CN108918499A - The method of Raman baseline drift is removed in Raman map - Google Patents
The method of Raman baseline drift is removed in Raman map Download PDFInfo
- Publication number
- CN108918499A CN108918499A CN201810683051.7A CN201810683051A CN108918499A CN 108918499 A CN108918499 A CN 108918499A CN 201810683051 A CN201810683051 A CN 201810683051A CN 108918499 A CN108918499 A CN 108918499A
- Authority
- CN
- China
- Prior art keywords
- raman
- trough
- abscissa
- baseline
- map
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/65—Raman scattering
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2201/00—Features of devices classified in G01N21/00
- G01N2201/12—Circuits of general importance; Signal processing
- G01N2201/127—Calibration; base line adjustment; drift compensation
Landscapes
- Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The invention discloses the methods that Raman baseline drift is removed in Raman map, including：The first step：Original Raman map is carried out smoothly, it is bigger than two boundary values by median, smoothed out Raman wave crest is acquired, it is smaller than two boundary values by median, smoothed out trough is acquired, smoothed out Raman map is obtained；Second step：The linear fit that sliding window formula is carried out to smoothed out Raman region, the Raman baseline after being fitted；Third step：The fluorescence packets in Raman baseline after removal fitting；4th step：Original Raman peaks map subtracts the Raman baseline of the removal fluorescence packets of step 3 acquisition to get the Raman map of final removal Raman baseline drift is arrived.This method can guarantee the stability in correction course, accelerate the speed of baseline correction.
Description
Technical field
The present invention relates to remove to draw in the processing method of the Raman baseline drift in optical field, in particular to Raman map
The method of graceful baseline drift.
Background technique
Raman spectrum plays an important role in bioanalysis and detection field.In recent years, Raman spectrum is employed for cell
In classification and analysis.This generally requires more accurate data to carry out algorithm analysis.In Raman detection, background fluorescence is one
The most common background interference light of kind, background fluorescence appearance are usually more much broader than Raman peaks, it appears that are like that Raman spectrum is slow
Slowly the baseline changed.Background fluorescence will lead to Raman baseline and drift about, so as to cause the inaccurate of Raman data.
Common baseline correction method has derivation, fitting of a polynomial, wavelet transformation etc..Savitzky-Golay derivative light
Spectrum is more commonly used derivative spectrum, it is to be fitted a multinomial according to a series of given spectrum points, calculates data segment
The derivative of midpoint polynomial fitting.However derivative spectrum can reduce signal-to-noise ratio or original spectrum is caused to become for baseline correction
Shape.The baseline correction method of fitting of a polynomial is the selection fitting data point from spectrum, and definition is suitable the characteristics of further according to spectrum
Fitting order, then according to the best approximating function of principle of least square digital simulation data point.However it is fitted the difficulty of order
With determination, it will increase operation time, while being fitted baseline to vibrate because of the change of fitting order.Wavelet transformation exists
There are the residuals that in problem, will lead to fluorescence background for fluorescence packets.
Baseline correction of the Wavelet transformation in Raman at present be it is universal, by Wavelet transformation carry out low-and high-frequency separation,
Principle of least square method fitting data point is used simultaneously, removes interference of the fluorescence background to spectrum.The use of this method is special
In a kind of Raman spectra pretreatment method of benefit 201310094703.0, also has mentioned and use.
Although current wavelet transformation removal Raman baseline has certain effect on the needle position misalignment caused by fluorescence,
There are problems in the processing of fluorescence packets, cannot remove fluorescence packets well.
Summary of the invention
The object of the present invention is to provide the method for removing Raman baseline drift in Raman map, this method can guarantee to correct
Stability in the process accelerates the speed of baseline correction.
Above-mentioned purpose of the invention is realized by following technical solution：The side of Raman baseline drift is removed in Raman map
Method, which is characterized in that this method comprises the following steps：
The first step：Original Raman map is carried out smoothly, it is bigger than two boundary values by median, acquire smoothed out Raman wave
Peak, it is smaller than two boundary values by median, smoothed out trough is acquired, smoothed out Raman map is obtained；
Second step：The linear fit that sliding window formula is carried out to smoothed out Raman region, the Raman base after being fitted
Line；
Third step：The fluorescence packets in Raman baseline after removal fitting；
4th step：Original Raman peaks map subtracts the Raman baseline of the removal fluorescence packets of step 3 acquisition to get to finally
Removal Raman baseline drift Raman map.
In the present invention, detailed process is as follows for the second step：Selecting abscissa in smoothed out Raman map is the wave of X
Paddy, the abscissa of next trough are X1, and the abscissa of next trough is X2, and the abscissa of next trough is X3, is acquired
The cosine value for the angle between the negative semiaxis of straight line and y-axis that the trough that abscissa is X is connected with the trough that abscissa is X1 is denoted as
A1 acquires the folder cosine of an angle between the trough straight line being connected with the trough that abscissa is X2 that abscissa is X and the negative semiaxis of y-axis
Value is denoted as a2, compares the size of a1 and a2, if a1>A2, the then trough for selecting abscissa to be X1 for the trough and abscissa of X carry out
Linear fit；If a1<A2 continues downwards relatively, to acquire the straight line that the trough that abscissa is X is connected with the trough that abscissa is X3
The cosine value of angle between the negative semiaxis of y-axis is denoted as a3, if a2>A3, does not continue to compare, then selects abscissa for the wave of X
The trough that paddy and abscissa are X2 carries out linear fit；If a2<A3 then continues downwards relatively, until there is a (N)>A (N+1), N
N-th trough under the trough for being X for abscissa, the then trough for selecting abscissa to be X (N) for the trough and abscissa of X carry out
Linear fit, the Raman baseline after being fitted continue to repeat the above steps with abscissa X (N) for trough starting point, Zhi Daosuo
There is trough to be all traversed, obtain the Raman baseline being finally fitted, linear fit is all made of the approximating method of existing least square method.
Detailed process is as follows for the third step：There may be fluorescence packets for the trough in Raman baseline after fitting, need
Removal, is arranged the entire value of a fluorescence coefficient, if the fluorescence coefficient of trough is more than this entire value, which can be considered as glimmering
Light packet,
When fluorescence coefficient is greater than entire value, the Raman baseline after fitting is pulled up to remove fluorescence packets, and it is glimmering to obtain removal
The Raman baseline of light packet.
Method of the invention is not in the feelings for being fitted excessive or poor fitting during correcting Raman baseline drift
Condition ensure that the stability in correction course, because being linear fit, not need the order for determining fitting, therefore significantly
The speed for accelerating baseline correction.
Detailed description of the invention
The present invention is described in further details in the following with reference to the drawings and specific embodiments.
Fig. 1 is to slide in the method for remove in Raman map of the present invention Raman baseline drift to smoothed out Raman region
The linear fit of dynamic window type, the diagram of the Raman baseline after being fitted are fitted baseline with it for smoothed out Raman map
Composite diagram, wherein full curve be smoothed out Raman map, dash-dotted curve be smoothed out Raman map fitting base
Line finds out the position for being appropriate for two troughs of linear fit by comparing the cosine value of two troughs and the negative semiaxis line of y
It sets；
Figure 1A is smoothed out Raman map in Fig. 1；
Figure 1B is the fitting baseline of smoothed out Raman map in Fig. 1；
Fig. 2 is in the Raman baseline in the method for remove in Raman map of the present invention Raman baseline drift after removal fitting
The diagram of fluorescence packets；To there is the fitting baseline of the smoothed out Raman map of fluorescence packets to be fitted with it composite diagram of baseline, wherein
Full curve is the smoothed out Raman map for having fluorescence packets, dash-dotted curve be have fluorescence packets smoothed out Raman map it is quasi-
Baseline is closed, which is the fitting baseline effects figure for having the smoothed out Raman map of fluorescence, it can be seen from the figure that smooth post-tensioning
The fluorescence packets of graceful map are fitted well, and the case where over-fitting and poor fitting do not occur in Raman peaks；
Fig. 2A is the smoothed out Raman map for having fluorescence packets in Fig. 2；
Fig. 2 B is the fitting baseline for having the smoothed out Raman map of fluorescence packets in Fig. 2；
Fig. 3 is the Raman baseline that removal fluorescence packets are obtained in the method for remove in Raman map of the present invention Raman baseline drift
Diagram；To there is the Raman map of fluorescence packets to be fitted with it composite diagram of baseline, wherein full curve is the Raman for having fluorescence packets
Map, dash-dotted curve are the fitting baseline for having the Raman map of fluorescence packets, and the baseline of acquisition and Raman map are mapped, from
As can be seen that the fluorescence packets of Raman map are fitted well in figure, and there is not over-fitting and poor fitting yet in Raman peaks
Situation；
Fig. 3 A is the Raman map for having fluorescence packets in Fig. 3；
Fig. 3 B is the fitting baseline for having the Raman map of fluorescence packets in Fig. 3；
Fig. 4 is the Raman map that baseline correction is later in the method for remove in Raman map of the present invention Raman baseline drift,
By Raman map subtract it is smooth after Raman map fitting baseline it can be seen from the figure that fluorescence packets are removed, and Raman peaks
There is not substantial loss in intensity.
Specific embodiment
The method that Raman baseline drift is removed in Raman map of the present invention, includes the following steps：
The first step：Original Raman map is carried out smoothly, it is bigger than two boundary values by median, acquire smoothed out Raman wave
Peak, it is smaller than two boundary values by median, smoothed out trough is acquired, smoothed out Raman map is obtained；
Second step：The linear fit that sliding window formula is carried out to smoothed out Raman region, the Raman base after being fitted
Line；
Selecting abscissa in smoothed out Raman map is the trough of X, and the abscissa of next trough is X1, next wave
The abscissa of paddy is X2, and the abscissa of next trough is X3, acquires the trough phase that trough that abscissa is X is X1 with abscissa
The cosine value of angle between the negative semiaxis of straight line and y-axis even is denoted as a1, and acquiring trough that abscissa is X and abscissa is X2's
The cosine value of angle between trough connected straight line and the negative semiaxis of y-axis is denoted as a2, compares the size of a1 and a2, if a1>A2, then
The trough for selecting abscissa to be X1 for the trough and abscissa of X carries out linear fit；If a1<A2 continues downwards relatively, to acquire cross
The cosine value for the angle between the negative semiaxis of straight line and y-axis that the trough that coordinate is X is connected with the trough that abscissa is X3 is denoted as a3,
If a2>A3, does not continue to compare, then the trough for selecting abscissa to be X2 for the trough and abscissa of X carries out linear fit；If a2
<A3 then continues downwards relatively, until there is a (N)>A (N+1), N are the n-th trough under the trough that abscissa is X, then select
The trough that the trough and abscissa that abscissa is X are X (N) carries out linear fit, and the Raman baseline after being fitted continues with cross
Coordinate X (N) is trough starting point, is repeated the above steps, and until all troughs are all traversed, obtains the Raman baseline being finally fitted,
Linear fit is all made of the approximating method of existing least square method；
As shown in Fig. 1, Figure 1A, Figure 1B, by taking the trough at (x=1651) as an example, we pass through the Linear Quasi of two troughs
Cooperation is Raman baseline, and next trough of the Raman is (x=1607), but this trough and is unsuitable as linear fit
Trough, because of the case where baseline at this moment drawn will appear poor fitting, and next trough (x=1554) is suitable as Linear Quasi
The trough of conjunction, next trough (x=1522) is not appropriate for the trough as linear fit, because Raman baseline can Raman
Above map, and there is the over-fitting of Raman baseline.
Judge (x=1651) and which trough carries out linear fit, acquires the straight line and y of (x=1651) Yu (x=1607)
The Cos note a1 (a1 is -0.906) of negative semiaxis, acquires the straight line of (x=1651) and (x=1554) and the Cos note a2 of the negative semiaxis of y
(a2 is -0.293), compares the size of a1 and a2, a1<A2 continues downwards relatively, to acquire the straight of (x=1651) and (x=1522)
The Cos of line and the negative semiaxis of y is denoted as a3 (a3 is -0.515), because of a2>A3, then not continuing to compare at this time, we select (x
=1651) trough and at (x=1544) carries out linear fit.
The benefit done so：It can prevent Raman from poor fitting occur, while over-fitting will not occur because of floating on baseline again
The case where.It can be good at realizing and removing the baseline drift of Raman.
Third step：The fluorescence packets in Raman baseline after removal fitting；
The trough in Raman baseline after fitting needs to remove there may be fluorescence packets.
As shown in Fig. 2, Fig. 2A, Fig. 2 B, (x=1242) trough and (x=1406) trough for being obtained after Raman subregion
There are apparent Raman fluorescence packets in region, and such fluorescence packets are practical and Raman peaks are not present, but if cannot pass through baseline
If removal, subsequent data analysis is influenced whether.
The entire value of one fluorescence coefficient, numerical value 1.5 are set, it, should if the fluorescence coefficient of trough is more than this entire value
Trough can be considered as fluorescence packets, and height of foundation is 0.03 in the formula of dissymmetry coefficient.
When fluorescence coefficient is greater than entire value, the Raman baseline after fitting is pulled up to remove fluorescence packets, and it is glimmering to obtain removal
The Raman baseline of light packet, as shown in Fig. 3, Fig. 3 A, Fig. 3 B；
4th step：Original Raman peaks map subtracts the Raman baseline of the removal fluorescence packets of step 3 acquisition to get to finally
Removal Raman baseline drift Raman map, the Raman map after removal is as shown in Figure 4.
The above embodiment of the present invention is not limiting the scope of the present invention, and embodiments of the present invention are not limited to
This, all this kind above content according to the present invention is not departing from this according to the ordinary technical knowledge and customary means of this field
Modification, replacement or change under the premise of inventing above-mentioned basic fundamental thought, to other diversified forms that above structure of the present invention is made
More, it should all fall within the scope and spirit of the invention.
Claims (3)
1. removing the method for Raman baseline drift in Raman map, which is characterized in that this method comprises the following steps：
The first step：Original Raman map is carried out smoothly, it is bigger than two boundary values by median, smoothed out Raman wave crest is acquired,
It is smaller than two boundary values by median, smoothed out trough is acquired, smoothed out Raman map is obtained；
Second step：The linear fit that sliding window formula is carried out to smoothed out Raman region, the Raman baseline after being fitted；
Third step：The fluorescence packets in Raman baseline after removal fitting；
4th step：The Raman baseline that original Raman peaks map subtracts the removal fluorescence packets of step 3 acquisition goes to get to final
Except the Raman map of Raman baseline drift.
2. removing the method for Raman baseline drift in Raman map according to claim 1, which is characterized in that described second
Detailed process is as follows for step：Select abscissa in smoothed out Raman map for the trough of X, the abscissa of next trough is
X1, the abscissa of next trough are X2, and the abscissa of next trough is X3, acquire trough and abscissa that abscissa is X
The cosine value of angle between the trough of X1 connected straight line and the negative semiaxis of y-axis is denoted as a1, acquire trough that abscissa is X with
The cosine value of angle between the trough that abscissa is X2 connected straight line and the negative semiaxis of y-axis is denoted as a2, compares the big of a1 and a2
It is small, if a1>A2, the then trough for selecting abscissa to be X1 for the trough and abscissa of X carry out linear fit；If a1<A2, continue to
Lower comparison acquires the angle between the trough straight line being connected with the trough that abscissa is X3 that abscissa is X and the negative semiaxis of y-axis
Cosine value is denoted as a3, if a2>A3, does not continue to compare, then the trough for selecting abscissa to be X2 for the trough and abscissa of X carries out
Linear fit；If a2<A3 then continues downwards relatively, until there is a (N)>A (N+1), N are the N under the trough that abscissa is X
A trough, the then trough for selecting abscissa to be X (N) for the trough and abscissa of X carry out linear fit, the Raman after being fitted
Baseline continues to repeat the above steps with abscissa X (N) for trough starting point, and until all troughs are all traversed, acquisition is final quasi-
The Raman baseline of conjunction, linear fit are all made of the approximating method of existing least square method.
3. removing the method for Raman baseline drift in Raman map according to claim 2, which is characterized in that the third
Detailed process is as follows for step：The trough in Raman baseline after fitting needs to remove there may be fluorescence packets, and a fluorescence is arranged
The entire value of coefficient, if the fluorescence coefficient of trough is more than this entire value, which can be considered as fluorescence packets,
When fluorescence coefficient is greater than entire value, the Raman baseline after fitting is pulled up to remove fluorescence packets, obtains removal fluorescence packets
Raman baseline.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810683051.7A CN108918499B (en) | 2018-06-28 | 2018-06-28 | The method of Raman baseline drift is removed in Raman map |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810683051.7A CN108918499B (en) | 2018-06-28 | 2018-06-28 | The method of Raman baseline drift is removed in Raman map |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108918499A true CN108918499A (en) | 2018-11-30 |
| CN108918499B CN108918499B (en) | 2019-08-02 |
Family
ID=64422893
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810683051.7A Active CN108918499B (en) | 2018-06-28 | 2018-06-28 | The method of Raman baseline drift is removed in Raman map |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108918499B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993155A (en) * | 2019-04-23 | 2019-07-09 | 北京理工大学 | For the characteristic peak extracting method of low signal-to-noise ratio uv raman spectroscopy |
| CN110108694A (en) * | 2019-05-17 | 2019-08-09 | 广西科技大学 | The method for improving of Raman spectrometer wave number migration measurement accuracy |
| CN113109318A (en) * | 2021-03-26 | 2021-07-13 | 中国科学院西安光学精密机械研究所 | Raman spectrum quantitative analysis method and system based on spectral peak height direct extraction |
| CN114993891A (en) * | 2022-07-14 | 2022-09-02 | 长光辰英(杭州)科学仪器有限公司 | Raman particulate matter detection method based on cosine similarity |
| CN116030032A (en) * | 2023-02-16 | 2023-04-28 | 四川省肿瘤医院 | Breast cancer analysis equipment, system and storage medium based on Raman spectrum data |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002003857A2 (en) * | 2000-07-11 | 2002-01-17 | Lightouch Medical, Inc. | Method of tissue modulation for noninvasive measurement of an analyte |
| CN103529012A (en) * | 2013-09-18 | 2014-01-22 | 武汉钢铁(集团)公司 | Raman spectrum quantitative determination method applicable to carbon sources in blast furnace gas dust |
| CN105806824A (en) * | 2016-03-22 | 2016-07-27 | 武汉轻工大学 | Quick edible oil variety identification method based on multisource spectral feature fusion |
| CN107345911A (en) * | 2017-06-06 | 2017-11-14 | 浙江大学 | A kind of method of histamine in SERS qualitative and quantitative analysis rice fish tissue |
-
2018
- 2018-06-28 CN CN201810683051.7A patent/CN108918499B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002003857A2 (en) * | 2000-07-11 | 2002-01-17 | Lightouch Medical, Inc. | Method of tissue modulation for noninvasive measurement of an analyte |
| CN103529012A (en) * | 2013-09-18 | 2014-01-22 | 武汉钢铁(集团)公司 | Raman spectrum quantitative determination method applicable to carbon sources in blast furnace gas dust |
| CN105806824A (en) * | 2016-03-22 | 2016-07-27 | 武汉轻工大学 | Quick edible oil variety identification method based on multisource spectral feature fusion |
| CN107345911A (en) * | 2017-06-06 | 2017-11-14 | 浙江大学 | A kind of method of histamine in SERS qualitative and quantitative analysis rice fish tissue |
Non-Patent Citations (1)
| Title |
|---|
| 覃赵军等: "《分段式线性拟合校正拉曼光谱基线漂移》", 《光谱学与光谱分析》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993155A (en) * | 2019-04-23 | 2019-07-09 | 北京理工大学 | For the characteristic peak extracting method of low signal-to-noise ratio uv raman spectroscopy |
| CN109993155B (en) * | 2019-04-23 | 2021-01-01 | 北京理工大学 | Characteristic peak extraction method for low signal-to-noise ratio ultraviolet Raman spectrum |
| CN110108694A (en) * | 2019-05-17 | 2019-08-09 | 广西科技大学 | The method for improving of Raman spectrometer wave number migration measurement accuracy |
| CN110108694B (en) * | 2019-05-17 | 2021-11-02 | 广西科技大学 | Method for improving wave number migration measurement precision of Raman spectrometer |
| CN113109318A (en) * | 2021-03-26 | 2021-07-13 | 中国科学院西安光学精密机械研究所 | Raman spectrum quantitative analysis method and system based on spectral peak height direct extraction |
| CN114993891A (en) * | 2022-07-14 | 2022-09-02 | 长光辰英(杭州)科学仪器有限公司 | Raman particulate matter detection method based on cosine similarity |
| CN114993891B (en) * | 2022-07-14 | 2024-04-19 | 长光辰英(杭州)科学仪器有限公司 | Particle Raman detection method based on cosine similarity |
| CN116030032A (en) * | 2023-02-16 | 2023-04-28 | 四川省肿瘤医院 | Breast cancer analysis equipment, system and storage medium based on Raman spectrum data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108918499B (en) | 2019-08-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108918499B (en) | The method of Raman baseline drift is removed in Raman map | |
| CN109696788B (en) | Quick automatic focusing method based on display panel | |
| CN110473189B (en) | Text image definition judging method and system | |
| CN115601367A (en) | LED lamp wick defect detection method | |
| CN105117725B (en) | A kind of car plate self-correction recognition methods based on car plate confidence level | |
| CN103886553B (en) | Method and system for non-local average value denoising of image | |
| CN108896278A (en) | A kind of optical filter silk-screen defect inspection method, device and terminal device | |
| CN108711158A (en) | The image of gauge with pointer recognition methods based on contour fitting and radially divided | |
| CN110111301A (en) | Metal based on frequency-domain transform aoxidizes surface defect visible detection method | |
| CN107369176B (en) | System and method for detecting oxidation area of flexible IC substrate | |
| CN105139391B (en) | A kind of haze weather traffic image edge detection method | |
| CN107730521B (en) | Method for rapidly detecting ridge type edge in image | |
| CN110033439B (en) | Visual detection method for material blockage of belt conveyor in wire making process | |
| WO2023070593A1 (en) | Line width measurement method and apparatus, computing processing device, computer program, and computer readable medium | |
| CN102663384A (en) | Curve identification method based on Bezier control point searching and apparatus thereof | |
| CN110533616A (en) | A kind of method of image sky areas segmentation | |
| CN107038444A (en) | A kind of image-recognizing method of pointer dial plate | |
| CN108918093A (en) | A kind of optical filter mirror defects detection method, device and terminal device | |
| CN117541582A (en) | IGBT insulation quality detection method for high-frequency converter | |
| CN112750089B (en) | Optical remote sensing image defogging method based on local block maximum and minimum pixel prior | |
| CN114140416A (en) | Glass edge detection method and system based on machine vision | |
| CN117689574A (en) | Medical image processing method for tumor radio frequency ablation diagnosis and treatment positioning | |
| CN110489810B (en) | Automatic trend extraction method based on data blocks | |
| CN108492345B (en) | Data block dividing method based on scale transformation | |
| CN109146900A (en) | Improve extracting method in the region Blob shown applied to line holographic projections |
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 |