CN105117730A - Rosewood discriminating method and system based on hyperspectral data - Google Patents
Rosewood discriminating method and system based on hyperspectral data Download PDFInfo
- Publication number
- CN105117730A CN105117730A CN201510316862.XA CN201510316862A CN105117730A CN 105117730 A CN105117730 A CN 105117730A CN 201510316862 A CN201510316862 A CN 201510316862A CN 105117730 A CN105117730 A CN 105117730A
- Authority
- CN
- China
- Prior art keywords
- data
- classification
- spectroscopic data
- dimension
- redwood
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/192—Recognition using electronic means using simultaneous comparisons or correlations of the image signals with a plurality of references
- G06V30/194—References adjustable by an adaptive method, e.g. learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention proposes a rosewood discriminating method based on hyperspectral data, and the method comprises the following steps: obtaining a training sample and a testing sample, wherein the type is known; classifying the spectral data of all n dimensions according to the training sample, and extracting spectral data with the maximum classification accuracy in m dimensions, wherein m is less than n; obtaining a characteristic point set S with the dimension n, wherein the characteristic point in the set S will serve as a characteristic selection waveband for the later classification, thereby replacing a later classification algorithm; classifying the spectral data of m dimensions according to a K-nearest neighbor classification algorithm, so as to determine the type of rosewood. The method can improve the classification accuracy of data, and reduces the processing time of a data classification algorithm. The invention also provides a rosewood discriminating system based on the hyperspectral data.
Description
Technical field
The present invention relates to hyperspectral data processing technical field, particularly a kind of redwood discrimination method based on high-spectral data and system.
Background technology
High light spectrum image-forming technology can obtain information more more than conventional imaging techniques, the problem utilizing these information can complete traditional imaging data cannot to process.Current most of imaging camera technology all records for scene image based on red, green, blue three look information, although three look sensing imaging techniques meet the imaging demand of human visual system, but from the angle of physical principle, reality scene not only has three look information so simple.Send from light source or light through reflections off objects has abundant wavelength, wherein visible ray covers from 390nm until the extensive region of 780nm, contains a large amount of information.Scene light line spectrum is just referring to the distribution of light light intensity in this section of wavelength coverage, and this spectral information can reflect the natural quality of light source, object and scene, and therefore spectra collection technology has become the effective tool carrying out scientific research and engineer applied.
And redwood classification is a newer field, market is also by virtue of experience knowledge or the check system that damages to its classification, can says that this is a urgently standardization and provide the market segment of Nondestructive Evaluation Techniques.The development of high-spectral data in recent years provides feasibility to the nondestructive test realizing redwood.By setting up the training sample of a known class, utilizing high-spectral data to carry out classification judgement to unknown redwood, utilizing the difference of high-spectral data between different redwood to realize the qualification of redwood classification.
But there is a large amount of data redundancies in high-spectral data, directly the phenomenon that data processing and classification can cause processing speed fast is not carried out to original spectral data, so before classifying to data and analyzing, carry out some pre-service to data and be very important.But traditional data mining and Data Dimensionality Reduction are (as PCA, LDP) method, utilize eigenmatrix to carry out eigentransformation to former data mostly, also be that former data are operated fundamentally, or full spectrum information can be used, and the not obvious contrast relationship one by one of the later matrix of eigentransformation and former data.Therefore, the classification accuracy of current sorting technique is lower.
Summary of the invention
The present invention is intended to solve one of technical matters in above-mentioned correlation technique at least to a certain extent.
For this reason, one object of the present invention is to propose a kind of redwood discrimination method based on high-spectral data, and the method can improve the accuracy rate of Data classification.
Second object of the present invention is to provide a kind of redwood identification system based on high-spectral data.
To achieve these goals, the embodiment of first aspect present invention proposes a kind of redwood discrimination method based on high-spectral data, comprise the following steps: obtain training sample and test sample book, wherein, described training sample and test sample book are all the spectroscopic data of n dimension, and the classification of described training sample is known; The spectroscopic data of the every one dimension in the spectroscopic data tieed up described n according to described training sample is classified, and extracts the highest m of classification accuracy rate and tie up spectroscopic data, and wherein, described m is less than described n; Obtain the unique point S set that dimension is n, the unique point in S set, as the feature selecting wave band of classification later, is classified to the spectroscopic data that described m ties up according to K arest neighbors sorting algorithm, to determine the classification of redwood.
According to the redwood discrimination method based on high-spectral data of the embodiment of the present invention, first training sample and test sample book is obtained, and the spectroscopic data of every one dimension in the spectroscopic data tieed up n according to training sample is classified, and extract the highest m dimension spectroscopic data of classification accuracy rate, then according to K arest neighbors sorting algorithm, the spectroscopic data that m ties up is classified, to determine the classification of redwood.Therefore, the method extracts the maximum dimension of contribution to classification while Data Dimensionality Reduction, and classifies to the maximum dimension of classification contribution in Algorithms of Selecting accordingly, improves the accuracy rate of Data classification.
In addition, the redwood discrimination method based on high-spectral data according to the above embodiment of the present invention can also have following additional technical characteristic:
In some instances, the spectroscopic data of the every one dimension in the spectroscopic data tieed up described n according to described training sample is classified, and extract the highest m dimension spectroscopic data of classification accuracy rate, comprise further: any one-dimensional data in the spectroscopic data tie up described n is classified, and checks the accuracy rate of classification; The spectroscopic data of that the highest for classification accuracy one dimension is joined in data set S; Same classification process is carried out until the spectral coverage in described data set S reaches m to the spectroscopic data of remaining n-1 dimension.
In some instances, in described K arest neighbors sorting algorithm, described K is 1,2 or 3.
Second aspect present invention embodiment still provides a kind of redwood identification system based on high-spectral data, comprise: acquisition module, described acquisition module is for obtaining training sample and test sample book, wherein, described training sample and test sample book are all the spectroscopic data of n dimension, and the classification of described training sample is known; Test module, the spectroscopic data of every one dimension that described test module is used in the spectroscopic data tieed up described n according to described training sample is classified, and extracts the highest m of classification accuracy rate and tie up spectroscopic data, and wherein, described m is less than described n; Sort module, described sort module is used for classifying to the spectroscopic data that described m ties up according to K arest neighbors sorting algorithm, to determine the classification of redwood.
According to the redwood identification system based on high-spectral data of the embodiment of the present invention, first training sample and test sample book is obtained, and the spectroscopic data of every one dimension in the spectroscopic data tieed up n according to training sample is classified, and extract the highest m dimension spectroscopic data of classification accuracy rate, then according to K arest neighbors sorting algorithm, the spectroscopic data that m ties up is classified, to determine the classification of redwood.Therefore, this system extracts the maximum dimension of contribution to classification while Data Dimensionality Reduction, and classifies to the maximum dimension of classification contribution in Algorithms of Selecting accordingly, improves the accuracy rate of Data classification.
In addition, the redwood identification system based on high-spectral data according to the above embodiment of the present invention can also have following additional technical characteristic:
In some instances, any one-dimensional data that described test module is used in the spectroscopic data tieed up described n is classified, and check the accuracy rate of classification, and joined in data set S by the spectroscopic data of that the highest for classification accuracy one dimension, and same classification process is carried out until the spectral coverage in described data set S reaches m to the spectroscopic data of remaining n-1 dimension.
In some instances, in described K arest neighbors sorting algorithm, described K is 1,2 or 3.
Additional aspect of the present invention and advantage will part provide in the following description, and part will become obvious from the following description, or be recognized by practice of the present invention.
Accompanying drawing explanation
Above-mentioned and/or additional aspect of the present invention and advantage will become obvious and easy understand from accompanying drawing below combining to the description of embodiment, wherein:
Fig. 1 is according to an embodiment of the invention based on the process flow diagram of the redwood discrimination method of high-spectral data;
Fig. 2 is the part redwood picture schematic diagram gathered according to an embodiment of the invention;
Fig. 3 is the full modal data curve map of fractional-sample point according to an embodiment of the invention;
Fig. 4 is according to an embodiment of the invention to the structural representation that n dimension data is classified; And
Fig. 5 is according to an embodiment of the invention based on the structured flowchart of the redwood identification system of high-spectral data.
Embodiment
Be described below in detail embodiments of the invention, the example of described embodiment is shown in the drawings, and wherein same or similar label represents same or similar element or has element that is identical or similar functions from start to finish.Being exemplary below by the embodiment be described with reference to the drawings, only for explaining the present invention, and can not limitation of the present invention being interpreted as.
Below in conjunction with accompanying drawing description according to the redwood discrimination method based on high-spectral data of the embodiment of the present invention and system.
Fig. 1 is according to an embodiment of the invention based on the process flow diagram of the redwood discrimination method of high-spectral data.As shown in Figure 1, the method comprises the following steps:
Step S101, obtains training sample and test sample book, and wherein, training sample and test sample book are all the spectroscopic data of n dimension, and the classification of training sample is known.The spectroscopic data of the n dimension of the type redwood to be identified that namely test sample book collects.In other words, be divided into two classes by data, a class is the data group (i.e. training sample) of known class.One class is the data group (i.e. test sample book) of unknown classification.
Step S102, the spectroscopic data of the every one dimension in the spectroscopic data tieed up n according to training sample is classified, and extracts the highest m of classification accuracy rate and tie up spectroscopic data, and wherein, m is less than n.
In one embodiment of the invention, step S102 comprises further:
Step 1: any one-dimensional data in the spectroscopic data of n dimension is classified, and checks the accuracy rate of classification;
Step 2: the spectroscopic data of that the highest for classification accuracy one dimension is joined in data set S;
Step 3: same classification process (namely step 1 is to step 2) is carried out until the spectral coverage in data set S reaches m to the spectroscopic data of remaining n-1 dimension.
Step S103, classifies to the spectroscopic data that m ties up according to K arest neighbors sorting algorithm, to determine the classification of redwood.Wherein, in some instances, the value of the K in K arest neighbors sorting algorithm is 1,2 or 3.
As example particularly, carry out specifically, particularly describing to the method for the above embodiment of the present invention below in conjunction with accompanying drawing 2-4.
Such as, the high-spectral data used in the above embodiment of the present invention is the redwood data as shown in Figure 2 of the spectrometer collection with ASD company, each pixel is a sampled point, each sampled point has the spectroscopic data of a n dimension, the full spectrum information of direct use carries out the classification of redwood not necessarily the most accurately, because the data of complete some dimension of spectrum information are beneficial to redwood classification, some dimension is unfavorable for that redwood is classified, and can be referred to as noise.Such as shown in Fig. 3, the curve of same color represents that these data are from same class redwood, and the shortcoming directly using full spectrum information another one maximum is that processing speed is slow, by the scope of 400nm to the 2500nm of the spectrometer of ASD, data volume is very huge, processing speed is slow, and if want by algorithm application in Embedded equipment later, the basis ensureing classifying quality should be reduced the use of data dimension.
In concrete example, the greedy algorithm of classics is applied in the data selection aspect of redwood data by embodiments of the invention.Suppose the redwood data always total n dimension collected, need to choose the maximum m dimension of classification contribution.In specific implementation process, first set up a training sample set and test sample book collection.Further, any one dimension in n dimension data is used to classify, check the accuracy of classification, choosing that the highest one-dimensional data of classification accuracy rate joins in data set S, so, now S concentrates and only has a spectral coverage, then continue to carry out same process to n-1 remaining dimension, in n-1 residue spectral coverage, choose a spectral coverage makes the classification accuracy rate of the new S collection formed the highest, now S concentrates two spectral coverages, operate successively, when the spectral coverage that S concentrates equals m, greedy algorithm just achieves the feature selecting that the present invention needs.Be exactly equidistant sampling compared to greedy algorithm the simplest Data Dimensionality Reduction choosing method.
Table 1
As concrete example, it is 1nm that table 1 to illustrate in the present invention's concrete example resolution, wavelength coverage is the Data classification result that the full modal data of 400-1700nm obtains when carrying out equidistant (from 5 ~ 50nm not etc.) sampling, the accuracy of classifying as can be seen from Table 1 is unstable, this is because equidistant sampling can not ensure can choose to the maximum dimension of classification contribution, so there is contingency at every turn.But the greedy algorithm used in the present invention ensure that the dimension chosen is maximum to classification contribution at every turn.Such as Fig. 4 is 10 dimensions using greedy algorithm to choose at 400 ~ 1700nm, utilizing these 20 dimensions to carry out Data classification just can ensure about 93% by accuracy, if use simple equidistant sampling to choose 20 dimensions between 400 ~ 1700nm, so sampling interval at least obtains 65nm, and accuracy can not ensure.
In the examples described above, the present invention utilizes greedy algorithm to choose m dimension data, then carries out Data classification to the dimension that the sample (test sample book) of later unknown classification still chooses S concentrated.More specifically, what the algorithm of Data classification adopted is K arest neighbors (k-NearestNeighbor, KNN) sorting algorithm.The thinking of the method is: if the great majority in the sample of K (namely the most contiguous in feature space) the most similar of a sample in feature space belong to some classifications, then this sample also belongs to this classification.In KNN algorithm, selected neighbours are the objects of correctly classification.The method is determining class decision-making only decides to treat the classification belonging to point sample according to the classification of one or several the most contiguous samples.Although KNN method also depends on limit theorem principle, when classification decision-making, only relevant with the adjacent sample of minute quantity.Because KNN method is mainly by the sample of the limited vicinity of surrounding, instead of by differentiating that the method for class field determines generic, therefore for the intersection of class field or overlap more treat a point sample set, KNN method is more applicable compared with other method.In data processing, preferably, choose K=1 (empirical tests, K Selection effect between 1 ~ 3 is better), what the distance in feature space adopted is that Euclidean distance is weighed.
To sum up, according to the redwood discrimination method based on high-spectral data of the embodiment of the present invention, first training sample and test sample book is obtained, and the spectroscopic data of every one dimension in the spectroscopic data tieed up n according to training sample is classified, and extract the highest m dimension spectroscopic data of classification accuracy rate, before classifying to such other, data selection is directly chosen this m dimension data and is carried out Data classification later, then according to K arest neighbors sorting algorithm, the spectroscopic data that m ties up is classified, to determine the classification of redwood.Therefore, the method extracts the maximum dimension of contribution to classification while Data Dimensionality Reduction, and classifies to the maximum dimension of classification contribution in Algorithms of Selecting accordingly, improves the accuracy rate of Data classification.
Further embodiment of the present invention additionally provides a kind of redwood identification system based on high-spectral data.
Fig. 5 is according to an embodiment of the invention based on the structured flowchart of the redwood identification system of high-spectral data.As shown in Figure 5, this system 100 comprises: acquisition module 110, test module 120 and sort module 130.
Wherein, acquisition module 110 is for obtaining training sample and test sample book, and wherein, training sample and test sample book are all the spectroscopic data of n dimension, and the classification of training sample is known.The spectroscopic data of the n dimension of the type redwood to be identified that namely test sample book collects.In other words, be divided into two classes by data, a class is the data group (i.e. training sample) of known class.One class is the data group (i.e. test sample book) of unknown classification.
Test module 120 is classified for the spectroscopic data of the every one dimension in the spectroscopic data tieed up n according to training sample, and extracts the highest m of classification accuracy rate and tie up spectroscopic data, and wherein, m is less than n.
More specifically, in one embodiment of the invention, test module 120 is classified for any one-dimensional data in the spectroscopic data tieed up n, and check the accuracy rate of classification, and joined in data set S by the spectroscopic data of that the highest for classification accuracy one dimension, and same classification process is carried out until the spectral coverage in data set S reaches m to the spectroscopic data of remaining n-1 dimension.
Sort module 130 for classifying to the spectroscopic data that m ties up according to K arest neighbors sorting algorithm, to determine the classification of redwood.Wherein, in some instances, the value of the K in K arest neighbors sorting algorithm is 1,2 or 3.
Exemplary description for this system 100 refers to the above-mentioned description part to method of the present invention, for reducing redundancy, repeats no more herein.
To sum up, according to the redwood identification system based on high-spectral data of the embodiment of the present invention, first training sample and test sample book is obtained, and the spectroscopic data of every one dimension in the spectroscopic data tieed up n according to test sample book is classified, and extract the highest m dimension spectroscopic data of classification accuracy rate, then according to K arest neighbors sorting algorithm, the spectroscopic data that m ties up is classified, to determine the classification of redwood.Therefore, this system extracts the maximum dimension of contribution to classification while Data Dimensionality Reduction, and classifies to the maximum dimension of classification contribution in Algorithms of Selecting accordingly, improves the accuracy rate of Data classification.
In describing the invention, it will be appreciated that, term " " center ", " longitudinal direction ", " transverse direction ", " length ", " width ", " thickness ", " on ", D score, " front ", " afterwards ", " left side ", " right side ", " vertically ", " level ", " top ", " end " " interior ", " outward ", " clockwise ", " counterclockwise ", " axis ", " radial direction ", orientation or the position relationship of the instruction such as " circumference " are based on orientation shown in the drawings or position relationship, only the present invention for convenience of description and simplified characterization, instead of indicate or imply that the device of indication or element must have specific orientation, with specific azimuth configuration and operation, therefore limitation of the present invention can not be interpreted as.
In addition, term " first ", " second " only for describing object, and can not be interpreted as instruction or hint relative importance or imply the quantity indicating indicated technical characteristic.Thus, be limited with " first ", the feature of " second " can express or impliedly comprise at least one this feature.In describing the invention, the implication of " multiple " is at least two, such as two, three etc., unless otherwise expressly limited specifically.
In the present invention, unless otherwise clearly defined and limited, the term such as term " installation ", " being connected ", " connection ", " fixing " should be interpreted broadly, and such as, can be fixedly connected with, also can be removably connect, or integral; Can be mechanical connection, also can be electrical connection; Can be directly be connected, also indirectly can be connected by intermediary, can be the connection of two element internals or the interaction relationship of two elements, unless otherwise clear and definite restriction.For the ordinary skill in the art, above-mentioned term concrete meaning in the present invention can be understood as the case may be.
In the present invention, unless otherwise clearly defined and limited, fisrt feature second feature " on " or D score can be that the first and second features directly contact, or the first and second features are by intermediary indirect contact.And, fisrt feature second feature " on ", " top " and " above " but fisrt feature directly over second feature or oblique upper, or only represent that fisrt feature level height is higher than second feature.Fisrt feature second feature " under ", " below " and " below " can be fisrt feature immediately below second feature or tiltedly below, or only represent that fisrt feature level height is less than second feature.
In the description of this instructions, specific features, structure, material or feature that the description of reference term " embodiment ", " some embodiments ", " example ", " concrete example " or " some examples " etc. means to describe in conjunction with this embodiment or example are contained at least one embodiment of the present invention or example.In this manual, to the schematic representation of above-mentioned term not must for be identical embodiment or example.And the specific features of description, structure, material or feature can combine in one or more embodiment in office or example in an appropriate manner.In addition, when not conflicting, the feature of the different embodiment described in this instructions or example and different embodiment or example can carry out combining and combining by those skilled in the art.
Although illustrate and describe embodiments of the invention above, be understandable that, above-described embodiment is exemplary, can not be interpreted as limitation of the present invention, and those of ordinary skill in the art can change above-described embodiment within the scope of the invention, revises, replace and modification.
Claims (6)
1., based on a redwood discrimination method for high-spectral data, it is characterized in that, comprise the following steps:
Obtain training sample and test sample book, wherein, described training sample and test sample book are all the spectroscopic data of n dimension, and the classification of described training sample is known;
The spectroscopic data of the every one dimension in the spectroscopic data tieed up described n according to described training sample is classified, and extracts the highest m of classification accuracy rate and tie up spectroscopic data, and wherein, described m is less than described n;
According to K arest neighbors sorting algorithm, the spectroscopic data that described m ties up is classified, to determine the classification of redwood.
2. the redwood discrimination method based on high-spectral data according to claim 1, it is characterized in that, the spectroscopic data of the every one dimension in the spectroscopic data tieed up described n according to described training sample is classified, and extracts the highest m of classification accuracy rate and tie up spectroscopic data, comprises further:
Any one-dimensional data in the spectroscopic data tie up described n is classified, and checks the accuracy rate of classification;
The spectroscopic data of that the highest for classification accuracy one dimension is joined in data set S;
Same classification process is carried out until the spectral coverage in described data set S reaches m to the spectroscopic data of remaining n-1 dimension.
3. the redwood discrimination method based on high-spectral data according to claim 1, is characterized in that, in described K arest neighbors sorting algorithm, described K is 1,2 or 3.
4., based on a redwood identification system for high-spectral data, it is characterized in that, comprising:
Acquisition module, described acquisition module is for obtaining training sample and test sample book, and wherein, described training sample and test sample book are all the spectroscopic data of n dimension, and the classification of described training sample is known;
Test module, the spectroscopic data of every one dimension that described test module is used in the spectroscopic data tieed up described n according to described training sample is classified, and extracts the highest m of classification accuracy rate and tie up spectroscopic data, and wherein, described m is less than described n;
Sort module, described sort module is used for classifying to the spectroscopic data that described m ties up according to K arest neighbors sorting algorithm, to determine the classification of redwood.
5. the redwood identification system based on high-spectral data according to claim 4, it is characterized in that, any one-dimensional data that described test module is used in the spectroscopic data tieed up described n is classified, and check the accuracy rate of classification, and joined in data set S by the spectroscopic data of that the highest for classification accuracy one dimension, and same classification process is carried out until the spectral coverage in described data set S reaches m to the spectroscopic data of remaining n-1 dimension.
6. the redwood identification system based on high-spectral data according to claim 4, is characterized in that, in described K arest neighbors sorting algorithm, described K is 1,2 or 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510316862.XA CN105117730A (en) | 2015-06-10 | 2015-06-10 | Rosewood discriminating method and system based on hyperspectral data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510316862.XA CN105117730A (en) | 2015-06-10 | 2015-06-10 | Rosewood discriminating method and system based on hyperspectral data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105117730A true CN105117730A (en) | 2015-12-02 |
Family
ID=54665713
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510316862.XA Pending CN105117730A (en) | 2015-06-10 | 2015-06-10 | Rosewood discriminating method and system based on hyperspectral data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105117730A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105842173A (en) * | 2016-06-06 | 2016-08-10 | 南京大学 | Method for identifying hyperspectral material |
| CN111144244A (en) * | 2019-12-13 | 2020-05-12 | 唐美(北京)文化科技发展有限公司 | Furniture wood type identification method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1936552A (en) * | 2006-10-12 | 2007-03-28 | 中国林业科学研究院木材工业研究所 | Method for identifying red sandalwood by near-infrared ray |
| CN103278467A (en) * | 2013-05-07 | 2013-09-04 | 江苏大学 | Rapid nondestructive high-accuracy method with for identifying abundance degree of nitrogen element in plant leaf |
| CN103714354A (en) * | 2014-01-16 | 2014-04-09 | 西安电子科技大学 | Hyperspectral image wave band selection method based on quantum-behaved particle swarm optimization algorithm |
-
2015
- 2015-06-10 CN CN201510316862.XA patent/CN105117730A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1936552A (en) * | 2006-10-12 | 2007-03-28 | 中国林业科学研究院木材工业研究所 | Method for identifying red sandalwood by near-infrared ray |
| CN103278467A (en) * | 2013-05-07 | 2013-09-04 | 江苏大学 | Rapid nondestructive high-accuracy method with for identifying abundance degree of nitrogen element in plant leaf |
| CN103714354A (en) * | 2014-01-16 | 2014-04-09 | 西安电子科技大学 | Hyperspectral image wave band selection method based on quantum-behaved particle swarm optimization algorithm |
Non-Patent Citations (2)
| Title |
|---|
| 李寒: "基于机器视觉的目标检测在精细农业中的关键技术研究", 《中国博士学位论文全文数据库信息科技辑》 * |
| 臧卓 等: "利用PCA算法进行乔木树种高光谱数据降维与分类", 《测绘科学》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105842173A (en) * | 2016-06-06 | 2016-08-10 | 南京大学 | Method for identifying hyperspectral material |
| CN105842173B (en) * | 2016-06-06 | 2018-05-08 | 南京大学 | A kind of EO-1 hyperion material discrimination method |
| CN111144244A (en) * | 2019-12-13 | 2020-05-12 | 唐美(北京)文化科技发展有限公司 | Furniture wood type identification method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | An automatic method for selecting the parameter of the RBF kernel function to support vector machines | |
| CN109271991B (en) | License plate detection method based on deep learning | |
| Youssef et al. | Fast traffic sign recognition using color segmentation and deep convolutional networks | |
| CN103646250B (en) | Pedestrian monitoring method and device based on distance image head and shoulder features | |
| CN105913095A (en) | Instrument recognition method for transformer substation patrol inspection robot | |
| CN109872335A (en) | A kind of automatic read tablet method and its system for PD-L1 antibody stained slice | |
| US8693772B2 (en) | System and method for digital image signal compression using intrinsic images | |
| CN102324047A (en) | High spectrum image atural object recognition methods based on sparse nuclear coding SKR | |
| CN108416379A (en) | Method and apparatus for handling cervical cell image | |
| DK2447884T3 (en) | A method for the detection and recognition of an object in an image and an apparatus and a computer program therefor | |
| CN105809151A (en) | Image recognition method for state of pressure gauge of fire extinguisher | |
| CN104331710A (en) | On-off state recognition system | |
| CN106529472B (en) | Object detection method and device based on large scale high-resolution high spectrum image | |
| US11881021B2 (en) | Method of providing carbon emission management information, and server and system using the same | |
| Farag | A lightweight vehicle detection and tracking technique for advanced driving assistance systems | |
| Tschentscher et al. | Video-based parking space detection | |
| Li et al. | GIS-based detection of grain boundaries | |
| CN115909059A (en) | Natural resource sample library establishing method and device | |
| CN112560845A (en) | Character recognition method and device, intelligent meal taking cabinet, electronic equipment and storage medium | |
| CN104680185A (en) | Hyperspectral image classification method based on boundary point reclassification | |
| CN105117730A (en) | Rosewood discriminating method and system based on hyperspectral data | |
| CN111931721B (en) | Method and device for detecting color and number of annual inspection label and electronic equipment | |
| CN114332534A (en) | Hyperspectral image small sample classification method | |
| Deepthi et al. | Classification of forensic hyperspectral paper data using hybrid spectral similarity algorithms | |
| CN112101274A (en) | License plate recognition method, device, equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into 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: 20151202 |