CN110097139A - A kind of intelligence rice washing method and device based on convolutional neural networks - Google Patents
A kind of intelligence rice washing method and device based on convolutional neural networks Download PDFInfo
- Publication number
- CN110097139A CN110097139A CN201910393635.5A CN201910393635A CN110097139A CN 110097139 A CN110097139 A CN 110097139A CN 201910393635 A CN201910393635 A CN 201910393635A CN 110097139 A CN110097139 A CN 110097139A
- Authority
- CN
- China
- Prior art keywords
- rice
- model
- water
- insect
- training
- 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.)
- Granted
Links
Classifications
-
- A—HUMAN NECESSITIES
- A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
- A47J—KITCHEN EQUIPMENT; COFFEE MILLS; SPICE MILLS; APPARATUS FOR MAKING BEVERAGES
- A47J43/00—Implements for preparing or holding food, not provided for in other groups of this subclass
- A47J43/04—Machines for domestic use not covered elsewhere, e.g. for grinding, mixing, stirring, kneading, emulsifying, whipping or beating foodstuffs, e.g. power-driven
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Food Science & Technology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Mechanical Engineering (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
- Adjustment And Processing Of Grains (AREA)
Abstract
The intelligence rice washing method and device based on convolutional neural networks that the invention discloses a kind of, belong to artificial intelligence field, the technical problem to be solved in the present invention be how to be replaced using convolutional neural networks people using visually select reject insect, realize intelligence rice washing, the technical solution of use are as follows: 1. this method is directed to respectively using the two different models of water-model and rice-model spreads out two kinds of situations of rice and flour in the plane by the horizontal plane peace for the rice that bubble is crossed to detect in rice whether have insect, when two models of water-model and rice-model make the not judgement of insect, then determine that rice is clean;Otherwise, operation will be repeated, until two models of water-model and rice-model make the judgement of not insect.2. the device includes server, camera, water filling water fits, picture processing module and rice washing machine, server is separately connected camera, picture processing module and rice washing machine, rice washing machine and is connected with water filling water fits.
Description
Technical field
The present invention relates to a kind of artificial intelligence field, specifically a kind of intelligent rice washing side based on convolutional neural networks
Method and device.
Background technique
With the development transformation and the progress of artificial intelligence technology of the mode of production and life, nothing repeats for the time being in many lives
Life style can be replaced by artificial intelligence.In various deep neural network structures, convolutional neural networks are most widely used
General one kind, it was proposed by LeCun in 1989.Convolutional neural networks are being successfully applied to the knowledge of hand-written character image in early days
Not.Deeper AlexNet network is succeeded within 2012, and hereafter convolutional neural networks flourish, and is widely used in each
A field all achieves current best performance in many problems.In machine vision and other many problems, convolutional Neural
Network achieves current best effect.Convolutional neural networks are by the automatic study image of convolution sum pondization operation at all levels
On feature, this meets the common sense we have appreciated that image.
Existing life style is still based on artificial, but artificial intelligence technology has but obtained significant progress.Manually
Intelligence should more service life, allow people to save the time worked and go to enjoy higher quality of the life.As this life of washing rice
In must but cumbersome activity more should go to solve by the mode of artificial intelligence.Rice is the indispensable master of people in life
Food, but have a worry, same day hot air heating always has many insects with after in a pile rice, must wash in a pan repeatedly before eating every time
Rice is just edible, but this process, without very long for the time being, people can only be more time-consuming by visually selecting rejecting, and
It rejects unclean.Therefore how to replace people using rejecting insect is visually selected using convolutional neural networks, realize intelligence rice washing
It is to continue the technical issues of solving in currently available technology.
Summary of the invention
Technical assignment of the invention is to provide a kind of intelligence rice washing method and device based on convolutional neural networks, to solve
How people using visually select rejecting insect, realize intelligence rice washing the problem of is replaced using convolutional neural networks.
Technical assignment of the invention realizes in the following manner, a kind of intelligent rice washing side based on convolutional neural networks
Method, this method are directed to the water for the rice crossed by bubble using the two different models of water-model and rice-model respectively
Plane peace spreads out two kinds of situations of rice and flour in the plane to detect in rice whether have insect, as water-model and rice-
When two models of model make the not judgement of insect, it is determined that rice is clean;Otherwise, operation will be repeated, until water-
Two models of model and rice-model make the judgement of not insect;Specific step is as follows:
S1, model is trained: including water-model training and rice-model training;
S2, rice washing work is carried out using two kinds of models of trained water-model and rice-model: by trained
The two models are put into raspberry by the model water-model and rice-model of the convolutional neural networks different to two
In group, this raspberry pie and camera are embedded into and are automatically injected in the rice washing machine to draw water, washed rice using rice washing machine.
Preferably, specific step is as follows for water-model training in the step S1:
S1-11, etc. the rice of capacity pour into rice utensil, pour into water, make the equal energy of insect for having in the rice heap of insect
Levitating comes;
S1-12, horizontal plane is taken into a picture, whether horizontal plane on have insect, while by picture if observing by the naked eye
Tagged, whether the content of label as has insect;
S1-13, under different light environments, the operation of repeated several times step S1-12;
S1-14, existing photo is selectively cut, is rotated, amplified and is reduced, by all with label
Photo be fabricated to a horizontal plane data set, horizontal plane data set is divided into horizontal plane training set and horizontal plane test set
S1-15, water-model (i.e. convolutional neural networks model is transformed based on ssd model) is put into the water handled well
Panel data collection, is trained using server;
S1-16, the water-model that training is completed is put into horizontal plane test data set and tested, and judge to test
As a result whether reach default effect:
If not up to default effect is fetched again according to collection training or is finely tuned to water-model, until water-
Model fitting.
More preferably, specific step is as follows for rice-model training in the step S1:
S1-21, it will be crossed every time with bubble in step S1-11 to step S1-16 and do not do the rice of any processing and divided
On surface plate, in order to avoid insect is coated in rice layer, therefore the surface plate size chosen can just be such that rice only covers
One layer;
S1-22, identify in the rice on surface plate whether there is insect by naked eyes, with camera by the rice in surface plate
It takes photos, and tagged to picture, whether the content of label as has insect;
S1-23, under different light environments, the operation of repeated several times step S1-22;
S1-24, existing photo is selectively cut, is rotated, amplified and is reduced, by all with label
Photo be fabricated to the rice flour data set divided, rice flour data set is divided into training set and test set;
S1-25, rice-model (i.e. convolutional neural networks model, based on ssd model be transformed) is put into handle well it is flat
The rice flour training dataset at booth, is trained using server;
S1-26, the rice-model that training is completed is put into the rice flour test data set divided and tested, and judged
Whether test result reaches default effect:
If not up to default effect is fetched again according to collection training or is finely tuned to rice-model, until rice-
Model fitting.
More preferably, the openning of the rice utensil and surface plate are rectangle.
Preferably, being carried out in the step S2 using two kinds of models of trained water-model and rice-model
Washing rice, specific step is as follows for work:
S201, rice is poured into rice washing machine, the rice that rice washing machine chooses fixed capacity is poured into rice washing machine
In alternative rice space;
S202, start automatic water filling after rice is completely covered, horizontal plane is taken photos with camera;
S203, the water-model that the photo in step S203 is passed in raspberry pie is judged:
If 1., the result that obtains of water-model be to have insect, then follow the steps S204;
If 2., the judging result of water-model be no insect, go to step S205;
The insect of horizontal plane itself, can be completely drawn out, in order to prevent by S204, all extractions automatically by water while drawing water
There may be part insect not to be sucked out, ensure that the insect in rice is thoroughly removed completely, repeat step S202 and step
The operation of S203, until water-model judges the result obtained as no insect;
S205, there may be part insect there is no emersion on the water surface in order to prevent, ensure the insect in rice by thoroughly clear
Except clean, rice and water are automatically separated according to granular size, water penetrates into another space, and remaining rice is divided
Rice on bottom surface is taken into a picture by the bottom of rectangular cylinder, then camera, and photo is passed in raspberry pie, is utilized
Rice-model carries out judging whether there is insect:
If 1., have insect, fill the water again, re-start the operation of step S202 and step S203;
If 2., without insect, confirm rice be it is clean, be put into clean rice space;
S206, the operation that step S201 to step S205 is carried out to remaining rice in rice washing machine, until each portion
Dividing rice is all clean, i.e. completion rice washing operation.
More preferably, alternative rice space is in rectangular cylinder structure in the rice washing machine.
A kind of intelligent rice-washing device based on convolutional neural networks, the device include that server, camera, water filling are drawn water work
Tool, picture processing module and rice washing machine, server are separately connected camera, picture processing module and rice washing machine, rice washer
Device is connected with water filling water fits.
Preferably, the server is used for the training of model, model includes water-model and rice-model;
Camera is for shooting photo;Picture processing module is used to carry out the place of camera shooting photo using raspberry pie
Reason, including cutting, rotation, amplification and the diminution to photo;
Rice washing machine is used to carry out rice washing work using two kinds of models of trained water-model and rice-model,
The model water-model and rice-model of two different convolutional neural networks are obtained by training, by the two models
It is put into raspberry pie, this raspberry pie and camera is embedded into and are automatically injected in the rice washing machine to draw water, utilize rice washing machine
It washes rice;
Automatic pumping and water filling of the water filling water fits for machine of washing rice.
Intelligence rice washing method and device based on convolutional neural networks of the invention has the advantage that
(1), currently most popular convolutional neural networks are big to identify using artificial intelligence computer visual field by the present invention
Whether there is this problem of insect in rice, and thoroughly reject the insect in rice according to certain process, realizes intelligent rice washing;
(2), the present invention is based on convolutional neural networks come two classification methods that judge whether to have insect in a pile rice, root
It is judged that result is filled the water or drawn water, so that insect be rejected, achieve the purpose that wash in a pan in rice clean;In order to which insect is picked
Except clean, the process of rice washing is decomposed, in the case where having water and without water, respectively with based on convolutional neural networks training
Model out carries out judging whether there is insect;When only all there is no insect there are two types of in the case of, it just can determine that meter Dui Zhong is clean
's.
Detailed description of the invention
The following further describes the present invention with reference to the drawings.
Attached drawing 1 is the intelligence rice washing method flow block diagram based on convolutional neural networks;
Attached drawing 2 is model structure schematic diagram.
Specific embodiment
Referring to Figure of description and specific embodiment to a kind of intelligent rice washing side based on convolutional neural networks of the invention
Method and device are described in detail below.
Embodiment 1:
Intelligent rice washing method based on convolutional neural networks of the invention, this method utilize water-model and rice-
The two different models of model are directed to the two kinds of situations of rice and flour of the horizontal plane peace booth for the rice crossed by bubble in the plane respectively
Detect in rice whether have insect, when two models of water-model and rice-model make the judgement of not insect
When, it is determined that rice is clean;Otherwise, operation will be repeated, haunted until two models of water-model and rice-model are done
There is the judgement of insect;Specific step is as follows:
S1, model is trained: including water-model training and rice-model training;Water-model training
Specific step is as follows:
S1-11, etc. the rice of capacity pour into rice utensil, pour into water, make the equal energy of insect for having in the rice heap of insect
Levitating comes;
S1-12, horizontal plane is taken into a picture, whether horizontal plane on have insect, while by picture if observing by the naked eye
Tagged, whether the content of label as has insect;
S1-13, under different light environments, the operation of repeated several times step S1-12;
S1-14, existing photo is selectively cut, is rotated, amplified and is reduced, by all with label
Photo be fabricated to a horizontal plane data set, horizontal plane data set is divided into horizontal plane training set and horizontal plane test set
S1-15, by water-model, (i.e. convolutional neural networks model is transformed based on ssd model, and attached drawing 2 is model knot
Structure) it is put into the horizontal plane data set handled well, it is trained using server;
S1-16, the water-model that training is completed is put into horizontal plane test data set and tested, and judge to test
As a result whether reach default effect:
If not up to default effect is fetched again according to collection training or is finely tuned to water-model, until water-
Model fitting.
Specific step is as follows for rice-model training:
S1-21, it will be crossed every time with bubble in step S1-11 to step S1-16 and do not do the rice of any processing and divided
On surface plate, in order to avoid insect is coated in rice layer, therefore the surface plate size chosen can just be such that rice only covers
One layer;
S1-22, identify in the rice on surface plate whether there is insect by naked eyes, with camera by the rice in surface plate
It takes photos, and tagged to picture, whether the content of label as has insect;
S1-23, under different light environments, the operation of repeated several times step S1-22;
S1-24, existing photo is selectively cut, is rotated, amplified and is reduced, by all with label
Photo be fabricated to the rice flour data set divided, rice flour data set is divided into training set and test set;
S1-25, by rice-model, (i.e. convolutional neural networks model is transformed based on ssd model, and attached drawing 2 is model knot
Structure) it is put into the rice flour training dataset divided handled well, it is trained using server;
S1-26, the rice-model that training is completed is put into the rice flour test data set divided and tested, and judged
Whether test result reaches default effect:
If not up to default effect is fetched again according to collection training or is finely tuned to rice-model, until rice-
Model fitting.
S2, rice washing work is carried out using two kinds of models of trained water-model and rice-model: by trained
The two models are put into raspberry by the model water-model and rice-model of the convolutional neural networks different to two
In group, this raspberry pie and camera are embedded into and are automatically injected in the rice washing machine to draw water, washed rice using rice washing machine;Such as
Shown in attached drawing 1, the specific steps are as follows:
S201, rice is poured into rice washing machine, the rice that rice washing machine chooses fixed capacity is poured into rice washing machine
In alternative rice space;
S202, start automatic water filling after rice is completely covered, horizontal plane is taken photos with camera;
S203, the water-model that the photo in step S203 is passed in raspberry pie is judged:
If 1., the result that obtains of water-model be to have insect, then follow the steps S204;
If 2., the judging result of water-model be no insect, go to step S205;
The insect of horizontal plane itself, can be completely drawn out, in order to prevent by S204, all extractions automatically by water while drawing water
There may be part insect not to be sucked out, ensure that the insect in rice is thoroughly removed completely, repeat step S202 and step
The operation of S203, until water-model judges the result obtained as no insect;
S205, there may be part insect there is no emersion on the water surface in order to prevent, ensure the insect in rice by thoroughly clear
Except clean, rice and water are automatically separated according to granular size, water penetrates into another space, and remaining rice is divided
Rice on bottom surface is taken into a picture by the bottom of rectangular cylinder, then camera, and photo is passed in raspberry pie, is utilized
Rice-model carries out judging whether there is insect:
If 1., have insect, fill the water again, re-start the operation of step S202 and step S203;
If 2., without insect, confirm rice be it is clean, be put into clean rice space;
S206, the operation that step S201 to step S205 is carried out to remaining rice in rice washing machine, until each portion
Dividing rice is all clean, i.e. completion rice washing operation.
Wherein, the openning of rice utensil and surface plate are rectangle.Alternative rice space is in rectangular column in rice washing machine
Body structure.
Embodiment 2:
Intelligent rice-washing device based on convolutional neural networks of the invention, the device include server, camera, water filling pumping
Water conservancy project tool, picture processing module and rice washing machine, server are separately connected camera, picture processing module and rice washing machine, wash in a pan
Rice machine is connected with water filling water fits.Server is used for the training of model, and model includes water-model and rice-
model;Camera is for shooting photo;Picture processing module is used to carry out the processing of camera shooting photo using raspberry pie,
Including cutting, rotation, amplification and the diminution to photo;Machine of washing rice is used to utilize trained water-model and rice-
Two kinds of models of model carry out rice washing work, obtain the model water-model of two different convolutional neural networks by training
And rice-model, the two models are put into raspberry pie, this raspberry pie and camera are embedded into be automatically injected and drawn water
Rice washing machine in, using rice washing machine wash rice;Automatic pumping and water filling of the water filling water fits for machine of washing rice.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations;To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement;And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (8)
1. a kind of intelligent rice washing method based on convolutional neural networks, which is characterized in that this method using water-model and
The two different models of rice-model are directed to the rice and flour two of the horizontal plane peace booth for the rice crossed by bubble in the plane respectively
Situation is planted to detect in rice whether have insect, when two models of water-model and rice-model make no insect
Judgement when, it is determined that rice is clean;Otherwise, operation will be repeated, until water-model and two models of rice-model are equal
Make the judgement of not insect;Specific step is as follows:
S1, model is trained: including water-model training and rice-model training;
S2, rice washing work is carried out using two kinds of models of trained water-model and rice-model: obtaining two by training
The model water-model and rice-model of a different convolutional neural networks, the two models are put into raspberry pie,
This raspberry pie and camera are embedded into and are automatically injected in the rice washing machine to draw water, is washed rice using rice washing machine.
2. the intelligent rice washing method according to claim 1 based on convolutional neural networks, which is characterized in that the step S1
Specific step is as follows for middle water-model training:
S1-11, etc. the rice of capacity pour into rice utensil, pour into water, make have the insect in the rice heap of insect can levitating
Come;
S1-12, horizontal plane is taken into a picture, whether have insect, while picture being stamped if observing by the naked eye on horizontal plane
Whether label, the content of label as have insect;
S1-13, under different light environments, the operation of repeated several times step S1-12;
S1-14, existing photo is selectively cut, is rotated, amplified and is reduced, by all photographs with label
Piece is fabricated to a horizontal plane data set, and horizontal plane data set is divided into horizontal plane training set and horizontal plane test set
S1-15, water-model is put into the horizontal plane data set handled well, is trained using server;
S1-16, the water-model that training is completed is put into horizontal plane test data set and tested, and judge test result
Whether default effect is reached:
If not up to default effect is fetched again according to collection training or is finely tuned to water-model, until water-model
Fitting.
3. the intelligent rice washing method according to claim 1 or 2 based on convolutional neural networks, which is characterized in that the step
Specific step is as follows for rice-model training in rapid S1:
S1-21, it will be crossed every time with bubble in step S1-11 to step S1-16 and do not do the rice of any processing and divided flat
On panel, in order to avoid insect is coated in rice layer, therefore the surface plate size chosen just can make rice only cover one layer;
S1-22, by naked eyes identify surface plate on rice in whether have insect, with camera by surface plate rice clap at
Photo, and it is tagged to picture, and whether the content of label as has insect;
S1-23, under different light environments, the operation of repeated several times step S1-22;
S1-24, existing photo is selectively cut, is rotated, amplified and is reduced, by all photographs with label
Piece is fabricated to the rice flour data set divided, and rice flour data set is divided into training set and test set;
S1-25, rice-model is put into the rice flour training dataset divided handled well, is trained using server;
S1-26, the rice-model that training is completed is put into the rice flour test data set divided and tested, and judge to test
As a result whether reach default effect:
If not up to default effect is fetched again according to collection training or is finely tuned to rice-model, until rice-model is quasi-
It closes.
4. the intelligent rice washing method according to claim 3 based on convolutional neural networks, which is characterized in that the rice device
The openning and surface plate of tool are rectangle.
5. the intelligent rice washing method according to claim 1 based on convolutional neural networks, which is characterized in that the step S2
It is middle using two kinds of models of trained water-model and rice-model carry out rice washing work specific step is as follows:
S201, rice is poured into rice washing machine, the rice that rice washing machine chooses fixed capacity is poured into alternative in rice washing machine
In rice space;
S202, start automatic water filling after rice is completely covered, horizontal plane is taken photos with camera;
S203, the water-model that the photo in step S203 is passed in raspberry pie is judged:
If 1., the result that obtains of water-model be to have insect, then follow the steps S204;
If 2., the judging result of water-model be no insect, go to step S205;
The insect of horizontal plane itself, can be completely drawn out while drawing water, in order to prevent may by S204, all extractions automatically by water
There is part insect not to be sucked out, ensures that the insect in rice is thoroughly removed completely, repeat step S202's and step S203
Operation, until water-model judges the result obtained as no insect;
S205, there may be part insect there is no emersion on the water surface in order to prevent, it is dry to ensure that the insect in rice is thoroughly removed
Only, rice and water are automatically separated according to granular size, water penetrates into another space, and remaining rice is divided in rectangle
Rice on bottom surface is taken into a picture by the bottom of cylinder, then camera, and photo is passed in raspberry pie, rice- is utilized
Model carries out judging whether there is insect:
If 1., have insect, fill the water again, re-start the operation of step S202 and step S203;
If 2., without insect, confirm rice be it is clean, be put into clean rice space;
S206, the operation that step S201 to step S205 is carried out to remaining rice in rice washing machine, until each section is big
Rice is all clean, i.e. completion rice washing operation.
6. the intelligent rice washing method according to claim 5 based on convolutional neural networks, which is characterized in that the rice washer
Alternative rice space is in rectangular cylinder structure in device.
7. a kind of intelligent rice-washing device based on convolutional neural networks, which is characterized in that the device include server, camera,
It fills the water water fits, picture processing module and rice washing machine, server and is separately connected camera, picture processing module and rice washer
Device, rice washing machine are connected with water filling water fits.
8. the intelligent rice-washing device according to claim 7 based on convolutional neural networks, which is characterized in that the server
For the training of model, model includes water-model and rice-model;
Camera is for shooting photo;Picture processing module is used to carry out the processing of camera shooting photo, packet using raspberry pie
Include cutting, rotation, amplification and the diminution to photo;
Rice washing machine is used to carry out rice washing work using two kinds of models of trained water-model and rice-model, passes through
Training obtains the model water-model and rice-model of two different convolutional neural networks, the two models are put into
Into raspberry pie, this raspberry pie and camera are embedded into and are automatically injected in the rice washing machine to draw water, carried out using rice washing machine
Rice washing;
Automatic pumping and water filling of the water filling water fits for machine of washing rice.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910393635.5A CN110097139B (en) | 2019-05-13 | 2019-05-13 | Intelligent rice washing method and device based on convolutional neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910393635.5A CN110097139B (en) | 2019-05-13 | 2019-05-13 | Intelligent rice washing method and device based on convolutional neural network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110097139A true CN110097139A (en) | 2019-08-06 |
CN110097139B CN110097139B (en) | 2022-12-13 |
Family
ID=67447854
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910393635.5A Active CN110097139B (en) | 2019-05-13 | 2019-05-13 | Intelligent rice washing method and device based on convolutional neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110097139B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113837049A (en) * | 2021-09-17 | 2021-12-24 | 山东浪潮科学研究院有限公司 | Intelligent road surface cleaning method based on convolutional neural network and genetic algorithm |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016195496A2 (en) * | 2015-06-05 | 2016-12-08 | Universiteit Van Amsterdam | Deep receptive field networks |
CN109740560A (en) * | 2019-01-11 | 2019-05-10 | 济南浪潮高新科技投资发展有限公司 | Human cellular protein automatic identifying method and system based on convolutional neural networks |
-
2019
- 2019-05-13 CN CN201910393635.5A patent/CN110097139B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016195496A2 (en) * | 2015-06-05 | 2016-12-08 | Universiteit Van Amsterdam | Deep receptive field networks |
CN109740560A (en) * | 2019-01-11 | 2019-05-10 | 济南浪潮高新科技投资发展有限公司 | Human cellular protein automatic identifying method and system based on convolutional neural networks |
Non-Patent Citations (1)
Title |
---|
邢艳芳等: "TensorFlow在图像识别系统中的应用", 《计算机技术与发展》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113837049A (en) * | 2021-09-17 | 2021-12-24 | 山东浪潮科学研究院有限公司 | Intelligent road surface cleaning method based on convolutional neural network and genetic algorithm |
CN113837049B (en) * | 2021-09-17 | 2023-06-20 | 山东浪潮科学研究院有限公司 | Intelligent pavement cleaning method based on convolutional neural network and genetic algorithm |
Also Published As
Publication number | Publication date |
---|---|
CN110097139B (en) | 2022-12-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106529564B (en) | A kind of food image automatic classification method based on convolutional neural networks | |
CN106295789A (en) | A kind of crop seed method of counting based on image procossing | |
CN109636772A (en) | The defect inspection method on the irregular shape intermetallic composite coating surface based on deep learning | |
CN108445746A (en) | A kind of intelligence feeds control method and device | |
CN109376636A (en) | Eye ground image classification method based on capsule network | |
CN102110294B (en) | Method and system for processing image of diseased fish body | |
CN107832780A (en) | Low confidence sample processing method and system are sorted based on artificial intelligence plank | |
CN108596338A (en) | A kind of acquisition methods and its system of neural metwork training collection | |
CN208064246U (en) | A kind of intelligence feeds control device | |
CN109584209A (en) | Vascular wall patch identifies equipment, system, method and storage medium | |
CN108921105A (en) | Identify the method, apparatus and computer readable storage medium of destination number | |
CN108108767A (en) | A kind of cereal recognition methods, device and computer storage media | |
CN108866934A (en) | A kind of laundry mode control system and method based on artificial intelligence | |
CN106295502A (en) | A kind of method for detecting human face and device | |
CN109344888A (en) | A kind of image-recognizing method based on convolutional neural networks, device and equipment | |
CN111524144A (en) | Intelligent pulmonary nodule diagnosis method based on GAN and Unet network | |
CN110458829A (en) | Image quality control method, device, equipment and storage medium based on artificial intelligence | |
CN109740721A (en) | Wheat head method of counting and device | |
CN113160966B (en) | Tongue picture diagnosis method and tongue picture diagnosis system based on multitask learning | |
CN109117703A (en) | It is a kind of that cell category identification method is mixed based on fine granularity identification | |
CN108596871A (en) | A kind of BGA air blister defect image detection methods based on deep learning | |
CN110097139A (en) | A kind of intelligence rice washing method and device based on convolutional neural networks | |
CN108734708A (en) | Gastric cancer recognition methods, device and storage medium | |
CN110060246A (en) | A kind of image processing method, equipment and storage medium | |
CN112906700B (en) | Self-compacting concrete image semantic segmentation method and device and data set generation method |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20221121 Address after: 250100 building S02, No. 1036, Langchao Road, high tech Zone, Jinan City, Shandong Province Applicant after: Shandong Inspur Scientific Research Institute Co.,Ltd. Address before: 250100 First Floor of R&D Building 2877 Kehang Road, Sun Village Town, Jinan High-tech Zone, Shandong Province Applicant before: JINAN INSPUR HIGH-TECH TECHNOLOGY DEVELOPMENT Co.,Ltd. |
|
GR01 | Patent grant | ||
GR01 | Patent grant |