CN109884088A - Fruit maturity compartment intelligent checking system and detection method - Google Patents
Fruit maturity compartment intelligent checking system and detection method Download PDFInfo
- Publication number
- CN109884088A CN109884088A CN201910217592.5A CN201910217592A CN109884088A CN 109884088 A CN109884088 A CN 109884088A CN 201910217592 A CN201910217592 A CN 201910217592A CN 109884088 A CN109884088 A CN 109884088A
- Authority
- CN
- China
- Prior art keywords
- data
- fruit
- maturity
- module
- cloud
- 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
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The present invention provides a kind of fruit maturity compartment intelligent checking system and detection methods, including sensor module, cloud service module and control module, the control module to be connected respectively with sensor module, cloud service module；The cloud service module includes Cloud Server submodule and cloud database subsystem module, and the sensor module includes ultra wide band millimetre-wave radar sensor；The control module includes single-chip microcontroller.This system, which makes fruit-picking not need removal protective bag in the process, can judge the maturity of fruit, can not only reduce artificial judgment bring economic loss, moreover it is possible to which guarantee picking is ripening fruits, improves working efficiency.
Description
Technical field
The present invention relates to a kind of non-contact types of materials detection system and detection method based on ultra wide band millimetre-wave radar,
More particularly, to a kind of fruit maturity compartment intelligent monitor system and detection method.
Background technique
In recent years, intelligent fruit maturity detection has developed into the one of agricultural in the quick forming process of wisdom agricultural
A important research direction.Requirement cannot use pesticide in organic fruits planting process, current most of orchards take fruit at
The mode for putting on paper bag for a long time makes destruction of the fruit in developmental process from pest and disease damage and birds, the paper bag once put on just
It being shut with staple, is just opened after picking fruit gets off, plucker can not see the growth situation of fruit by paper bag,
Weight by weighing fruit in the hand is unable to judge accurately fruit maturity, can only carry out the judgement of maturity by gently pinching fruit at present,
And the dynamics pinched is slightly heavy to bring scar to fruit.Therefore fruit in picking process how across paper bag judge fruit at
Ripe degree becomes very difficult.
Existing fruit maturity detection method includes near infrared spectroscopy, gas sensor array method, digital image method
Deng.Near infrared spectroscopy is to be associated with spectral value with the foundation of the internal component content of fruit with a variety of Mathematical Modeling Methods, is selected
Unsuitable method will lead to crossing for model built and adapt to or owe adapt to；The overfitting phenomenon of regression analysis will lead to certain heavy
The wave band wanted is ignored, and the correctness of data can be by the SNR influence of spectrometer itself；The accuracy of spectrographic detection is easy quilt
Peel thickness influences to bring certain error.Near-infrared spectrum technique is largely only in laboratory scope to fruit
Quality is detected, and real commercial applications are seldom formed.Gas sensor array method is mainly the air-sensitive passed through built in it
The volatile gas of sensor array and tested sample occurs transient response and obtains sample information.Digital image method measurement is accurate,
It is the relationship established between the maturity of fruit and colouring information, but has radiation residual, and instrument and equipment is expensive, measurement
Process and fruit eat process to body nocuousness.Although these existing methods can reach the non-destructive testing of fruit maturity,
But it uses, fruit maturity can not be examined across paper bag in picking process when can only remove protective bag after picking fruit
It surveys.
The main cause that loss late of the fruit in multiple links such as picking, packaging, preservation, transport and processing increases
It is: does not know maturity and the storage that mixed after fruit picking.And it can according to the fruit that orchard worker's experience is picked
Can be overdone or not yet mature, this will bring great loss to orchard worker.Therefore urgently occurs a kind of non-contact compartment water in the market
Fruit maturity intelligent checking system.
Summary of the invention
The present invention is overcomes the prior art to provide a kind of fruit maturity compartment intelligence based on ultra wide band millimetre-wave radar
Energy detection system and detection method, detectable range include the thin skins class fruit such as apple, orange, bergamot pear and peach.This system makes
Do not need removal protective bag during fruit-picking and can judge the maturity of fruit, can not only reduce artificial judgment and bring
Economic loss, moreover it is possible to guarantee mechanical picking is ripening fruits, improve working efficiency.
In order to achieve the above object, the technical scheme is that
A kind of fruit maturity compartment intelligent checking system, including sensor module, cloud service module and control module, institute
The control module stated is connected with sensor module, cloud service module respectively；The cloud service module includes Cloud Server submodule
Block and cloud database subsystem module, the sensor module include ultra wide band millimetre-wave radar sensor；The control module
Including single-chip microcontroller.
A kind of detection method of fruit maturity compartment intelligent checking system: the single-chip microcontroller in the control module is utilized
Control ultra wide band millimetre-wave radar sensor emission millimeter-wave signal simultaneously receives reflection signal progress pericarp information data acquisition,
Data prediction is carried out to the reflection signal that ultra wide band millimetre-wave radar sensor obtains using the single-chip microcontroller simultaneously and is deposited
Storage；Wherein the fruit of differing maturity has the sugar of different content, acid ultra wide band millimeter wave thunder different with hardness, described
Be the equal of the data of different material up to the data that sensor receives, obtain the current signal strength of surveyed fruit, it is then right
Detect fruit for a period of time in signal detection peak value point line, the stable signal intensity envelope of the fruit and its steady can be obtained
Fixed signal envelope waveform diagram data；Above-mentioned all data are entered into the storage of cloud database subsystem module；It is then based on pretreated mesh
Data are marked, are classified using the Cloud Server submodule to data, the cloud database subsystem module calling classification device
The fruit data measured and canned data are subjected to data analysis and comparison, to judge fruit maturity；Last cloud service module
Interface display can be made to analysis result and data feedback is entered into control module.
The data prediction is to carry out target detection, including filtering processing using the single-chip microcontroller of the control module
And extract two steps of waveform frequency domain character；Radar data is screened first, using Kalman filtering to noise-containing detection
Signal is handled, and the interference of data is removed, and the smallest actual signal intensity R of error is acquired in the sense that average, wherein believing
Number maximum of intensity, that is, each crest location is labeled as Q；Then frequency domain character, the waveform diagram medium wave of return are extracted on this waveform
Peak is signal most strength, and one section of Wave data of specific length is intercepted centered on Q, is extracted according to the interphase of certain wave crest number
Feature forms test sample, and the training set that a certain number of sample constitutes input is randomly selected from test sample, is taken for cloud
Business device identifies classification to the maturity of the detection fruit.
Canned data source in the cloud database subsystem module are as follows: return the pretreated target data of control module
Incoming cloud database subsystem module storage；The Cloud Server carries out big data analysis so that it is determined that fruit maturity to training set
Judge parameter, then classify to a large amount of measurement data of differing maturity fruit by sorting algorithm, the cloud
The fruit data measured and canned data are carried out data analysis and comparison by database subsystem module calling classification device, are joined in conjunction with judgement
Number description fruit maturity classification results.
The sorting algorithm uses KNN (k-NearestNeighbor nearest neighbor algorithm), SVM (Support Vector
Machine, algorithm of support vector machine), CNN (Convolutional Neural Networks, convolutional neural networks algorithm) calculate
One of method.
The invention has the benefit that the differing maturity that the present invention is received by ultra wide band millimetre-wave radar sensor
The difference of fruit signal data, received signal strength acquire later into cloud database and carry out data comparison, can detect differentiation
The thin skins class fruit such as apple, orange, bergamot pear and peach of differing maturity.The present invention is sensed based on ultra wide band millimetre-wave radar
The fruit maturity compartment detection system of device, detectable signal use millimeter wave, and wave form is continuous wave (CW) signal；With it is existing
Visual light imaging compared with infrared thermal imaging technique, millimeter wave detection accuracy is higher, and have penetrate cigarette, dust and mist
Ability, can all weather operations；There is low-power consumption, low cost, without contact compared with other traditional fruit maturity detection techniques
Fruit surface and the advantages of without destroying fruit structure；The present invention without remove the fruit growth stage cover upper protective bag can be
Fruit picking carries out the judgement of maturity before getting off, can not only reduce artificial judgment bring economic loss, moreover it is possible to guarantee
Picking is ripening fruits, and orchard worker can according to testing result classify the fruit of differing maturity, improves working efficiency.
This method strong flexibility and there is practical value, the identification of high-precision maturity may be implemented, there is boundless application
Prospect.
Detailed description of the invention
Fig. 1 is system module connection figure of the invention；
Fig. 2 is the schematic diagram of millimetre-wave radar data fusion of the present invention；
Fig. 3 is maturity testing process analysis chart of the present invention；
Fig. 4 is the present invention using CNN convolutional neural networks as the data classification flow chart of example；
Fig. 5 is the signal envelope figure that fruit X does not cover paper bag measurement in experimentation of the present invention；
Fig. 6 is the signal envelope figure that fruit X puts on paper bag measurement in experimentation of the present invention；
Fig. 7 is the signal envelope figure that fruit Y does not cover paper bag measurement in experimentation of the present invention；
Fig. 8 is the signal envelope figure that fruit Y puts on paper bag measurement in experimentation of the present invention；
Fig. 9 is the present invention using fruit Z as 4 mature side waveform diagrams of example；
Figure 10 is the present invention using fruit Z as 9 mature side waveform diagrams of example；
Figure 11 is the present invention using KNN nearest neighbor algorithm as the data classification flow chart of example；
Figure 12 is the present invention using SVM algorithm of support vector machine as the classifier training figure of example；
Figure 13 is the present invention using SVM algorithm of support vector machine as the assorting process figure of example.
Specific embodiment
Embodiment 1
A kind of fruit maturity compartment intelligent checking system of the present embodiment, as shown in Figure 1, including sensor module, cloud
Service module and control module, the control module are connected with sensor module, cloud service module respectively；The cloud service
Module includes Cloud Server submodule and cloud database subsystem module, and the sensor module includes low-power consumption ultra wide band millimeter wave
Radar sensor；The control module includes single-chip microcontroller.
The detection method of a kind of fruit maturity compartment intelligent checking system of the present embodiment: as shown in figure 3, described in utilizing
Control module in single-chip microcontroller control ultra wide band millimetre-wave radar sensor emission millimeter-wave signal and receive reflection signal into
The acquisition of row pericarp information data, while the reflection signal that ultra wide band millimetre-wave radar sensor is obtained using the single-chip microcontroller
It carries out data prediction and stores；Wherein the fruit of differing maturity has the sugar of different content, acid, institute different with hardness
The data that the ultra wide band millimetre-wave radar sensor stated receives are the equal of the data of different material, obtain working as surveyed fruit
Front signal intensity, then the signal detection peak value point line interior for a period of time to detection fruit, can obtain the stable letter of the fruit
Number variation envelope and its stable signal envelope waveform diagram data, the present embodiment is as shown in Fig. 5, Fig. 6, Fig. 7, Fig. 8, Fig. 5, figure
7 be respectively in the present embodiment experimentation fruit X and fruit Y do not cover the signal envelope figure that paper bag measures, horizontal axis indicates to survey
Span is from the longitudinal axis indicates measuring amplitude；Fig. 6, Fig. 8 be respectively in the present embodiment experimentation fruit X and fruit Y put on paper bag into
The signal envelope figure of row measurement, horizontal axis indicate measurement distance, and the longitudinal axis indicates measuring amplitude.As shown in Fig. 5, Fig. 6, Fig. 7, Fig. 8, it is
System can carry out compartment measurement fruit maturity to fruit, and waveform diagram only has minimum error.Above-mentioned all data are passed to
The storage of cloud database subsystem module；It is then based on pretreated target data, carries out data using the Cloud Server submodule
Classification, the present embodiment as shown in figure 4, Fig. 4 be of the invention with convolutional neural networks algorithm (CNN) be example data classification process
Pretreated data training set as input is established convolutional neural networks model by figure, and training data obtains maturity point
Class device.The cloud database subsystem module calling classification device judges fruit maturity；Last cloud service module can be to analysis result
It makes interface display and data feedback is entered into control module.
The data prediction is to carry out target detection, including filtering processing using the single-chip microcontroller of the control module
And extract two steps of waveform frequency domain character；Radar data is screened first, using Kalman filtering to noise-containing detection
Signal is handled, and the interference of data is removed, and acquires the smallest actual signal intensity R of error in the sense that average；Then exist
Frequency domain character is extracted on this waveform, the present embodiment will be located in advance as shown in figure 4, wherein feature extraction is realized by convolutional neural networks
Data training set as input after reason, establishes convolutional neural networks model, is received by the convolutional layer of convolutional neural networks defeated
The feature vector of the training set entered calculates characteristic results by function, and formula is as follows:
Indicate the feature vector of convolutional layer k convolution kernel j, DjIndicate the receptive field of neuron,It is convolutional layer k convolution
I-th of weighting coefficient of core j,It is the biasing coefficient of convolutional layer k convolution kernel j.The training as input of the feature set of extraction
Collection identifies classification for maturity of the Cloud Server to the detection fruit.
Canned data source in the cloud database subsystem module are as follows: return the pretreated target data of control module
Incoming cloud database subsystem module storage；The Cloud Server carries out big data analysis so that it is determined that fruit maturity to training set
Judge parameter, then classify to a large amount of measurement data of differing maturity fruit by sorting algorithm, the cloud
Database subsystem module calling classification device judges fruit maturity grade, finally shows that the maturity of fruit analyzes knot in equipment end
Fruit.
Fig. 9 is the present embodiment using fruit Z as 4 mature side waveform diagrams of example, and horizontal axis indicates measurement distance, and the longitudinal axis indicates
Measuring amplitude；Figure 10 is the present embodiment using fruit Z as 9 mature side waveform diagrams of example, and horizontal axis indicates measurement distance, longitudinal axis table
Show measuring amplitude.To same fruit but the different individual of maturity detects, and the millimeter-wave signal being reflected back can be a certain
Occur peak value in wavelength, the variation of this peak value can with the content of internal physiological index such as soluble sugar when fruit maturation, can
The content or hardness for dripping acid are associated, and the fruit of differing maturity is just different substance for millimetre-wave radar, according to
This characteristic, as shown in Figure 9, Figure 10, it is feasible for carrying out compartment detection.
The differing maturity fruit signal data that the present embodiment is received by ultra wide band millimetre-wave radar sensor is not
Together, acquire after received signal strength into cloud database and carry out data comparison, can detect distinguish differing maturity apple,
The thin skins class fruit such as orange, bergamot pear and peach.The present embodiment carries out data analysis with the detection data that sensor obtains for foundation,
It establishes data set and is stored in cloud database for the comparison of late detection process, the maturity after analysis relatively is as the result is shown in terminal
In equipment.The present embodiment to the maturity of fruit carried out it is more intelligent, more comprehensively, it is more accurate test and analyze, solve existing
Non-destructive testing scheme can not judge before fruit picking maturity and can not compartment detection the problem of, high degree reduce
Waste degree of the fruit in picking process convenient for storing fruit according to maturity grade difference reduces orchard worker
Economic loss.
Embodiment 2
A kind of fruit maturity compartment intelligent checking system of the present embodiment, as shown in Figure 1, including sensor module, cloud
Service module and control module, the control module are connected with sensor module, cloud service module respectively；The cloud service
Module includes Cloud Server submodule and cloud database subsystem module, and the sensor module includes low-power consumption ultra wide band millimeter wave
Radar sensor；The control module includes single-chip microcontroller.
The detection method of a kind of fruit maturity compartment intelligent checking system of the present embodiment: as shown in figure 3, described in utilizing
Control module in single-chip microcontroller control ultra wide band millimetre-wave radar sensor emission millimeter-wave signal and receive reflection signal into
The acquisition of row pericarp information data, while the reflection signal that ultra wide band millimetre-wave radar sensor is obtained using the single-chip microcontroller
It carries out data prediction and stores, finally the comparative analysis of database progress data beyond the clouds.
The data prediction includes filtering processing, extracts waveform frequency domain character and data classification three parts.Specifically:
Radar data is screened first, and noise-containing detection signal is handled using Kalman filtering, removes the interference of data,
The smallest actual signal intensity R of error is acquired in the sense that average；Then on this waveform extract frequency domain character refer to data into
Row classification.The present embodiment is as shown in figure 11, and assorting process is realized by KNN (k-Nearest Neighbor nearest neighbor algorithm), Figure 11
For the classification process figure of KNN algorithm in the present embodiment, the waveform frequency domain character value of extraction is as unfiled sample data set
Input calculates in the data and the cloud database of unfiled sample data set each data in classification samples data set
Distance, calculation formula selects Euclidean distance, and formula is as follows:
Preceding K most like data in classification samples data set are selected, wherein the most sorting parameter of frequency of occurrence is made
For the sorting parameter of unfiled sample data.Fruit data that the cloud database subsystem module calling classification device will measure with
Deposit data carries out data analysis and comparison, and data analysis and comparison treatment process need ultra wide band millimetre-wave radar sensor radar
Thread and cloud database subsystem module cloud handle the two-part data of thread.As shown in Fig. 2, data are transmitted to cloud service mould by sensor
Block carries out data storage analysis, while legacy data can be added in buffer queue for cloud convergence analysis, and combining classification parameter is retouched
Fruit maturity classification results are stated, finally show that the maturity of fruit analyzes result in equipment end.
Embodiment 3
A kind of fruit maturity compartment intelligent checking system of the present embodiment, as shown in Figure 1, including sensor module, cloud
Service module and control module, the control module are connected with sensor module, cloud service module respectively；The cloud service
Module includes Cloud Server submodule and cloud database subsystem module, and the sensor module includes low-power consumption ultra wide band millimeter wave
Radar sensor；The control module includes single-chip microcontroller.
The detection method of a kind of fruit maturity compartment intelligent checking system of the present embodiment: as shown in figure 3, described in utilizing
Control module in single-chip microcontroller control ultra wide band millimetre-wave radar sensor emission millimeter-wave signal and receive reflection signal into
The acquisition of row pericarp information data, while the reflection signal that ultra wide band millimetre-wave radar sensor is obtained using the single-chip microcontroller
It carries out data prediction and stores；Wherein the fruit of differing maturity has the sugar of different content, acid, institute different with hardness
The data that the ultra wide band millimetre-wave radar sensor stated receives are the equal of the data of different material, obtain working as surveyed fruit
Front signal intensity, then the signal detection peak value point line interior for a period of time to detection fruit, can obtain the stable letter of the fruit
Number variation envelope and its stable signal envelope waveform diagram data.Based on pretreated target data, taken using the cloud
Device submodule of being engaged in carries out data classification, the present embodiment such as Figure 12, and shown in Figure 13, it with svm classifier algorithm is real that Figure 12, which is the present invention,
Pretreated data training set as input is established SVM classifier model by the classifier training figure of example, is obtained applicable
In the classifier of fruit maturity classification.Figure 13 is the present invention using svm classifier algorithm as the assorting process figure of example, the cloud
Data are sent into the classifier of maturity classification by database subsystem module calling classification device, and fruit maturation is judged by classification results
Degree.Last cloud service module can make interface display to analysis result and data feedback is entered control module.
The data prediction is to carry out target detection, including filtering processing using the single-chip microcontroller of the control module
And extract two steps of waveform frequency domain character；Radar data is screened first, using Kalman filtering to noise-containing detection
Signal is handled, and the interference of data is removed, and acquires the smallest actual signal intensity R of error in the sense that average；Then exist
Frequency domain character is extracted on this waveform, the present embodiment such as Figure 12, shown in Figure 13, wherein feature extraction is realized by Fourier transformation, will
Pretreated data training set as input, establishes SVM classifier model, formula is as follows:
What wherein y was indicated is the signal strength returned, and w indicates weight matrix representated by data, and what x was represented is distance
Value.The purpose of argmax is to extract optimal classification solution.The feature set of extraction training set as input is used for cloud service
Device carries out machine learning training export classifier.
Canned data source in the cloud database subsystem module are as follows: return the pretreated target data of control module
Incoming cloud database subsystem module storage；The Cloud Server carries out big data analysis so that it is determined that fruit maturity to training set
Judge parameter, then classify to a large amount of measurement data of differing maturity fruit by sorting algorithm, the cloud
Database subsystem module calling classification device judges that fruit maturity is classified, and finally shows that the maturity of fruit analyzes knot in equipment end
Fruit.
The differing maturity fruit signal data that the present embodiment is received by ultra wide band millimetre-wave radar sensor is not
Together, it is acquired after received signal strength and carries out data classification using classifier, apple, the tangerine for distinguishing differing maturity can be detected
The thin skins class fruit such as son, bergamot pear and peach.The present embodiment carries out data analysis with the detection data that sensor obtains for foundation, builds
Vertical data set is stored in cloud database and uses for later period training classifier, and the maturity that classifier obtains is set in terminal as the result is shown
It is standby upper.The present embodiment to the maturity of fruit carried out it is more intelligent, more comprehensively, it is more accurate test and analyze, solve existing
Non-destructive testing scheme can not judge before fruit picking maturity and can not compartment detection the problem of, high degree reduces
Waste degree of the fruit in picking process reduces orchard worker's convenient for storing fruit according to maturity grade difference
Economic loss.
Claims (5)
1. a kind of fruit maturity compartment intelligent checking system, it is characterised in that including sensor module, cloud service module and control
Molding block, the control module are connected with sensor module, cloud service module respectively；The cloud service module includes cloud clothes
Business device submodule and cloud database subsystem module, the sensor module include ultra wide band millimetre-wave radar sensor；Described
Control module includes single-chip microcontroller.
2. a kind of fruit maturity compartment intelligent checking system, it is characterised in that: the system uses following detection method: benefit
With the single-chip microcontroller control ultra wide band millimetre-wave radar sensor emission millimeter-wave signal in the control module and receive reflection
Signal carries out the acquisition of pericarp information data, while it is anti-to utilize the single-chip microcontroller to obtain ultra wide band millimetre-wave radar sensor
Signal is penetrated to carry out data prediction and store；Wherein the fruit of differing maturity has the sugar of different content, acid and hardness
Difference, the data that the ultra wide band millimetre-wave radar sensor receives are the equal of the data of different material, obtain being surveyed
The current signal strength of fruit, then the signal detection peak value point line interior for a period of time to detection fruit, can obtain the fruit
Stable signal intensity envelope and its stable signal envelope waveform diagram data；Above-mentioned all data are entered into cloud database submodule
Block storage；It is then based on pretreated target data, is classified using the Cloud Server submodule to data, it is described
The fruit data measured and canned data are carried out data analysis and comparison by cloud database subsystem module calling classification device, to judge water
Fruit maturity；Last cloud service module can make interface display to analysis result and data feedback is entered control module.
3. a kind of fruit maturity compartment intelligent checking system as claimed in claim 2, it is characterised in that: the data are pre-
Processing is to carry out target detection using the single-chip microcontroller of the control module, including be filtered and extract waveform frequency domain character
Two steps；Radar data is screened first, noise-containing detection signal is handled using Kalman filtering, removes data
Interference, acquire the smallest actual signal intensity R of error in the sense that average1, wherein signal strength maximum value, that is, each wave
Peak position is labeled as Q；Then frequency domain character is extracted on this waveform, the waveform diagram medium wave peak of return is signal most strength, is with Q
Center intercepts one section of Wave data of specific length, extracts feature according to the interphase of certain wave crest number and forms test sample, from
Randomly select the training set that a certain number of sample constitutes input in test sample, for Cloud Server to the detection fruit at
Ripe degree identification classification.
4. a kind of fruit maturity compartment intelligent checking system as claimed in claim 2, it is characterised in that: the cloud data
Canned data source in the submodule of library are as follows: the pretreated target data of control module is returned and is deposited into cloud database subsystem module
Storage；The Cloud Server carries out big data analysis to training set so that it is determined that fruit maturity judges parameter, then by dividing
Class algorithm classify to a large amount of measurement data of differing maturity fruit, the cloud database subsystem module calling classification
The fruit data measured and canned data are carried out data analysis and comparison by device, in conjunction with judge that parameter describes fruit maturity and classify
As a result.
5. a kind of fruit maturity compartment intelligent checking system as claimed in claim 4, it is characterised in that: the classification is calculated
Method using k-Nearest Neighbor nearest neighbor algorithm, Support Vector Machine algorithm of support vector machine,
One of Convolutional Neural Networks convolutional neural networks algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910217592.5A CN109884088A (en) | 2019-03-21 | 2019-03-21 | Fruit maturity compartment intelligent checking system and detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910217592.5A CN109884088A (en) | 2019-03-21 | 2019-03-21 | Fruit maturity compartment intelligent checking system and detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109884088A true CN109884088A (en) | 2019-06-14 |
Family
ID=66933460
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910217592.5A Pending CN109884088A (en) | 2019-03-21 | 2019-03-21 | Fruit maturity compartment intelligent checking system and detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109884088A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110969090A (en) * | 2019-11-04 | 2020-04-07 | 口碑(上海)信息技术有限公司 | Fruit quality identification method and device based on deep neural network |
| CN111487247A (en) * | 2020-04-28 | 2020-08-04 | 前海国兴(深圳)高科技有限公司 | Fruit maturity AI identification system |
| CN114062505A (en) * | 2022-01-17 | 2022-02-18 | 季华实验室 | Fruit maturity detection method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104515751A (en) * | 2014-12-24 | 2015-04-15 | 河南农业大学 | Flight time spectral fruit maturity non-destructive and rapid detection method and system |
| US20160011162A1 (en) * | 2012-04-16 | 2016-01-14 | Eugenio Minvielle | Logistic Transport System for Nutritional Substances |
| CN107121407A (en) * | 2017-06-02 | 2017-09-01 | 中国计量大学 | The method that near-infrared spectrum analysis based on PSO RICAELM differentiates Cuiguan pear maturity |
| CN207992039U (en) * | 2018-03-28 | 2018-10-19 | 成都光数云联科技有限公司 | A kind of multi-functional hand-held analytical instrument |
| CN109220226A (en) * | 2018-10-31 | 2019-01-18 | 哈尔滨理工大学 | Fruit automatic recognition classification and the orchard intellectualizing system of picking |
-
2019
- 2019-03-21 CN CN201910217592.5A patent/CN109884088A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160011162A1 (en) * | 2012-04-16 | 2016-01-14 | Eugenio Minvielle | Logistic Transport System for Nutritional Substances |
| CN104515751A (en) * | 2014-12-24 | 2015-04-15 | 河南农业大学 | Flight time spectral fruit maturity non-destructive and rapid detection method and system |
| CN107121407A (en) * | 2017-06-02 | 2017-09-01 | 中国计量大学 | The method that near-infrared spectrum analysis based on PSO RICAELM differentiates Cuiguan pear maturity |
| CN207992039U (en) * | 2018-03-28 | 2018-10-19 | 成都光数云联科技有限公司 | A kind of multi-functional hand-held analytical instrument |
| CN109220226A (en) * | 2018-10-31 | 2019-01-18 | 哈尔滨理工大学 | Fruit automatic recognition classification and the orchard intellectualizing system of picking |
Non-Patent Citations (3)
| Title |
|---|
| KOJI MIZUNO等: "New Applications of Millimeter-Wave Incoherent Imaging", 《IEEE》 * |
| WILSON CASTRO等: "Classification of Cape Gooseberry Fruit According to Its Level of Ripeness Using Machine Learning Techniques and Different Color Spaces", 《IEEE》 * |
| 贺艳楠等: "水果成熟度无损检测技术研究进展", 《北方园艺》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110969090A (en) * | 2019-11-04 | 2020-04-07 | 口碑(上海)信息技术有限公司 | Fruit quality identification method and device based on deep neural network |
| CN111487247A (en) * | 2020-04-28 | 2020-08-04 | 前海国兴(深圳)高科技有限公司 | Fruit maturity AI identification system |
| CN114062505A (en) * | 2022-01-17 | 2022-02-18 | 季华实验室 | Fruit maturity detection method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bhargava et al. | Fruits and vegetables quality evaluation using computer vision: A review | |
| Kodagali et al. | Computer vision and image analysis based techniques for automatic characterization of fruits-a review | |
| Stajnko et al. | Modelling apple fruit yield using image analysis for fruit colour, shape and texture | |
| CN109884088A (en) | Fruit maturity compartment intelligent checking system and detection method | |
| Valdez | Apple defect detection using deep learning based object detection for better post harvest handling | |
| Lin et al. | Detection of strawberry flowers in outdoor field by deep neural network | |
| Pereira et al. | Recent advances in image processing techniques for automated harvesting purposes: A review | |
| CN106706546A (en) | Analysis method for artificial intelligence learning materials on basis of infrared and Raman spectrum data | |
| Lal et al. | Identification and counting of mature apple fruit based on BP feed forward neural network | |
| Gill et al. | A review of automatic fruit classification using soft computing techniques | |
| Yuan et al. | Detection of early bruise in apple using near-infrared camera imaging technology combined with deep learning | |
| US20230214982A1 (en) | Machine learning-based assessment of food item quality | |
| WO2021154624A1 (en) | System and method for performing machine vision recognition of dynamic objects | |
| Caceres-Hernandez et al. | Recent advances in automatic feature detection and classification of fruits including with a special emphasis on Watermelon (Citrillus lanatus): A review | |
| Koç et al. | Classification of some fruits using image processing and machine learning | |
| Aslam et al. | A survey of deep learning methods for fruit and vegetable detection and yield estimation | |
| Ji et al. | Apple color automatic grading method based on machine vision | |
| Kangune et al. | Automated estimation of grape ripeness | |
| Xing et al. | Application of computer vision technology in agricultural products and food inspection | |
| CN109100311B (en) | Strawberry maturity rapid identification method and device | |
| Elaoud et al. | Multi-View-Based Apple Maturity Classification Using Similarity Network Fusion Versus Classical Machine Learning Classifiers | |
| Shah et al. | Crop protection and disease detection using artificial intelligence and computer vision: a comprehensive review | |
| Raut et al. | Assessment of Fruit Maturity using Direct Color Mapping | |
| Rodríguez-Álvarez et al. | Noninvasive system for weight estimation in cactus crops: A YOLOv5-decision tree approach based on interval type-2 fuzzy sets | |
| Beltran et al. | Exploring computer vision, machine learning, and robotics applications in banana grading: A review |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190614 |