CN112433339B - Microscope fine focusing method based on random forest - Google Patents
Microscope fine focusing method based on random forest Download PDFInfo
- Publication number
- CN112433339B CN112433339B CN202011438427.1A CN202011438427A CN112433339B CN 112433339 B CN112433339 B CN 112433339B CN 202011438427 A CN202011438427 A CN 202011438427A CN 112433339 B CN112433339 B CN 112433339B
- Authority
- CN
- China
- Prior art keywords
- image
- focusing
- random forest
- function
- calculating
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B7/00—Mountings, adjusting means, or light-tight connections, for optical elements
- G02B7/28—Systems for automatic generation of focusing signals
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B21/00—Microscopes
- G02B21/24—Base structure
- G02B21/241—Devices for focusing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
Abstract
The invention discloses a microscope fine focusing method and a microscope fine focusing system based on random forests, wherein the method comprises the following steps: 1) focusing for the first time: the lens is at any position, an image Img1 is collected, the lens advances by a distance A, and an image Img2 is collected again; 2) calculating nine definition evaluation function characteristic values of the images Img1 and Img 2; 3) classifying the random forest classification model to obtain defocusing distances L1 and L2; if L2 is 0, finishing focusing, otherwise, carrying out secondary focusing; 4) and (3) focusing for the second time: calculating an energy gradient function; 5) collecting an image; 6) and calculating the peak value of the energy gradient function, finding a focal plane and finishing focusing. The invention has the advantages that the number of images needing to be collected is less, the time for searching the focal plane is shorter, and the focusing efficiency can be effectively improved; the invention combines nine definition evaluation functions to judge the focal length, can reduce the probability of falling into a local extreme value, and can improve the robustness and the precision of the focusing method.
Description
Technical Field
The invention relates to the technical field of microscopic imaging, in particular to a microscope fine focusing method based on random forests.
Background
The auto-focusing technique is a key technique of various automatic imaging systems, and is widely applied to a large number of imaging systems such as an endoscope, an electron microscope, a laser scanning confocal microscope and the like. At present, an automatic focusing method is mainly based on an image processing theory, firstly, definition evaluation function calculation is carried out on an acquired image, image definition quality evaluation is obtained through calculation, then a certain search strategy is combined to drive a hardware adjusting system to find a peak value, and common search strategies include a Fibonacci search method, a curve fitting method, a hill climbing search method and the like. These conventional auto-focusing methods basically use a single sharpness evaluation function, which results in that the search tends to fall into local extrema. In addition, in order to find an extreme value of the definition function evaluation, the objective lens needs to be moved to acquire a large number of pictures, so that the timeliness of the whole focusing process is reduced.
A more reliable solution is now needed.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a microscope fine focusing method based on random forest aiming at the defects in the prior art.
In order to solve the technical problems, the invention adopts the technical scheme that: a microscope fine focusing method based on random forests comprises the following steps:
1) focusing for the first time: initially, the lens to be focused is at any position, an image Img1 is collected, then the lens is advanced by a distance A, and an image Img2 is collected again;
2) firstly, filtering difference processing is carried out on the image Img1 and the image Img2, then nine definition evaluation function characteristic values of the image Img1 are calculated to be used as a sample 1, and nine definition evaluation function characteristic values of the image Img2 are calculated to be used as a sample 2;
3) classifying the sample 1 and the sample 2 by using a random forest classification model obtained by pre-training so as to obtain a defocus distance L1 of the image Img1 and a defocus distance L2 of the image Img 2;
if L2 is 0, the current image is located in the focal plane, focusing is completed, and if L2 is not 0, the next step is carried out, and secondary focusing is carried out;
4) and (3) focusing for the second time: calculating energy gradient functions of Img1 and Img2, which are respectively marked as EOG1 and EOG 2;
if EOG1-EOG2 is less than or equal to 0, advancing the lens by a distance L2, otherwise, retreating the lens by a distance L2;
5) acquiring images at a current position, and then respectively acquiring the images at positions which are +/-A, + -2A, + -3A, + -2N and A away from the current position, wherein N is a positive integer;
6) and calculating the energy gradient function of each of the acquired 1+2N images, taking the position corresponding to the image with the maximum energy gradient function as a focal plane, and moving the lens to the focal plane to finish focusing.
Preferably, a ═ 2 μm.
Preferably, a ═ 1 μm.
Preferably, wherein N ═ 5.
Preferably, in the step 3), the method for training the random forest model in advance includes the following steps:
3-1) respectively collecting a plurality of images with out-of-focus distances of 1T, 2T, 3T, M T by taking the step length as T, and establishing a training data set by taking the out-of-focus distances as labels of the images;
3-2) carrying out image preprocessing on the training data set;
3-3) for each image, calculating nine definition evaluation function characteristic values as sample values of the image;
and 3-4) inputting the sample values of all the images into a random forest model for training to finally obtain a random forest classification model.
Preferably, the nine sharpness evaluation function feature values specifically include:
the method comprises the following steps of calculating the square of the difference between adjacent gray values in the horizontal direction and the vertical direction of an image through an EOG function, calculating the square sum of the difference between diagonal pixel gray values of the image through a Roberts function, extracting gradient values of pixel points of the image in the horizontal direction and the vertical direction through a Tenengrad function by utilizing a Sobel operator, calculating the difference of two pixel points in the x direction of the image through a Brenner function, obtaining the discrete degree of image gray distribution through a Varian function, obtaining the square sum of gradient matrixes of the image through convolution of a Laplace operator and the pixel points through the Laplace function, obtaining a definition evaluation function based on two-dimensional discrete Fourier transform, obtaining a definition evaluation function based on discrete cosine transform and obtaining an evaluation function based on information entropy.
Preferably, the image preprocessing in step 3-2) includes performing an average filtering process, a median filtering process, a gaussian difference process and a down-sampling process on the image.
Preferably, T is 10 μ M, and M is 10.
Preferably, T is 2 μ M, and M is 10.
The invention also provides a microscope fine focusing system which adopts the random forest based microscope fine focusing method to carry out microscope focusing.
The invention has the beneficial effects that:
according to the microscope fine focusing method based on the random forest, the number of images needing to be collected is small, the time for finding the focal plane is short, and the focusing efficiency can be effectively improved; the invention combines nine definition evaluation functions to judge the focal length, can reduce the probability of falling into a local extreme value, and can improve the robustness and the precision of the focusing method.
Drawings
FIG. 1 is a flowchart of a method for fine focusing of a microscope based on random forest in embodiment 1 of the present invention;
FIG. 2 is a flowchart of training a random forest model in embodiment 1 of the present invention;
FIG. 3 is a schematic flow chart of a random forest model in embodiment 1 of the present invention;
fig. 4 is a detailed diagram of a data set in embodiment 3 of the present invention.
Detailed Description
The present invention is further described in detail below with reference to examples so that those skilled in the art can practice the invention with reference to the description.
It will be understood that terms such as "having," "including," and "comprising," as used herein, do not preclude the presence or addition of one or more other elements or groups thereof.
Example 1
Referring to fig. 1, the method for fine focusing of a microscope based on a random forest according to the present embodiment includes the following steps:
1) focusing for the first time: initially, the lens to be focused is at an arbitrary position, an image Img1 is collected, then the lens is advanced by 2 μm, and an image Img2 is collected again;
2) preprocessing the filtering difference of the image Img1 and the image Img2, calculating nine feature values of the sharpness evaluation function of the image Img1 as a sample 1, and calculating nine feature values of the sharpness evaluation function of the image Img2 as a sample 2;
3) classifying the sample 1 and the sample 2 by using a random forest classification model obtained by pre-training so as to obtain a defocus distance L1 of the image Img1 and a defocus distance L2 of the image Img 2;
if L2 is 0, the current image is located in the focal plane, focusing is completed, and if L2 is not 0, the next step is carried out, and secondary focusing is carried out;
4) and (3) focusing for the second time: calculating energy gradient functions of Img1 and Img2, which are respectively marked as EOG1 and EOG 2;
if EOG1-EOG2 is less than or equal to 0, advancing the lens by a distance L2, otherwise, retreating the lens by a distance L2;
5) acquiring images at the current position, and then acquiring images at positions which are +/-2 μm, +/-4 μm, +/-6 μm, +/-8 μm and +/-10 μm away from the current position respectively (namely N is 5);
6) and calculating respective energy gradient functions of the 11 acquired images, taking the position corresponding to the image with the maximum energy gradient function as a focal plane, and moving the lens to the focal plane to finish focusing.
Referring to fig. 2, in step 3), the method for training the random forest model in advance includes the following steps:
3-1) respectively collecting a plurality of images with out-of-focus distances of 1T, 2T, 3T, M T by taking the step length as T, and establishing a training data set by taking the out-of-focus distances as labels of the images; in this example, T is 10 μ M, and M is 10.
3-2) carrying out image preprocessing on the training data set;
the image preprocessing comprises the steps of carrying out average filtering processing, median filtering processing, Gaussian difference processing and down-sampling processing on the image. Noise interference in the defocused picture can be removed through average filtering processing and median filtering processing; the Gaussian difference processing can inhibit interference caused by color depth, strengthen edge information in the image and serve as an important representation of the defocusing degree of the image; and the image is processed by down-sampling, so that the calculated amount can be reduced, and the automatic focusing speed is improved. Through image preprocessing, the defocused pictures with different degrees are more distinguishable, and the subsequent training of the random forest model is facilitated.
3-3) for each image, calculating nine definition evaluation function characteristic values as sample values of the image;
in the invention, 9 definition evaluation functions are combined, and one definition function value of a picture is taken as a characteristic value of an image sample, so that the probability of falling into a local extreme value is reduced, and the robustness of the algorithm is improved. The nine definition evaluation function characteristic values specifically include:
the method comprises the steps of calculating the square of the difference between adjacent gray values in the horizontal direction and the vertical direction of an image through an EOG (energy of gradient) function, calculating the square sum of the difference between diagonal pixel gray values of the image through a Roberts function, extracting the gradient value of a pixel point in the horizontal direction and the vertical direction of the image through a Tenengrad function by using a Sobel operator, calculating the difference between two pixel points in the x direction of the image through a Brenner function (namely calculating a second-order gradient), obtaining the dispersion degree of image gray distribution through a Variance function, obtaining the square sum of a gradient matrix of the image through convolution of a Laplace function and the pixel points, obtaining a definition evaluation function based on two-dimensional discrete Fourier transform, obtaining the definition evaluation function based on discrete cosine transform and obtaining the evaluation function based on information entropy.
And 3-4) inputting the sample values of all the images into a random forest model for training to finally obtain a random forest classification model.
The random forest model bagging method obtains a training sample subset, a node classification characteristic is obtained by combining a random subspace method, a data set is divided into a plurality of subsets, a decision tree is generated on each sub data set, and the final result is subjected to joint prediction by a plurality of decision trees, which is shown in fig. 3. In the embodiment, the accuracy and the average defocusing error are used as indexes for evaluating the performance of the classification model.
The embodiment also provides a microscope fine focusing system, which adopts the random forest based microscope fine focusing method as described in embodiment 1 to perform microscope focusing.
Example 2
This embodiment differs from embodiment 1 only in that: in this example, a is 1 μ M, T is 2 μ M, and M is 10.
Example 3
In this example, the method of example 1 was used to perform fine focusing of the microscope.
The data set established in this embodiment is: the data set is acquired from lily bud slices, the defocusing distance is 0-100 mu m, and the step length is 5 mu m; about 200 pictures, 4397 pictures in total, were acquired for each defocus distance, and the data set details are shown in fig. 4. A 10-fold cross-validation experiment was performed on this data set.
Based on the data set, the number of decision trees in the random forest in this embodiment is set to 100, and the detailed results of ten-fold cross-validation are shown in table 1. The model accuracy represents the proportion of correct pictures classified by the random forest model, the focusing accuracy represents the proportion of correctly finding the focal plane after the second focusing, and the result shows that the EOG function can basically find the correct focusing position in the positions of +/-2 mu m, +/-4 mu m, +/-6 mu m, +/-8 mu m and +/-10 mu m, and the final focusing accuracy can reach 91.7 percent, thus the method has good robustness and accuracy.
TABLE 1 Ten-fold cross-validation classification results
While embodiments of the invention have been disclosed above, it is not limited to the applications listed in the description and the embodiments, which are fully applicable in all kinds of fields of application of the invention, and further modifications may readily be effected by those skilled in the art, so that the invention is not limited to the specific details without departing from the general concept defined by the claims and the scope of equivalents.
Claims (10)
1. A microscope fine focusing method based on random forests is characterized by comprising the following steps:
1) focusing for the first time: initially, the lens to be focused is at any position, an image Img1 is collected, then the lens is advanced by a distance A, and an image Img2 is collected again;
2) firstly, filtering difference processing is carried out on the image Img1 and the image Img2, then nine definition evaluation function characteristic values of the image Img1 are calculated to be used as a sample 1, and nine definition evaluation function characteristic values of the image Img2 are calculated to be used as a sample 2;
3) classifying the sample 1 and the sample 2 by using a random forest classification model obtained by pre-training so as to obtain a defocus distance L1 of the image Img1 and a defocus distance L2 of the image Img 2;
if L2 is 0, the current image is located in the focal plane, focusing is completed, and if L2 is not 0, the next step is carried out, and secondary focusing is carried out;
4) and (3) focusing for the second time: calculating energy gradient functions of Img1 and Img2, which are respectively marked as EOG1 and EOG 2;
if EOG1-EOG2 is less than or equal to 0, advancing the lens by a distance L2, otherwise, retreating the lens by a distance L2;
5) acquiring images at a current position, and then respectively acquiring the images at positions which are +/-A, + -2A, + -3A, + -2N and A away from the current position, wherein N is a positive integer;
6) and calculating the energy gradient function of each of the acquired 1+2N images, taking the position corresponding to the image with the maximum energy gradient function as a focal plane, and moving the lens to the focal plane to finish focusing.
2. The random forest based microscope fine focusing method of claim 1, wherein a ═ 2 μm.
3. The random forest based microscope fine focusing method of claim 1, wherein a ═ 1 μm.
4. A random forest based microscope fine focusing method as claimed in claim 2 or 3 wherein N-5.
5. The random forest based microscope fine focusing method as claimed in claim 1, wherein in the step 3), the method for pre-training the random forest model comprises the following steps:
3-1) respectively collecting a plurality of images with out-of-focus distances of 1T, 2T, 3T, M T by taking the step length as T, and establishing a training data set by taking the out-of-focus distances as labels of the images;
3-2) carrying out image preprocessing on the training data set;
3-3) for each image, calculating nine definition evaluation function characteristic values as sample values of the image;
and 3-4) inputting the sample values of all the images into a random forest model for training to finally obtain a random forest classification model.
6. The random forest based microscope fine focusing method as claimed in claim 5, wherein the nine sharpness evaluation function feature values specifically comprise:
the method comprises the following steps of calculating the square of the difference between adjacent gray values in the horizontal direction and the vertical direction of an image through an EOG function, calculating the square sum of the difference between diagonal pixel gray values of the image through a Roberts function, extracting gradient values of pixel points of the image in the horizontal direction and the vertical direction through a Tenengrad function by utilizing a Sobel operator, calculating the difference of two pixel points in the x direction of the image through a Brenner function, obtaining the discrete degree of image gray distribution through a Varian function, obtaining the square sum of gradient matrixes of the image through convolution of a Laplace operator and the pixel points through the Laplace function, obtaining a definition evaluation function based on two-dimensional discrete Fourier transform, obtaining a definition evaluation function based on discrete cosine transform and obtaining an evaluation function based on information entropy.
7. The random forest based microscope fine focusing method as claimed in claim 5, wherein the image preprocessing in the step 3-2) comprises performing an average filtering process, a median filtering process, a Gaussian difference processing and a down-sampling process on the image.
8. The random forest based microscope fine focusing method of claim 5, wherein T-10 μ M and M-10.
9. The random forest based microscope fine focusing method of claim 5, wherein T-2 μ M and M-10.
10. A microscope fine focusing system, characterized in that it uses the random forest based microscope fine focusing method as claimed in any of claims 1-9 for microscope focusing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011438427.1A CN112433339B (en) | 2020-12-10 | 2020-12-10 | Microscope fine focusing method based on random forest |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011438427.1A CN112433339B (en) | 2020-12-10 | 2020-12-10 | Microscope fine focusing method based on random forest |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112433339A CN112433339A (en) | 2021-03-02 |
| CN112433339B true CN112433339B (en) | 2022-04-15 |
Family
ID=74692402
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011438427.1A Active CN112433339B (en) | 2020-12-10 | 2020-12-10 | Microscope fine focusing method based on random forest |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112433339B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113114934B (en) * | 2021-03-31 | 2022-08-05 | 太原理工大学 | Multi-focus video acquisition method and system for urine red blood cells |
| CN114760419B (en) * | 2022-06-15 | 2022-09-20 | 深圳深知未来智能有限公司 | Automatic focusing method and system based on deep learning |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102169275A (en) * | 2010-04-28 | 2011-08-31 | 上海盈方微电子有限公司 | Automatic focusing system of digital camera for non-uniform sampling window planning based on golden section |
| CN106373107A (en) * | 2016-12-06 | 2017-02-01 | 四川长虹电器股份有限公司 | Automatic image deblurring system and automatic image deblurring method of smart phone |
| CN107509023A (en) * | 2017-07-24 | 2017-12-22 | 上海理工大学 | A kind of auto-focusing searching algorithm |
| CN109507792A (en) * | 2018-12-27 | 2019-03-22 | 湖南品信生物工程有限公司 | A kind of optical microscopy auto focusing method based on machine learning |
| CN109598704A (en) * | 2018-11-19 | 2019-04-09 | 电子科技大学 | A kind of night soil microscopic image clarity evaluation method based on BP neural network |
| JP2019186911A (en) * | 2018-04-10 | 2019-10-24 | キヤノン株式会社 | Image processing apparatus, image processing method, and imaging apparatus |
| CN111007661A (en) * | 2019-12-02 | 2020-04-14 | 湖南国科智瞳科技有限公司 | Microscopic image automatic focusing method and device based on deep learning |
| DE102018133188A1 (en) * | 2018-12-20 | 2020-06-25 | Carl Zeiss Microscopy Gmbh | DISTANCE DETERMINATION OF A SAMPLE LEVEL IN A MICROSCOPE SYSTEM |
| CN111345025A (en) * | 2018-08-29 | 2020-06-26 | 深圳市大疆创新科技有限公司 | Camera device and focusing method |
| CN111432129A (en) * | 2020-04-14 | 2020-07-17 | 广东欧谱曼迪科技有限公司 | Automatic focusing hill climbing search algorithm based on threshold value method and local maximum value |
-
2020
- 2020-12-10 CN CN202011438427.1A patent/CN112433339B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102169275A (en) * | 2010-04-28 | 2011-08-31 | 上海盈方微电子有限公司 | Automatic focusing system of digital camera for non-uniform sampling window planning based on golden section |
| CN106373107A (en) * | 2016-12-06 | 2017-02-01 | 四川长虹电器股份有限公司 | Automatic image deblurring system and automatic image deblurring method of smart phone |
| CN107509023A (en) * | 2017-07-24 | 2017-12-22 | 上海理工大学 | A kind of auto-focusing searching algorithm |
| JP2019186911A (en) * | 2018-04-10 | 2019-10-24 | キヤノン株式会社 | Image processing apparatus, image processing method, and imaging apparatus |
| CN111345025A (en) * | 2018-08-29 | 2020-06-26 | 深圳市大疆创新科技有限公司 | Camera device and focusing method |
| CN109598704A (en) * | 2018-11-19 | 2019-04-09 | 电子科技大学 | A kind of night soil microscopic image clarity evaluation method based on BP neural network |
| DE102018133188A1 (en) * | 2018-12-20 | 2020-06-25 | Carl Zeiss Microscopy Gmbh | DISTANCE DETERMINATION OF A SAMPLE LEVEL IN A MICROSCOPE SYSTEM |
| CN109507792A (en) * | 2018-12-27 | 2019-03-22 | 湖南品信生物工程有限公司 | A kind of optical microscopy auto focusing method based on machine learning |
| CN111007661A (en) * | 2019-12-02 | 2020-04-14 | 湖南国科智瞳科技有限公司 | Microscopic image automatic focusing method and device based on deep learning |
| CN111432129A (en) * | 2020-04-14 | 2020-07-17 | 广东欧谱曼迪科技有限公司 | Automatic focusing hill climbing search algorithm based on threshold value method and local maximum value |
Non-Patent Citations (8)
| Title |
|---|
| Autofocus for 3D imaging;Lee-Elkin;《Proc. of SPIE》;20080415;全文 * |
| Learning to autofocus based on Gradient Boosting Machine for optical microscopy;Yixiong Liang;《Optik - International Journal for Light and Electron Optics》;20190623;全文 * |
| Sentiment classification of online consumer reviews using word vector representations;Barkha Bansal;《Procedia Computer Science》;20180529;全文 * |
| Yoon Kyeong Ho ; Lee Dae Jong ; Chun Myung Genu.Development of an Auto-focusing Algorithm Based on Random Forest.《THE TRANSACTION OF THE KOREAN INSTITUTE OF ELECTRICAL ENGINEERS P》.2020,全文. * |
| 单幅图像估计离焦量的航空摄像机自动调焦系统;王昊;《液晶与显示》;20160531;全文 * |
| 应用于一体化摄像机的自动聚焦搜索算法研究;曹永鹏;《应用光学》;20180731;全文 * |
| 数字PCR仪成像系统的自动对焦算法研究;陈善雄;《重庆大学学报》;20190930;全文 * |
| 由粗到精的虹膜图像离焦模糊评价方法;李星光;《Journal of Frontiers of Computer Science and Technology》;20140831;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112433339A (en) | 2021-03-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112433339B (en) | Microscope fine focusing method based on random forest | |
| Pertuz et al. | Analysis of focus measure operators for shape-from-focus | |
| US9297995B2 (en) | Automatic stereological analysis of biological tissue including section thickness determination | |
| CN110531484B (en) | Microscope automatic focusing method with settable focusing process model | |
| Yousefi et al. | A new auto-focus sharpness function for digital and smart-phone cameras | |
| Sigdel et al. | Focusall: Focal stacking of microscopic images using modified harris corner response measure | |
| US20120249770A1 (en) | Method for automatically focusing a microscope on a predetermined object and microscope for automatic focusing | |
| CN105354817A (en) | Noise image automatic focusing method | |
| Xu et al. | A comparison of contrast measurements in passive autofocus systems for low contrast images | |
| Dogan et al. | Automatic panorama with auto-focusing based on image fusion for microscopic imaging system | |
| CN111105354A (en) | Depth image super-resolution method and device based on multi-source depth residual error network | |
| Madali et al. | Automatic depth map retrieval from digital holograms using a depth-from-focus approach | |
| Hao et al. | Improving the performances of autofocus based on adaptive retina-like sampling model | |
| CN109507792B (en) | Optical microscope automatic focusing method based on machine learning | |
| CN109151319B (en) | Method and device for processing focusing sequence of mark points of multi-mark-point target object | |
| CN112817118B (en) | Infrared automatic focusing fast searching method | |
| Li et al. | Scene-adaptive image acquisition for focus stacking | |
| CN114384681A (en) | Rapid and accurate automatic focusing method and system for microscope, computer equipment and medium | |
| Dogan et al. | Auto-focusing with multi focus color image fusion based on curvelet transform on microscopic imaging | |
| Zhang et al. | Autofocus method based on multi regions of interest window for cervical smear images | |
| CN112950698A (en) | Depth estimation method, device, medium, and apparatus based on binocular defocused image | |
| CN117452619B (en) | Sparse target microscopic imaging automatic focusing method, system and storage medium | |
| Song et al. | A new auto-focusing algorithm for optical microscope based automated system | |
| Aslantas et al. | Multi focus image fusion by differential evolution algorithm | |
| CN110610470B (en) | Camera multi-focus clear image extraction method based on multi-azimuth gradient comparison |
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 |