CN109765194A - Fructus lycii Production area recognition method based on high light spectrum image-forming technology - Google Patents
Fructus lycii Production area recognition method based on high light spectrum image-forming technology Download PDFInfo
- Publication number
- CN109765194A CN109765194A CN201910082990.0A CN201910082990A CN109765194A CN 109765194 A CN109765194 A CN 109765194A CN 201910082990 A CN201910082990 A CN 201910082990A CN 109765194 A CN109765194 A CN 109765194A
- Authority
- CN
- China
- Prior art keywords
- data
- fructus lycii
- spectral
- area recognition
- production
- 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
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses the fructus lycii Production area recognition methods based on high light spectrum image-forming technology.It includes the following steps: to carry out the fructus lycii subsample of the same kind of different sources spectral scan, the high-spectral data of 1000~2400nm of collection；RAD correction, black and white correction are carried out, handles as relative reflectance data, Threshold segmentation is then carried out to it, delete small area operation；Region of interesting extraction is carried out to data, obtains area-of-interest average light spectrum；Three parts are divided into, training set, verifying collection and test set spectroscopic data are denoted as；Three parts data are handled using ZCA albefaction；The dominant spectral information obtained by training set spectrum and place of production use of information Partial Least Squares Regression are modeled, fructus lycii place of production prediction model is obtained；Utilize spectroscopic data verifying collection and test set spectroscopic data debugging model；Fructus lycii Production area recognition is carried out by the Production area recognition model finally established.The present invention can reduce manual identified cost to Production area recognition, improve the efficiency, accuracy and science of identification.
Description
Technical field
The present invention relates to the fructus lycii Production area recognition methods based on high light spectrum image-forming technology, belong to Materia Medica Identification field.
Background technique
Fructus lycii resource is widely distributed in China, describes according to " Chinese Plants will ", originates in China the north such as northern Hebei, interior
Mongolia, North of Shanxi, North Shaanxi, Gansu, Ningxia, Qinghai, Xinjiang have wild, are cultivated due to fruit medicine, now divided by
Upper provinces and regions have cultivation outer, many provinces and regions of Central China and south also introducing and planting, especially Ningxia and Efficiency in Buildings in Tianjin Area cultivation it is more,
Yield is high.In many ways it is investigated, Ningxia is positioned to the Genuine producing area of medicinal fructus lycii.But since the place of production is more, quality control
The merchandise resources of difficulty, market circulation are unable to ensure, until making fructus lycii market confusion, are adulterated, with other place of production offset roads
Ground producing region product phenomenon takes place frequently.During marketing, experience mirror method for distinguishing is mostly used for the identification of fructus lycii quality,
The method error is larger, subjective, and the kind of fructus lycii is originally a variety of in addition, therefore only only empirically identifying can
Reliability is lower.And chemical analysis, Molecular Detection due to operating method complexity it is time-consuming and laborious, can not popularize.
High light spectrum image-forming technology achieves development at full speed in recent years, is applied only for aerospace field from earliest.Again
It is developed to geological prospecting, ore identification.Followed by again step into agriculture field, the quality of crops is identified, type into
Row is distinguished.In this way, which the every aspect that high light spectrum image-forming technology has been gone into the thick of life, sets foot in seldom in tcm field only.
Summary of the invention
The object of the present invention is to provide a kind of the fructus lycii Production area recognition method based on high light spectrum image-forming technology, skill of the present invention
Art operating process is conducive to the market circulation monitoring of genunie medicinal materials；The cost of manual identified is also reduced, the effect of identification is improved
Rate, accuracy and science.
A kind of fructus lycii Production area recognition method based on high light spectrum image-forming technology provided by the invention, includes the following steps:
1) spectral scan is carried out to the fructus lycii of the same kind of different sources, every time each high-spectral data for collecting 1000~2400nm；
2) the original high-spectral data of sample is subjected to RAD correction；
3) data after RAD correction in step 2) are subjected to black and white correction, handled as relative reflectance data；
4) Threshold segmentation is carried out to the relative reflectance data, deletes small area operation；
5) to treated in step 4), data carry out region of interesting extraction, and it is average that area-of-interest is then calculated
Spectral value；
6) the area-of-interest average light spectrum that the data of spectral scan obtain after step 5) processing is divided into three
Point, it is denoted as training set, verifying collection and test set spectroscopic data；
7) training set, verifying collection and test set spectroscopic data that step 6) processing obtains are handled using ZCA albefaction；
It 8) will be inclined by step 7) treated dominant spectral information that training set spectroscopic data obtains and place of production use of information
Least square regression is modeled, and fructus lycii Production area recognition model is obtained, and utilizes spectroscopic data verifying collection and test light harvesting
Modal data debugging model；Fructus lycii Production area recognition is carried out by the Production area recognition model finally established.
In above-mentioned method, the sample size is more than or equal to 100；
The spectral scan is carried out using hyperspectral imager；
The condition of the spectral scan is as follows: the camera lens of the hyperspectral imager can be 20 at a distance from the fructus lycii
~30cm；Platform movement speed can be 1.5mm/s；When collected spectral region is in 1000~2400nm, the time of integration can be
4500 μ s, frame time can be 46928；
The number of the spectral scan is 3 times.
In above-mentioned method, RAD is corrected to Radiometric calibration radiometric calibration in step 2), is instrument
Included calibration software.
In above-mentioned method, black and white updating formula is as follows in step 3):
The relative reflectance of image after R indicates corrected in formula, IRIndicate the energy value of original image, IWIndicate white
The energy value of plate image, IBIndicate the energy value of blackboard image.
In above-mentioned method, step 4) carries out the Threshold segmentation using MATLAB software, deletes small area operation.
In above-mentioned method, step 5) carries out the region of interesting extraction and averaged spectrum meter using MATLAB software
It calculates.
In step 5) of the present invention, the extraction standard of the region of interesting extraction is according to common sense well known in the art, originally
What invention was specifically extracted is the spectroscopic data of fructus lycii subdivision.
In above-mentioned method, in step 6) area-of-interest average light spectrum be divided into three parts operation it is as follows: utilize
Randperm function in matlab software, the data of spectral scan are divided into three parts, wherein two numbers accordingly 2:1 ratio into
Row training set and verifying collection spectroscopic data are grouped at random, and remaining a data of collecting are as test set.In specific embodiment
By the data being collected into twice in the data of 3 spectral scans with 2:1 ratio be trained collection and verifying collection spectroscopic data with
In addition machine grouping once collects data as test set.
In the present invention, ZCA albefaction is used in step 7) in Pretreated spectra；
The ZCA whitening processing method is defined as rotating back to original on the basis of PCA albefaction multiplied by eigenvectors matrix
Data space obtains the new feature close to initial data.The algorithm of ZCA albefaction realizes following formula:
X is input data in formula, and dimension is m × n, and m representative sample number, n represents input feature vector dimension；Pass through calculating
The covariance matrix of input data XThen it carries out SVD to decompose to obtain left eigenvector matrix U and eigenvectors matrix S, most
New eigenmatrix X is calculated afterwardsnew。
In above-mentioned method, step 8) carries out the Partial Least-Squares Regression Model foundation using MATLAB software.Partially most
Small two multiply return it is as follows as multiple linear regression, canonical correlation analysis and the set of principal component analysis and evolution, thinking: from
Extract component t in independent variable set Xh(h=1,2 ...), it is mutually indepedent between each ingredient.Then set up extract component thWith dependent variable
Regression equation between Y.
In above-mentioned method, the place of production of the fructus lycii of the same kind of different sources can be Xinjiang, Inner Mongol, Gansu, blueness
At least one of sea and Ningxia；
The kind of the fructus lycii of the same kind of different sources is peaceful Qi 7.
The invention has the following advantages that
The present invention uses high light spectrum image-forming technology, is applied to Chinese medicine Production area recognition field, not only contributes to genuine
The market circulation of medicinal material monitors；The cost of manual identified is also reduced, the accuracy and science of identification are improved.The present invention will
EO-1 hyperion identifies applied to traditional Chinese medicinal materials assortment, key point be have found bloom spectral curve and home environment, medicinal material itself character,
Relationship between characteristic component.
Detailed description of the invention
Fig. 1 is that the present invention is based on the flow charts that high light spectrum image-forming spectrometer identifies different sources fructus lycii.
Fig. 2 is single unit system used in the present invention.
Fig. 3 is that fructus lycii puts original image.
Fig. 4 is Threshold segmentation image.
Fig. 5 is the different sources curve of spectrum, and XJ, NX, QH, NM, GS respectively indicate Xinjiang and produce peaceful Qi 7, Ningxia production in Fig. 5
Peaceful Qi 7, Qinghai produce Ningxia 7, Inner Mongol and produce the curve of spectrum in Ningxia 7 and the peaceful Qi in Gansu 7.
Fig. 6 is verifying collection accuracy rate with PLS number of principal components variation diagram.
Specific embodiment
Experimental method used in following embodiments is conventional method unless otherwise specified.
The materials, reagents and the like used in the following examples is commercially available unless otherwise specified.
Embodiment,
According to flow chart shown in FIG. 1 and Fig. 2 shown device is used, it is same to different sources based on high light spectrum image-forming spectrometer
One kind fructus lycii is identified, the specific steps are as follows:
1, spectral scan is carried out to the fructus lycii of the same kind of different sources by hyperspectral imager, when scanning, 1000-
The work of 2400nm camera lens, collects high-spectral data.
The fructus lycii Sample Scan of the same kind of different sources is taken, is tried not when scanning every time beyond camera lens range.Pendulum
When putting fructus lycii, the blank for being used for black and white correction is placed in by the feature of prominent each particle, close pendulum unfolded as far as possible
At the 5cm of sample rear.Wait instrument connection, self-test.Hyperspectral imager sweep parameter, distance of camera lens 30cm are set, and platform moves
Dynamic speed 1.5mm/s, the 1000-2400nm camera lens time of integration are 4500 μ s, frame time 46928.Wherein the time of integration is unit
Enter the number of photons of camera lens in time, in the case where not generating overexposure point, the time of integration is longer, and the quality of image is higher.Frame
Time response be image length-width ratio, numerical value is bigger, then the ratio of scanned object in the horizontal direction is widened, need into
Row is debugged repeatedly, finds optimal proportion, carries out data record.
2, the result after scanning is corrected using the RAD correction software that spectrometer carries, this correction can be eliminated because sweeping
When retouching external environment it is unstable caused by band and noise, keep picture quality more preferable.
3, high-spectral data is imported using matlab software, using black and white updating formula, is by original image data processing
Relative reflectance data.
4, the spectrum picture for carrying out black and white correction is subjected to Threshold segmentation, deletes small area, obtains the interested of image
Region mask, as shown in Figure 4.
5, the mask images determined are carried out to the extraction of area-of-interest using matlab software, are calculated in ROI region
Average light spectrum.
6, data are divided into three parts: training set, verifying collection, test set；The division of spectroscopic data collection, including using with
Machine extracts spectrum without the mode put back to, and firstly generates 1 to nRRandom manifold A, label information is corresponding with random data set, together
When spectroscopic data again it is corresponding with every strip label.Spectral information is distinguished by extracting different label informations according to set proportion
It is subdivided into: training set, verifying collection and test set.Wherein training set carries out the training of model, and verifying collection carries out the adjusting of parameter, surveys
Examination collection carries out the test of model performance.
Point 3 parallel acquisitions, 225 fructus lycii samples, by the data being collected into twice in the data of 3 spectral scans with
2:1 ratio is trained collection and verifying collection spectroscopic data is grouped at random, in addition once collects data as test set.By sample point
For training set, verifying collection, test set, specific classification chart such as the following table 1.
1 data distribution of table
7, the normalization of ZCA vector is carried out to data, filters out main spectral information, concrete principle is as follows:
The ZCA whitening processing method is defined as rotating back to original on the basis of PCA albefaction multiplied by eigenvectors matrix
Data space obtains the new feature close to initial data.The algorithm of ZCA albefaction realizes following formula:
X is input data in formula, and dimension is m × n；M representative sample number, n represent input feature vector dimension；Pass through calculating
The covariance matrix of input data XThen it carries out SVD to decompose to obtain left eigenvector matrix U and eigenvectors matrix S, most
New eigenmatrix X is calculated afterwardsnew。
According to flow chart shown in FIG. 1 and use Fig. 2 shown device, using the method for the present invention identify Xinjiang produce peaceful Qi 7,
Ningxia produces peaceful Qi 7, Qinghai produces Ningxia 7, specific step is as follows for Inner Mongol production Ningxia 7 and the peaceful Qi in Gansu 7:
1, fructus lycii subsample each 90 of peaceful Qi 7 for taking Xinjiang, Ningxia, Gansu, Qinghai, Inner Mongol to produce, are placed into movement
On platform, try not beyond camera lens range.The fructus lycii subsample of the same kind of different sources each 225 are taken, every time 75 points 3
Secondary scanning is tried not when scanning beyond camera lens range, as shown in Figure 3 every time.When putting fructus lycii, the spy of prominent each particle
The blank for being used for black and white correction is placed at the 5cm of sample rear by sign, close pendulum unfolded as far as possible.Wait instrument to connect,
Self-test.Hyperspectral imager sweep parameter, distance of camera lens 30cm, platform movement speed 1.5mm/s are set.400-1000nm is set
The camera lens time of integration is 4350 μ s, frame time 18000.The 1000-2400nm camera lens time of integration is 4500 μ s, frame time 46928.
2, the result after scanning is corrected using the RAD correction software that spectrometer carries, this correction can be eliminated because sweeping
When retouching external environment it is unstable caused by band and noise, keep picture quality more preferable.
3, high-spectral data is imported using matlab software, using black and white updating formula, is by original image data processing
Relative reflectance data.
4, it takes relative reflectance data to carry out Threshold segmentation, deletes small area, obtain the area-of-interest mask of image.
The sampled grey range that image is set as f (x, y) image is mainly by the threshold segmentation method mentioned in this step
[0, L] selects a suitable gray value T between 0 and L, then image can carry out point between background according to gray value T
It cuts, specific formula is as follows:
G (x, the y) image obtained at this time is bianry image, is felt using obtained bianry image to original spectrum picture
The extraction in interest region, as shown in Figure 4.
5, the mask images determined are subjected to area-of-interest (the spectrum number of fructus lycii subdivision using matlab software
According to) extraction, calculate ROI region in average light spectrum, as shown in Figure 5.
6, the data being collected into twice in the data of 3 spectral scans are trained collection and verifying with 2:1 ratio
Collection spectroscopic data is grouped at random, in addition once collects data as test set.
7, ZCA whitening processing is carried out to data, filters out main spectral information.
8, Partial Least Squares Regression is established using training set sample differentiate (PLS-DA) model；Collected using verifying, test set
Inspection result；As a result as shown in Figure 6.
Training set has been pre-processed to be modeled using PLS-DA, and training set Average Accuracy 100% verifies ensemble average accuracy rate
99.29%, test set accuracy rate is 91.04%, and test set accuracy rate standard deviation is 0.0105.
Claims (9)
1. a kind of fructus lycii Production area recognition method based on high light spectrum image-forming technology, includes the following steps: 1) same to different sources
The fructus lycii of one kind carries out spectral scan, every time each high-spectral data for collecting 1000~2400nm；
2) the original high-spectral data of sample is subjected to RAD correction；
3) data after RAD correction in step 2) are subjected to black and white correction, handled as relative reflectance data；
4) Threshold segmentation is carried out to the relative reflectance data, deletes small area operation；
5) to treated in step 4), data carry out region of interesting extraction, and area-of-interest averaged spectrum is then calculated
Value；
6) the area-of-interest average light spectrum that the data of spectral scan obtain after step 5) processing is divided into three parts, remembered
For training set, verifying collection and test set spectroscopic data；
7) training set, verifying collection and test set spectroscopic data that step 6) processing obtains are handled using ZCA albefaction；
It 8) will be partially minimum by step 7) treated dominant spectral information that training set spectroscopic data obtains and place of production use of information
Two, which multiply recurrence, is modeled, and fructus lycii Production area recognition model is obtained, and utilizes spectroscopic data verifying collection and test set spectrum number
According to debugging model；Fructus lycii Production area recognition is carried out by the Production area recognition model finally established.
2. according to the method described in claim 1, it is characterized by: the sample size is more than or equal to 100；
The spectral scan is carried out using hyperspectral imager；
The condition of the spectral scan is as follows: the camera lens of the hyperspectral imager at a distance from the fructus lycii for 20~
30cm；Platform movement speed is 1.5mm/s；When using 1000~2400nm camera lens, the time of integration is 4500 μ s, and frame time is
46928；
The number of the spectral scan is 3 times.
3. method according to claim 1 or 2, it is characterised in that: RAD is corrected to Radiometric in step 2)
Calibration radiometric calibration.
4. method according to any one of claim 1-3, it is characterised in that: in above-mentioned method, black and white in step 3)
Updating formula is as follows:
The relative reflectance of image after R indicates corrected in formula, IRIndicate the energy value of original image, IWIndicate blank figure
The energy value of picture, IBIndicate the energy value of blackboard image.
5. method according to any of claims 1-4, it is characterised in that: in above-mentioned method, step 4) is used
MATLAB software carries out the Threshold segmentation, deletes small area operation.
6. method according to any one of claims 1-5, it is characterised in that: in step 5), using MATLAB software into
The row region of interesting extraction and averaged spectrum calculate.
7. method according to claim 1 to 6, it is characterised in that: area-of-interest averaged spectrum in step 6)
The operation that value is divided into three parts is as follows: utilizing the randperm function in matlab software, the data average mark of spectral scan
At three parts, wherein 2:1 ratio is trained collection to two numbers accordingly and verifying collection spectroscopic data is grouped at random, remaining a collection
Data are as test set.
8. method according to any one of claims 1-7, it is characterised in that: step 8) carries out institute using MATLAB software
State Partial Least-Squares Regression Model foundation.
9. method according to claim 1 to 8, it is characterised in that: the fructus lycii of the same kind of different sources
The place of production of son is at least one of Xinjiang, Inner Mongol, Gansu, Qinghai and Ningxia；
The kind of the fructus lycii of the same kind of different sources is lycium barbarum.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910082990.0A CN109765194B (en) | 2019-01-25 | 2019-01-25 | Wolfberry producing area identification method based on hyperspectral imaging technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910082990.0A CN109765194B (en) | 2019-01-25 | 2019-01-25 | Wolfberry producing area identification method based on hyperspectral imaging technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109765194A true CN109765194A (en) | 2019-05-17 |
| CN109765194B CN109765194B (en) | 2021-05-11 |
Family
ID=66454389
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910082990.0A Active CN109765194B (en) | 2019-01-25 | 2019-01-25 | Wolfberry producing area identification method based on hyperspectral imaging technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109765194B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112085781A (en) * | 2020-09-08 | 2020-12-15 | 中国农业科学院农业资源与农业区划研究所 | Method for extracting winter wheat planting area based on spectrum reconstruction technology |
| CN112861627A (en) * | 2021-01-07 | 2021-05-28 | 中国科学院西安光学精密机械研究所 | Pathogenic bacteria species identification method and system based on microscopic hyperspectral technology |
| CN113505661A (en) * | 2021-06-22 | 2021-10-15 | 中国农业大学 | Method, device, electronic equipment and storage medium for origin identification |
| CN114720420A (en) * | 2022-03-24 | 2022-07-08 | 中国中医科学院中药研究所 | Method and system for identifying production area of Chinese prickly ash based on hyperspectral imaging technology |
| CN114858801A (en) * | 2022-05-25 | 2022-08-05 | 中国科学院西北生态环境资源研究院 | Automatic carbon dust statistical method based on image spectrum principle |
| WO2023142256A1 (en) * | 2022-01-28 | 2023-08-03 | 深圳市现代农业装备研究院 | Early identification and sorting method and system for male sterile seedlings in hybrid seed production, and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103822879A (en) * | 2014-02-24 | 2014-05-28 | 西北农林科技大学 | Nondestructive detection method of swelled kiwi fruits based on hyperspectral imaging technology |
| CN104215584A (en) * | 2014-08-29 | 2014-12-17 | 华南理工大学 | Hyper-spectral image technology-based detection method for distinguishing rice growing areas |
| CN106546541A (en) * | 2016-10-31 | 2017-03-29 | 浙江大学 | A kind of identifying device and method based on EO-1 hyperion transgenic corns seed |
| CN109001218A (en) * | 2018-09-03 | 2018-12-14 | 贵阳学院 | Apple surface defect quick nondestructive recognition methods based on high light spectrum image-forming technology |
-
2019
- 2019-01-25 CN CN201910082990.0A patent/CN109765194B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103822879A (en) * | 2014-02-24 | 2014-05-28 | 西北农林科技大学 | Nondestructive detection method of swelled kiwi fruits based on hyperspectral imaging technology |
| CN104215584A (en) * | 2014-08-29 | 2014-12-17 | 华南理工大学 | Hyper-spectral image technology-based detection method for distinguishing rice growing areas |
| CN106546541A (en) * | 2016-10-31 | 2017-03-29 | 浙江大学 | A kind of identifying device and method based on EO-1 hyperion transgenic corns seed |
| CN109001218A (en) * | 2018-09-03 | 2018-12-14 | 贵阳学院 | Apple surface defect quick nondestructive recognition methods based on high light spectrum image-forming technology |
Non-Patent Citations (2)
| Title |
|---|
| 王庆国等: "基于高光谱图像的玉米种子产地与年份鉴别", 《食品与生物技术学报》 * |
| 王润博: "基于高光谱图像技术的枸杞品质检测方法研究", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112085781A (en) * | 2020-09-08 | 2020-12-15 | 中国农业科学院农业资源与农业区划研究所 | Method for extracting winter wheat planting area based on spectrum reconstruction technology |
| CN112085781B (en) * | 2020-09-08 | 2021-05-11 | 中国农业科学院农业资源与农业区划研究所 | Method for extracting winter wheat planting area based on spectrum reconstruction technology |
| CN112861627A (en) * | 2021-01-07 | 2021-05-28 | 中国科学院西安光学精密机械研究所 | Pathogenic bacteria species identification method and system based on microscopic hyperspectral technology |
| CN113505661A (en) * | 2021-06-22 | 2021-10-15 | 中国农业大学 | Method, device, electronic equipment and storage medium for origin identification |
| WO2023142256A1 (en) * | 2022-01-28 | 2023-08-03 | 深圳市现代农业装备研究院 | Early identification and sorting method and system for male sterile seedlings in hybrid seed production, and storage medium |
| CN114720420A (en) * | 2022-03-24 | 2022-07-08 | 中国中医科学院中药研究所 | Method and system for identifying production area of Chinese prickly ash based on hyperspectral imaging technology |
| CN114858801A (en) * | 2022-05-25 | 2022-08-05 | 中国科学院西北生态环境资源研究院 | Automatic carbon dust statistical method based on image spectrum principle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109765194B (en) | 2021-05-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109765194A (en) | Fructus lycii Production area recognition method based on high light spectrum image-forming technology | |
| CN101881726B (en) | Nondestructive detection method for comprehensive character living bodies of plant seedlings | |
| Pang et al. | Improved crop row detection with deep neural network for early-season maize stand count in UAV imagery | |
| CN109540831A (en) | Fructus lycii variety ecotype method based on high light spectrum image-forming technology | |
| Wang et al. | Landscape-level vegetation classification and fractional woody and herbaceous vegetation cover estimation over the dryland ecosystems by unmanned aerial vehicle platform | |
| Liu et al. | Estimating potato above-ground biomass by using integrated unmanned aerial system-based optical, structural, and textural canopy measurements | |
| Liang et al. | Improved estimation of aboveground biomass in rubber plantations by fusing spectral and textural information from UAV-based RGB imagery | |
| CN109522929A (en) | It is a kind of based on multi-source image fusion natural environment under wheat weeds recognition methods | |
| CN104697943A (en) | Rape water stress nondestructive testing method | |
| CN102855485B (en) | The automatic testing method of one grow wheat heading | |
| Fang et al. | High-throughput volumetric reconstruction for 3D wheat plant architecture studies | |
| Chen et al. | Diagnosing of rice nitrogen stress based on static scanning technology and image information extraction | |
| CN109827957A (en) | A kind of rice leaf SPAD value estimating and measuring method based on computer vision and system | |
| CN111751376A (en) | Rice nitrogen nutrition estimation method based on canopy image feature derivation | |
| CN110082310A (en) | A kind of near infrared band EO-1 hyperion diagnostic method of rubber tree LTN content | |
| CN106568730B (en) | A kind of rice yin-yang leaf fringe recognition methods based on Hyperspectral imaging near the ground | |
| CN116862456A (en) | Traditional Chinese medicine production monitoring control system and method based on image processing | |
| CN102788796B (en) | Nutrient diagnosis device and nutrient diagnosis method for nitrogen of crops based on multi-information integration of high spectral images and fluorescent images | |
| Cheng et al. | Hyperspectral and imagery integrated analysis for vegetable seed vigor detection | |
| CN105527241A (en) | Non-destructive method for detecting authenticity of raw cordyceps sinensis | |
| CN105548037A (en) | Method for nondestructively detecting raw traditional Chinese medicinal materials | |
| CN105606544A (en) | Nondestructive detection method of insect bodies of Cordyceps sinensis | |
| CN116863341A (en) | Crop classification and identification method and system based on time sequence satellite remote sensing image | |
| Wang et al. | Characterization and classification of urban weed species in northeast China using terrestrial hyperspectral images | |
| CN105466865A (en) | Nondestructive ophiocordyceps sinensis stroma adulteration detection 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |