CN112540456A - Microscope precision automatic focusing method based on human-simulated definition judgment - Google Patents
Microscope precision automatic focusing method based on human-simulated definition judgment Download PDFInfo
- Publication number
- CN112540456A CN112540456A CN202011410109.4A CN202011410109A CN112540456A CN 112540456 A CN112540456 A CN 112540456A CN 202011410109 A CN202011410109 A CN 202011410109A CN 112540456 A CN112540456 A CN 112540456A
- Authority
- CN
- China
- Prior art keywords
- definition
- images
- image
- focusing
- microscope
- 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
Images
Classifications
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B21/00—Microscopes
- G02B21/24—Base structure
- G02B21/241—Devices for focusing
- G02B21/244—Devices for focusing using image analysis techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Optics & Photonics (AREA)
- Microscoopes, Condenser (AREA)
- Automatic Focus Adjustment (AREA)
Abstract
The invention relates to a microscope precision automatic focusing method based on humanoid definition judgment, and belongs to the technical field of experiments. The invention creates a human-simulated definition judgment model based on an artificial intelligence deep learning method, trains a neural network model with definition judgment capability, enables a microscope to have human-level precise automatic focusing capability, and drives the microscope to realize high-quality and stable automatic focusing photographing effect in each observation visual field in a rapid, continuous and precise automatic focusing manner, thereby reducing the cost, improving the efficiency and providing a high-definition image for subsequent detection and analysis.
Description
Technical Field
The invention belongs to the technical field of experiments, and relates to a microscope precision automatic focusing method based on humanoid definition judgment.
Background
The theory and methodology of microscope focusing techniques have been rapidly developed, but have not matured in the practical application of some imaging systems, particularly in the medical field where microscopes are frequently used.
The good focusing mode plays an important role in acquiring clear image information. The existing microscope focusing methods mainly comprise:
the first prior art is as follows: manual focusing: and an operator manually rotates the fine adjustment knob to focus according to the observation result of the microscope eyepiece.
Has the advantages that: the price is low, the equipment is not needed to be matched for focusing, and the shot image is clear.
The disadvantages are: the labor cost is too high, the efficiency is low, the focusing result needs to be judged according to the experience of an individual, and uncertainty exists.
The second prior art is: conventional electric focusing of microscope: under the control of a computer, the microscope is electrically adjusted to automatically focus up and down by calculating definition or object distance.
Has the advantages that: the labor cost is low, and the speed is high.
The disadvantages are: one is high in price, particularly the best automatic focusing microscopic photographing equipment in the market at present, wherein the cost of an electric objective table which is one of important accessories is hundreds of thousands, and the price of the whole equipment is hundreds of thousands; secondly, the definition of the shot image is uncertain, and especially when a plurality of images need to be shot in a multi-dimensional manner for a detection object containing liquid matrixes such as white bands, urine and the like with different volume sizes, the focusing mode cannot meet the requirement.
Disclosure of Invention
In view of the above, the present invention provides a method for precisely auto-focusing a microscope based on human-like sharpness determination.
In order to achieve the purpose, the invention provides the following technical scheme:
a microscope precision automatic focusing method based on human-simulated definition judgment comprises the following steps:
step 1: establishing a definition grading mechanism in the early stage, and learning the definition grading mechanism of an expert on an image by utilizing the characterization capability of deep learning on the detail characteristics of the bottom layer of the image and the extraction capability of deep patterns so as to enable a machine to obtain the definition grading capability similar to human eyes;
step 2: placing the prepared slide to be detected on an electric objective table, and fixing the slide to be detected by a fixing device;
and step 3: the computer drives the electric objective table with the slide to move to a set fixed origin, namely reset, the computer controls an X-axis motor and a Y-axis motor of the microscope, and the motor drives a sample to be detected of the slide to move to the center of the electric objective table;
and 4, step 4: the computer controls the electric objective table to move, the microscope firstly shoots a microscopic image at a first visual field position in the Z-axis direction, and the acquired image is quickly transmitted to a convolution depth neural network of the computer;
and 5: judging to obtain an image definition value under the focusing position through a convolution depth neural network, comparing the definition value with a score value of a definition grading mechanism, collecting images for multiple times, circularly judging, and collecting to obtain a first group of n images;
step 6: and controlling the electric objective table to move along an axis X, Y by the computer, continuously searching the next field of view for focusing and photographing, acquiring images and performing definition contrast, acquiring a second group of n images, continuously moving to the mth position, and acquiring m images.
Optionally, in step 5, after receiving the image, the computer determines to obtain an image sharpness value at the focusing position through a convolutional deep neural network, compares the sharpness value with a score value of a sharpness scoring mechanism, and directly outputs the image if the sharpness value is greater than the score value; if the number of the acquired images is less than the fraction value, a new Z-axis direction position is obtained through computer calculation, the images are acquired again, and the first group of n images are acquired through cyclic judgment.
Optionally, after receiving the image, the computer determines to obtain an image sharpness value at the focusing position through the convolutional depth neural network, and compares the sharpness value with a score value of a sharpness scoring mechanism, wherein the score value of the sharpness scoring mechanism is 90 scores, and the error between the sharpness scoring mechanism and the human eye judgment is judged only within a minimum basic scale of the fine-adjustment focusing disc by a representative algorithm of more than 90 scores, namely, the precision cannot be improved on the basis of switching the manual focusing.
Optionally, the first set of n images comprises at least 3 images.
Optionally, the mth location comprises at least 16 locations.
The invention has the beneficial effects that:
the invention creates a human-simulated definition judgment model based on an artificial intelligence deep learning method, trains a neural network model with definition judgment capability, enables a microscope to have human-level precise automatic focusing capability, and drives the microscope to realize high-quality and stable automatic focusing photographing effect in each observation visual field in a rapid, continuous and precise automatic focusing manner, thereby reducing the cost, improving the efficiency and providing a high-definition image for subsequent detection and analysis.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
For the purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:
FIG. 1 is a flow chart of the present invention;
fig. 2 is a schematic diagram of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.
Wherein the showings are for the purpose of illustrating the invention only and not for the purpose of limiting the same, and in which there is shown by way of illustration only and not in the drawings in which there is no intention to limit the invention thereto; to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The same or similar reference numerals in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if there is an orientation or positional relationship indicated by terms such as "upper", "lower", "left", "right", "front", "rear", etc., based on the orientation or positional relationship shown in the drawings, it is only for convenience of description and simplification of description, but it is not an indication or suggestion that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes, and are not to be construed as limiting the present invention, and the specific meaning of the terms may be understood by those skilled in the art according to specific situations.
Referring to fig. 1 and 2, the apparatus of the present invention includes: computers and microscopes.
The computer mainly controls a motor of the microscope to drive the electric objective table to move to a corresponding position and is responsible for calculating and controlling functions; two important accessories of the microscope are an electric stage for placing a sample to be tested and a camera for taking an image and transmitting the image to a computer.
The method comprises the following specific steps:
1. establishing a definition grading mechanism in the early stage, and learning the definition grading mechanism of an expert on an image by utilizing the characterization capability of deep learning on the detail characteristics of the bottom layer of the image and the extraction capability of deep patterns so as to enable a machine to obtain the definition grading capability similar to human eyes;
2. placing the prepared slide to be detected on an electric objective table, and fixing the slide by a pressing device;
3. the computer drives the electric objective table with the slide to move to a set fixed origin, namely reset, the computer controls the X-axis motor and the Y-axis motor of the microscope, and the motor drives the sample to be detected of the slide to move to the center of the electric objective table;
4. the computer controls the electric objective table to move, a microscope firstly shoots a microscope image at a first visual field position in the Z-axis direction, and the acquired image is quickly transmitted to a convolution depth neural network of the computer;
5. judging to obtain an image definition value under the focusing position through a convolution depth neural network, comparing the definition value with a certain score value of a definition scoring mechanism, and if the definition value is greater than the score value, directly outputting an image; if the number of the acquired images is less than the fraction value, obtaining a new Z-axis direction position through computer calculation, acquiring the images again, focusing and photographing, and performing cyclic judgment, so that a first group of n images can be acquired;
6. and the computer controls the motor to move along an X, Y axis, continuously searches the next field of view for focusing and photographing, acquires images and contrasts the images in definition, acquires a second group of n images, continuously moves to the mth field of view position, and acquires m images.
A certain point value of the definition scoring mechanism is 90 points, and more than 90 points represent that the error between the algorithm judgment and the human eye judgment is only within a minimum basic scale of the fine adjustment focusing disc, namely, the precision cannot be improved on the basis of switching the manual focusing.
And 5, searching a new Z-axis direction position, adopting a hill climbing algorithm, starting climbing along a certain direction from a starting point by simulating the climbing of the blind person, changing the focal length by starting climbing with a certain larger step length, comparing by calculating a definition evaluation value every time, reducing the step length to climb reversely when a falling edge of the slope is detected, repeatedly turning back the climbing in the way, stopping climbing until the step length is reduced to a preset termination step length, and obtaining the maximum value searched by the blind person hill climbing method at the peak position in the last climbing process.
And n of the n images is at least 3.
The m positions are taken to be m at least 16.
The invention is mechanically and automatically operated, does not need manual focusing and reduces the labor cost;
according to the invention, a unique definition scoring mechanism is creatively created through an artificial intelligence deep learning method, so that the focusing error of the microscope is indicated, and the microscope is driven to automatically and precisely focus continuously, stably and quickly.
The invention has mobility, can realize rapid and accurate automatic focusing on a common microscope, does not need to be provided with expensive microscope equipment, and equivalently reduces the equipment cost.
Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.
Claims (5)
1. A microscope precision automatic focusing method based on human-simulated definition judgment is characterized in that: the method comprises the following steps:
step 1: establishing a definition grading mechanism in the early stage, and learning the definition grading mechanism of an expert on an image by utilizing the characterization capability of deep learning on the detail characteristics of the bottom layer of the image and the extraction capability of deep patterns so as to enable a machine to obtain the definition grading capability similar to human eyes;
step 2: placing the prepared slide to be detected on an electric objective table, and fixing the slide to be detected by a fixing device;
and step 3: the computer drives the electric objective table with the slide to move to a set fixed origin, namely reset, the computer controls an X-axis motor and a Y-axis motor of the microscope, and the motor drives a sample to be detected of the slide to move to the center of the electric objective table;
and 4, step 4: the computer controls the electric objective table to move, the microscope firstly shoots a microscopic image at a first visual field position in the Z-axis direction, and the acquired image is quickly transmitted to a convolution depth neural network of the computer;
and 5: judging to obtain an image definition value under the focusing position through a convolution depth neural network, comparing the definition value with a score value of a definition grading mechanism, collecting images for multiple times, circularly judging, and collecting to obtain a first group of n images;
step 6: and controlling the electric objective table to move along an axis X, Y by the computer, continuously searching the next field of view for focusing and photographing, acquiring images and performing definition contrast, acquiring a second group of n images, continuously moving to the mth position, and acquiring m images.
2. The method for precisely automatically focusing the microscope according to claim 1, wherein the method comprises the following steps: in the step 5, after receiving the image, the computer judges through a convolutional depth neural network to obtain an image definition value at the focusing position, compares the definition value with a score value of a definition scoring mechanism, and directly outputs the image if the definition value is greater than the score value; if the number of the acquired images is less than the fraction value, a new Z-axis direction position is obtained through computer calculation, the images are acquired again, and the first group of n images are acquired through cyclic judgment.
3. The method for precisely automatically focusing the microscope according to claim 2, wherein the method comprises the following steps: after receiving the image, the computer judges through the convolution depth neural network to obtain the image definition value under the focusing position, compares the definition value with the score value of the definition scoring mechanism, the score value of the definition scoring mechanism is 90 scores, the error between the algorithm judgment represented by more than 90 scores and the human eye judgment is only within a minimum basic scale of the fine adjustment focusing disc, namely, the precision cannot be improved on the basis of switching manual focusing.
4. The method for precisely automatically focusing the microscope according to claim 1, wherein the method comprises the following steps: the first set of n images comprises at least 3 images.
5. The method for precisely automatically focusing the microscope according to claim 1, wherein the method comprises the following steps: the mth location includes at least 16 locations.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011410109.4A CN112540456A (en) | 2020-12-03 | 2020-12-03 | Microscope precision automatic focusing method based on human-simulated definition judgment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011410109.4A CN112540456A (en) | 2020-12-03 | 2020-12-03 | Microscope precision automatic focusing method based on human-simulated definition judgment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112540456A true CN112540456A (en) | 2021-03-23 |
Family
ID=75016021
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011410109.4A Pending CN112540456A (en) | 2020-12-03 | 2020-12-03 | Microscope precision automatic focusing method based on human-simulated definition judgment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112540456A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113538545A (en) * | 2021-07-16 | 2021-10-22 | 上海大学 | Monocular depth estimation method based on electro-hydraulic adjustable-focus lens and corresponding camera and storage medium |
| CN113741021A (en) * | 2021-07-19 | 2021-12-03 | 南方医科大学南方医院 | Automatic focusing method and device for microscope |
| CN113837079A (en) * | 2021-09-24 | 2021-12-24 | 苏州贝康智能制造有限公司 | Automatic focusing method and device for microscope, computer equipment and storage medium |
| CN115242979A (en) * | 2022-07-22 | 2022-10-25 | 湖南伊鸿健康科技有限公司 | Focusing method and system applied to blood shooting, intelligent equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW200620990A (en) * | 2004-12-10 | 2006-06-16 | Hon Hai Prec Ind Co Ltd | System and method for focusing images automatically |
| US20150219979A1 (en) * | 2012-10-18 | 2015-08-06 | Suzhou Hyssen Electronics Co., Ltd | Microscopic image detecting instrument and automatic focusing method therefor |
| CN106324795A (en) * | 2016-09-29 | 2017-01-11 | 电子科技大学 | Automatic fast focusing method for multiple views of detector's microscope |
| 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 |
| CN110531484A (en) * | 2019-07-24 | 2019-12-03 | 中国地质大学（武汉） | A kind of microscope Atomatic focusing method that focus process model can be set |
| CN111552069A (en) * | 2020-04-21 | 2020-08-18 | 中国人民解放军国防科技大学 | Microscopic image automatic focusing method and system based on deep reinforcement learning |
-
2020
- 2020-12-03 CN CN202011410109.4A patent/CN112540456A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW200620990A (en) * | 2004-12-10 | 2006-06-16 | Hon Hai Prec Ind Co Ltd | System and method for focusing images automatically |
| US20150219979A1 (en) * | 2012-10-18 | 2015-08-06 | Suzhou Hyssen Electronics Co., Ltd | Microscopic image detecting instrument and automatic focusing method therefor |
| CN106324795A (en) * | 2016-09-29 | 2017-01-11 | 电子科技大学 | Automatic fast focusing method for multiple views of detector's microscope |
| 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 |
| CN110531484A (en) * | 2019-07-24 | 2019-12-03 | 中国地质大学（武汉） | A kind of microscope Atomatic focusing method that focus process model can be set |
| CN111552069A (en) * | 2020-04-21 | 2020-08-18 | 中国人民解放军国防科技大学 | Microscopic image automatic focusing method and system based on deep reinforcement learning |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113538545A (en) * | 2021-07-16 | 2021-10-22 | 上海大学 | Monocular depth estimation method based on electro-hydraulic adjustable-focus lens and corresponding camera and storage medium |
| CN113741021A (en) * | 2021-07-19 | 2021-12-03 | 南方医科大学南方医院 | Automatic focusing method and device for microscope |
| CN113741021B (en) * | 2021-07-19 | 2024-01-16 | 南方医科大学南方医院 | Automatic focusing method and device for microscope |
| CN113837079A (en) * | 2021-09-24 | 2021-12-24 | 苏州贝康智能制造有限公司 | Automatic focusing method and device for microscope, computer equipment and storage medium |
| CN113837079B (en) * | 2021-09-24 | 2024-05-14 | 苏州贝康智能制造有限公司 | Automatic focusing method, device, computer equipment and storage medium of microscope |
| CN115242979A (en) * | 2022-07-22 | 2022-10-25 | 湖南伊鸿健康科技有限公司 | Focusing method and system applied to blood shooting, intelligent equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112540456A (en) | Microscope precision automatic focusing method based on human-simulated definition judgment | |
| CN111007661B (en) | Microscopic image automatic focusing method and device based on deep learning | |
| DE102004051159B4 (en) | Face identification device, face identification method and face identification program | |
| US10964232B2 (en) | Apparatus, method and system for microsurgical suture training | |
| CN109559275B (en) | Microscopic image stitching method of urine analyzer | |
| CN105682310B (en) | Combined lighting device and method based on image quality control | |
| CN110531484B (en) | Microscope automatic focusing method with settable focusing process model | |
| JP5040191B2 (en) | Microinjection apparatus and automatic focus adjustment method | |
| CN109085113B (en) | Automatic focusing method and device for cervical exfoliated cell detection device | |
| CN107797262B (en) | Microscope different multiples camera lens based on image texture combines focus method | |
| CN109031638B (en) | Quick automatic focusing method of biological microscope | |
| CN113406076A (en) | Smear image acquisition and analysis system | |
| CN106873142B (en) | High-quality image acquisition device and method of tubercle bacillus detector | |
| CN110766656B (en) | Method, device, equipment and storage medium for screening fundus macular region abnormality | |
| CN105004723A (en) | Pathological section scanning 3D imaging and fusion device and method | |
| WO2018127509A1 (en) | Method for generating a three-dimensional model of a sample in a digital microscope and a digital microscope | |
| US10812701B2 (en) | High-speed tag lens assisted 3D metrology and extended depth-of-field imaging | |
| DE102019133756A1 (en) | SYSTEMS AND METHODS FOR GENERATING A LIGHTING DESIGN | |
| CN110765835A (en) | Operation video flow identification method based on edge information | |
| CN111474103B (en) | Automatic focusing scanning method and system for bone marrow cell glass slide | |
| CN113777769B (en) | Automatic focusing method and device for microscopic instrument, intelligent terminal and storage medium | |
| EP3355101A1 (en) | Computer-implemented method for determining a representation of a spectacle socket rim or a representation the edges of the glasses of a pair of spectacles | |
| KR20200053718A (en) | Blood diagnostic apparatus | |
| EP1262810A2 (en) | Method of distance measurement of extended objects with an optical viewing device and microscope applying it | |
| CN110764220B (en) | Method and system for acquiring focusing curve of zoom lens and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210323 |
|
| RJ01 | Rejection of invention patent application after publication |