generation method for weight management |
CN108877896B (en) * | 2018-06-01 | 2022-02-22 | 四川黑石曼吉健康科技有限公司 | Artificial intelligence generated weight management method |
CN109300526A (en) * | 2018-11-06 | 2019-02-01 | 维沃移动通信有限公司 | A kind of recommended method and mobile terminal |
CN109887574A (en) * | 2019-01-31 | 2019-06-14 | 广州市格利网络技术有限公司 | Health preserving with food and dietetic therapy guidance method and device based on food materials identification |
CN109727658A (en) * | 2019-02-22 | 2019-05-07 | 上海鹰瞳医疗科技有限公司 | Dietary recommendations continued method, apparatus and system |
CN110718282A (en) * | 2019-10-14 | 2020-01-21 | 杭州睿琪软件有限公司 | Packaging food identification method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106777882A (en) | A kind of diet reminding method and device | |
CN107610087B (en) | Tongue coating automatic segmentation method based on deep learning | |
EP2973402B1 (en) | Systems, methods, and computer-readable media for identifying when a subject is likely to be affected by a medical condition | |
WO2021139471A1 (en) | Health status test method and device, and computer storage medium | |
CN108629306B (en) | Human body posture recognition method and device, electronic equipment and storage medium | |
CN108875606A (en) | A kind of classroom teaching appraisal method and system based on Expression Recognition | |
Hu et al. | Deep neural network-based speaker-aware information logging for augmentative and alternative communication | |
CN109002766A (en) | A kind of expression recognition method and device | |
CN109711422A (en) | Image real time transfer, the method for building up of model, device, computer equipment and storage medium | |
WO2021242368A1 (en) | Analysis and sorting in aquaculture | |
WO2017161734A1 (en) | Correction of human body movements via television and motion-sensing accessory and system | |
WO2024051597A1 (en) | Standard pull-up counting method, and system and storage medium therefor | |
CN107873097A (en) | Method suitable for handling asynchronous signal | |
CN109939432A (en) | A kind of intelligence rope skipping method of counting | |
Xu et al. | Optic disc detection via deep learning in fundus images | |
CN110659398A (en) | Visual question-answering method based on mathematical chart data set | |
CN111860091A (en) | Face image evaluation method and system, server and computer readable storage medium | |
CN106485266A (en) | A kind of ancient wall classifying identification method based on extraction color characteristic | |
CN106529470A (en) | Gesture recognition method based on multistage depth convolution neural network | |
Tran et al. | Adaptive pulsatile plane for robust noncontact heart rate monitoring | |
CN109726713B (en) | User region-of-interest detection system and method based on consumption-level sight tracker | |
CN110473176A (en) | Image processing method and device, method for processing fundus images, electronic equipment | |
EP3074844B1 (en) | Estimating gaze from un-calibrated eye measurement points | |
JP6732271B1 (en) | Meat quality discrimination program and system | |
Maheshwari et al. | Performance Analysis of Mango Leaf Disease using Machine Learning Technique |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170531 |
|
RJ01 | Rejection of invention patent application after publication |