CN120802484A

CN120802484A - Super-depth-of-field synthesis intelligent identification microscopic device

Info

Publication number: CN120802484A
Application number: CN202510619075.6A
Authority: CN
Inventors: 陈健; 李婷婷; 邱德义; 魏晓雅; 刘德星; 仇永军
Original assignee: Guangzhou Gaoming Biotechnology Co ltd; Zhongshan Customs Technical Center
Current assignee: Guangzhou Gaoming Biotechnology Co ltd; Zhongshan Customs Technical Center
Priority date: 2025-05-14
Filing date: 2025-05-14
Publication date: 2025-10-17

Abstract

The present invention discloses a super-depth-of-field synthetic intelligent identification microscope device, which relates to the field of identification technology. The device includes: a microscope unit, a high-definition camera unit and a super-depth-of-field program control unit. The super-depth-of-field program control unit includes a controller processing module, an image data processing module and an intelligent identification and recognition module, wherein the controller processing module is used to drive the imaging camera to capture the overall and detailed structure of the sample to obtain image data of the sample, and the image data includes multiple pictures or multiple video clips of the sample at different focal planes; the image data processing module is used to generate synthetic image data of the sample according to the image data through a preset multi-focal plane adaptive stacking algorithm, and the synthetic image data includes multiple synthetic pictures of the sample; the intelligent identification and recognition module is used to obtain an intelligent identification result by comparing the synthetic image data with known sample data in a pre-established digital specimen library.

Description

Super-depth-of-field synthesis intelligent identification microscopic device

Technical Field

The invention relates to the technical field of identification, in particular to an intelligent identification microscopic device for super-depth-of-field synthesis

Background

The microscopic photographing device is a common tool in laboratories, plays a key role in research and practice in numerous fields of biology, medicine, chemistry, industry, material science and the like, and is mainly used for observing and photographing structures of fine samples. For example, the method is used for observing the morphology of cells and microorganisms in the biological field, assisting pathological analysis in medicine, researching microscopic crystal structure in the chemical field, industrially detecting tiny flaws of products, analyzing microscopic structures of materials in the material science, and the like.

Automated imaging systems are limited by optical resolution and image algorithms. The insufficient optical resolution makes the synthesized picture have low definition and difficult to present a sample microstructure, the blurring of a partial region of the sample is caused by the blurring of a partial structure, and the deviation of the picture color and the actual color of the sample is caused by the low color rendition. These problems can lead to large errors in extracting morphological features from pictures, and in subsequent authentication or transmission authentication, deviation of authentication results is easily caused by inaccurate feature extraction.

At present, professional technicians are relied on to manually shoot a synthesized image, and although details can be well reserved, the efficiency is low, and the average shooting time of a single sample is long. And the focal plane synthesis restriction degree is high, and a plurality of difficulties are faced when images of different focal planes are synthesized into a complete clear image, so that the final imaging effect is influenced.

Species morphological identification mainly depends on morphological taxonomy expert, however, the expert talent echelon fault phenomenon is serious, and the micro-anatomical knowledge has higher cognitive barrier, so that the identification work is difficult to be widely and efficiently carried out. Many laboratories are not able to independently perform accurate species identification without the expertise of a person.

Sub-centimeter grade samples (< 5 mm) had surface microstructures that easily flaked off during transport, resulting in compromised sample characteristics. Meanwhile, the microscopic photographing link has insufficient signal-to-noise ratio of equipment, so that the image is easy to mix into noise interference, a blind area exists in microscopic feature identification by an operator, important features can be omitted, and texture loss can be caused by lossy compression of pictures. These factors combine to reduce the integrity of the multi-dimensional features of the sample and increase the difficulty of species identification.

Disclosure of Invention

The invention aims to provide an intelligent identification microscopic device for super-depth-of-field synthesis, which solves the problems faced by the background technology.

In order to achieve the above purpose, the present invention provides the following technical solutions:

A super depth of field synthetic intelligent identification microscopy device comprising:

the microscope unit comprises an objective table assembly, a Z-axis moving frame and an objective lens assembly, wherein the objective table assembly is used for placing a sample, the Z-axis moving frame is connected with the objective lens assembly and the objective table assembly in a supporting manner, and the objective lens assembly is connected with an ocular lens and an objective lens;

The high-definition camera unit comprises an imaging camera with a variable-focus regulation lens;

The super-depth-of-field program control unit comprises a controller processing module, an image data processing module and an intelligent identification module, wherein,

The controller processing module is used for driving the imaging camera to shoot the whole and detail structure of the sample to obtain image data of the sample, wherein the image data comprises a plurality of pictures or multi-section video recordings of the sample at different focal planes;

the image data processing module is used for generating synthetic image data of the sample through a preset multi-focal-plane self-adaptive stacking algorithm according to the image data, wherein the synthetic image data comprises a plurality of synthetic pictures of the sample;

The intelligent identification module is used for obtaining an intelligent identification result by comparing the synthesized image data with known sample data of a pre-established digital sample library.

The above-mentioned super-depth-of-field synthesis intelligent identification microscopic device, further, drives the imaging camera to shoot the image data of the whole and detail structure of the sample, specifically includes:

The controller processing module sends shooting distance and shooting speed shooting instructions of an imaging camera for controlling the variable-focus regulation and control lens to the high-definition camera unit;

and the imaging camera shoots the image data of the whole sample and the detailed structure according to the received shooting instruction.

The above-mentioned super-depth-of-field synthetic intelligent identification microscopy device further, the generating synthetic image data of the sample according to the image data by a preset multi-focal-plane adaptive stacking algorithm specifically includes:

sliding the preset displacement windows on the picture row by row or column by column according to a set step length, extracting the characteristic information of an image area in each window position, classifying and integrating the extracted characteristic information, and obtaining a layered characteristic diagram;

Carrying out quantitative analysis on the layered feature images, calculating the complexity and the size of the layered feature images to obtain details and global information of the layered feature images, and automatically judging whether each layered feature image belongs to a high-definition image or not by utilizing a preset algorithm and a set high-definition image standard according to the details and the global information of the layered feature images;

according to the hierarchical feature images belonging to the high-definition images, carrying out self-adaptive stacking to obtain a composite image, wherein when the matching degree of the current hierarchical feature image and the previous hierarchical feature image is smaller than a preset threshold, deleting the current hierarchical feature image and carrying out re-shooting;

And (3) carrying out stain elimination and noise reduction treatment on all the synthesized pictures to obtain synthesized image data.

The super depth of field synthesis intelligent identification microscopic device is further used for digitizing known sample data of a specimen library, and is specifically built by the following steps:

And (3) carrying out comprehensive image acquisition on each known physical sample in the digital sample library by using shooting equipment, creating a single folder for each sample, acquiring layered feature graphs with different angles and different feature structures in the folder, taking the layered feature graph of each known physical sample as known sample data, and storing the known sample data in an image information library.

The above-mentioned super depth of field synthesis intelligent identification microscopic device, further, according to the synthetic image data, obtains an intelligent identification result by comparing with known sample data of a pre-established digital sample library, specifically including:

And according to the layered characteristic diagram of the synthesized image data, comparing the layered characteristic diagram with the layered characteristic diagram of the known sample data of the image information base of the pre-established digital sample base, and listing a comparison result as an intelligent identification result.

The super depth of field synthesis intelligent identification microscopic device further comprises a USB data processing module, wherein the USB data processing module is used for converting data transmitted between the digital specimen library and the intelligent identification recognition module according to a set format.

The above-described super depth of field synthetic intelligent identification microscopy apparatus, further, the stage assembly, comprising:

A main stage;

The rotary objective table is arranged on the main objective table, and an X-axis moving objective table and a Y-axis moving objective table are arranged on the rotary objective table.

The super depth of field synthesis intelligent identification microscopic device is characterized in that the light source device is arranged near the objective lens, and the controller processing module is connected with the light source device in a control signal manner.

The super depth of field synthesis intelligent identification microscopic device further comprises a manipulator, wherein the manipulator is in control signal connection with the controller of the rotary objective table, the X-axis moving objective table, the Y-axis moving objective table, the Z-axis moving frame and the light source device.

The super depth of field synthesis intelligent identification microscopic device further comprises an upper computer, and the controller processing module is connected with the upper computer in a control signal manner.

Compared with the prior art, the invention has the beneficial effects that:

At present, professional technicians are relied on to manually shoot a single Zhang Zhaopian, and although details can be better reserved, the efficiency is lower, and the average shooting time of a single sample is longer. And the focal plane synthesis restriction degree is high, and a plurality of difficulties are faced when images of different focal planes are synthesized into a complete clear image, so that the final imaging effect is influenced. The invention obtains the image data of the sample by driving the imaging camera to shoot the whole sample and the detail structure, wherein the image data comprises a plurality of pictures or a plurality of sections of videos of the sample at different focal planes, and the image data is generated into the synthetic image data of the sample by a preset multi-focal-plane self-adaptive stacking algorithm, and the synthetic image data comprises a plurality of synthetic pictures of the sample so as to solve the problem.

Species morphological identification mainly depends on morphological taxonomy expert, however, the expert talent echelon fault phenomenon is serious, and the micro-anatomical knowledge has higher cognitive barrier, so that the identification work is difficult to be widely and efficiently carried out. Many laboratories are not able to independently perform accurate species identification without the expertise of a person. The invention solves the problem by comparing the synthesized image data with known sample data of a pre-established digital sample library to obtain an intelligent identification result.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following description will briefly explain the drawings needed in the embodiments, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic diagram of a super depth of field synthetic intelligent identification microscopic device in an embodiment of the invention;

FIG. 2 is a schematic diagram of a manipulator according to an embodiment of the present invention;

FIG. 3 is a schematic view of another view of a super depth of field synthetic intelligent authentication microscopy device according to an embodiment of the present invention;

FIG. 4 is a logic schematic diagram of a control unit of a super depth of field program according to an embodiment of the present invention;

fig. 5 is a schematic product diagram of a super depth of field program control unit according to an embodiment of the invention.

In the drawings, a high-definition camera device, a 2, a microscope device, a 3, an upper computer with a super depth of field program control unit, a 4, a light source device, a 5, a Z-axis moving frame, a 6, an X-axis moving object table, a 7, a Y-axis moving object table, an 8, a rotating object table, a 9, a main object table, a 10, a controller, a 11, a brightness adjusting knob, a 12, a Y-axis moving object table button, a 13, an X-axis moving object table button, a 14, a light source switch button, a 15, an XY-axis reset button, a 16, a Z-axis moving frame knob, a 17, a Z-axis reset button, an 18, 360-degree rotating object table knob, a 19, a rotating reset button are shown.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

Examples:

It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.

In the description of the present invention, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise. Furthermore, unless explicitly specified and limited otherwise, the terms "mounted," "connected," "coupled," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected, mechanically connected, electrically connected, directly connected, or indirectly connected via an intervening medium, or may be in communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.

Fig. 1 is a schematic structural diagram of a super-depth-of-field synthetic intelligent identification microscopic device according to an embodiment of the present invention, fig. 2 is a schematic structural diagram of a manipulator according to an embodiment of the present invention, fig. 3 is a schematic structural diagram of another view angle of the super-depth-of-field synthetic intelligent identification microscopic device according to an embodiment of the present invention, fig. 4 is a logic schematic diagram of a super-depth-of-field program control unit according to an embodiment of the present invention, and fig. 4 is a logic schematic diagram of the super-depth-of-field program control unit according to an embodiment of the present invention.

As shown in fig. 1 to 5, an embodiment of the present invention provides a super-depth-of-field synthetic intelligent identification microscopic device, including:

The microscope unit comprises an objective table assembly, a Z-axis moving frame and an objective lens assembly, wherein the objective table assembly is used for placing a sample slide, the Z-axis moving frame is in supporting connection with the objective lens assembly and the objective table assembly, and the objective lens assembly is connected with an ocular lens and an objective lens;

In an embodiment, the method for driving the imaging camera to shoot the image data of the whole and detail structure of the sample specifically comprises the steps that a controller processing module sends shooting distance and shooting speed shooting instructions of the imaging camera for controlling the variable-focus adjusting lens to the high-definition camera unit, and the imaging camera shoots the image data of the whole and detail structure of the sample according to the received shooting instructions.

Specifically, in the super-depth-of-field synthesis intelligent identification microscopic device, the high-definition camera unit consists of a high-resolution imaging camera and a variable-focus regulation lens. The variable focus control lens allows an operator to change the focal length by simply adjusting the focal length ring without changing the lens. Illustratively, when observing the insect specimen, the image of the insect specimen is focused, and the detail of the insect specimen scale or the hair cluster can be clearly presented by rotating the focal length ring. The controller processing module can regulate and control the shooting speed of the high-definition camera unit, and interacts with the hardware of the high-definition camera unit through an internal algorithm and an instruction system to set and adjust the shooting speed and the shooting distance. Illustratively, the high-definition camera unit shoots for 4 seconds at maximum to synthesize a high-definition picture, and the minimum shooting distance of each picture is 5um.

In an embodiment, the method comprises the steps of sliding a preset displacement window on a picture line by line or column by column according to a preset step length, extracting feature information of an image area in the window at each window position, classifying and integrating the extracted feature information to obtain a layered feature map, carrying out quantization analysis on the layered feature map, calculating complexity and size of the layered feature map to obtain details and global information of the layered feature map, automatically judging whether each layered feature map belongs to a high-definition image according to the details and global information of the layered feature map, carrying out self-adaption stacking according to the preset algorithm and a preset high-definition image standard, obtaining a synthesized picture according to the layered feature map belonging to the high-definition image, deleting the current layered feature map and carrying out re-shooting when the matching degree of the photographed current layered feature map and the previous layered feature map is smaller than a preset threshold, carrying out gradual superposition on the current layered feature map and the current layered feature map to obtain the synthesized picture, and carrying out noise reduction processing on all synthesized picture to obtain the synthesized picture.

Specifically, a sample photo taken by the high-definition camera is processed through a displacement window. The displacement window is a small area moving on the image according to a certain rule, and the image is divided by the displacement window to construct a layered characteristic diagram. Thus, different levels and features of the sample can be extracted separately, and each hierarchical feature map represents information of the sample at a specific level or angle. For the constructed hierarchical feature map, the image data processing module calculates its complexity and size. The complexity of the image reflects the richness of the image content and the detail, and the high complexity means that the image contains more detail information. The image size represents the data amount of the image, which is related to factors such as resolution of the image. By calculating these parameters, the features of the image can be fully understood, and the details and global information of the image can be captured. Based on the information of the complexity, the size and the like of the image obtained by the previous calculation, the image data processing module can automatically judge whether the image is a high-definition image or not. If the image is of high enough complexity, contains rich detail, and is of a size that meets certain criteria while also meeting requirements in other respects (e.g., resolution, sharpness, etc.), it is determined to be a high definition image, and conversely, it is not.

In the synthesis process, the image data processing module compares the matching degree of the currently shot layered characteristic diagram and the last layered characteristic diagram. The matching degree is mainly determined by comparing the feature similarity of the two images, such as the similarity degree in texture, color, shape and the like. If the matching degree is small, the feature difference of the two images is larger, the superposition effect is poor, at the moment, the image data processing module can send out an instruction to delete the current layered feature image, and the high-definition camera is controlled again to shoot so as to acquire one layered feature image with higher matching degree with the last image. If the matching degree is high and the superposition effect is good, the image data processing module sends out an instruction to superpose the current layered characteristic image and the previous image, and a complete sample image is gradually constructed.

When all the layered feature maps are overlapped to form a composite map, the composite map may have stains and noise due to the influence of various factors (such as light interference, sensor noise, etc.) during the shooting process. The system automatically performs stain removal and noise reduction on the composite map. The stain elimination is to identify and remove impurities or abnormal pixel points in the image by an algorithm, and the noise reduction treatment is to reduce noise in the image by adopting a specific method (such as a filtering algorithm) so as to make the image clearer and cleaner.

The high-definition sample pictures after stain removal and noise reduction treatment can be automatically stored by the system. The stored images can be conveniently viewed, analyzed and identified later, and can also be stored as data for further study or comparison with other sample images.

In one embodiment, the known sample data of the digital sample library is specifically established by using a photographing device to comprehensively collect images of each known physical sample in the digital sample library, creating a single folder for each sample, collecting layered feature maps of different angles and different feature structures in the folder, and taking the layered feature map of each known physical sample as the known sample data.

In the above embodiment, further, according to the synthetic image data, the intelligent identification result is obtained by comparing the synthetic image data with the known sample data of the pre-established digital sample library, specifically including that according to the hierarchical feature map of the synthetic image data, the comparison result is listed by comparing the synthetic image data with the hierarchical feature map of the known sample data of the image information library of the pre-established digital sample library, and the comparison result is used as the intelligent identification result.

Specifically, using a photographing device, a comprehensive image acquisition is performed on each known physical sample in the digitized specimen library. A separate folder is created for each sample, and 1000 layered feature images of different angles, different features are acquired in each folder. This operation is intended to record as comprehensively as possible the various features of the sample, for example, for an insect sample, taken from different angles and features of front, side, back, and microscopic wing texture, leg joints, etc., to obtain a rich image information.

And extracting the characteristics of the acquired sample picture and quantifying the characteristics into a characteristic matrix. The feature extraction may involve analysis of various visual features such as color, texture, shape, etc., and the quantized feature matrix stores the feature information of the sample in a digitized form, so as to facilitate subsequent processing and comparison. For example, the color features of the sample are converted into specific numerical representations, the texture features calculate corresponding parameters through an algorithm, and the numerical values and the parameters form a feature matrix.

The pictures in each physical sample folder are provided to the intelligent identification module, and the intelligent identification module learns and analyzes a large number of sample pictures. By constantly processing and memorizing the feature matrices of these samples, the intelligent authentication recognition module is increasingly capable of recognizing all known samples in the digitized sample library.

When a high-definition camera shoots a new sample photo, the intelligent identification module firstly performs image acquisition on the sample and constructs a characteristic structure of the sample. This step is similar to the operation in building the image information base, and also extracts various features of the sample to form a unique feature representation of the sample. For example, for an unknown insect sample, after an image is acquired, the characteristics of the overall appearance structure, the shape of the veins, the color and the spots of the body of the insect are extracted, and corresponding characteristic structures are constructed.

The characteristics of the new sample are quickly compared to the known sample characteristics in the digitized sample library. In the comparison process, the sample characteristics are tracked in a digital sample library, and a sample with high matching degree is found out. The process utilizes the characteristic matrix constructed before, and the similarity between samples is judged by calculating the similarity between different characteristic matrices. For example, if a feature of a new sample is very similar in value to a corresponding feature of a sample in the library, then their similarity is high.

And finally, the intelligent identification module enumerates the comparison result according to the comparison and tracking results. These results are typically ranked from high to low in similarity, exhibiting the most similar known samples to the new sample, and the degree of similarity between them. Scientific researchers can judge the information such as the kind, the attribute and the like of the new sample according to the comparison results, and finish the intelligent identification of the sample. For example, the comparison result shows that the similarity between the new sample and one insect sample in the sample library reaches 98-99%, so that the new sample is primarily judged to belong to the sample.

In an embodiment, the super depth program control unit further includes a USB data processing module, where the USB data processing module is configured to convert data transmitted between the digitized sample library and the intelligent authentication and identification module according to a set format.

Specifically, the USB data processing module can convert the instruction sent by the intelligent identification module from the original format into the format which can be understood and processed by the functional layer. Conversely, when the digital specimen library returns data, the USB data processing module also converts the data into a format that can be recognized and used by the smart identification module.

In one embodiment, the stage assembly comprises:

A main stage;

Specifically, the stage assembly is mainly composed of a main stage, an X-axis moving stage, a Y-axis moving stage, a rotary stage, and a Z-axis moving frame. The X-axis moving object stage can horizontally move left and right on the main object stage relative to the main object stage. The Y-axis moving object stage can horizontally move up and down on the main object stage relative to the main object stage. The rotary stage may be rotated 360 degrees on the main stage relative to the main stage level. The rotary stage may be rotated 360 degrees on the main stage relative to the main stage level. The Z-axis moving frame can horizontally and vertically move on the main objective table relative to the main objective table.

In this embodiment, the lifting or moving structure of the X moving stage, the Y moving stage, the rotary stage, and the Z moving frame relative to the main stage may be a rack engaging structure, a track moving structure, or the like, and is driven to move by a driving member such as a motor, which is not described herein.

In one embodiment, a light source device is arranged near the objective lens, and the controller processing module is connected with the light source device in a control signal manner.

Specifically, the light source device realizes stepless adjustment of the light source brightness through PWM (Pulse Width Modulation ) technology, namely, the light source brightness adjustment can be continuously and smoothly changed within a certain range. And the controller processing module can generate an instruction to the light source device according to shooting requirements and real-time image feedback, and the light source device automatically adjusts the brightness of the light source according to the received instruction.

In the above embodiment, further comprising a manipulator, the manipulator is connected with the rotary stage, the X-axis moving stage, the Y-axis moving stage, the Z-axis moving frame and the controller of the light source device by control signals.

Specifically, the manipulator is provided with a plurality of control keys for realizing the operation and adjustment of different parts of the device to meet various requirements of microscopic observation and photographing, for example, an X-axis moving stage button, a Y-axis moving stage button, a Z-axis moving frame knob, a 360-degree rotating stage knob, a rotating reset button, an XY-axis reset button, a Z-axis reset button, a light source switch button, and a brightness adjusting knob.

The X-axis moving object stage button is pressed, the moving object stage can be controlled to horizontally move left and right along the X axis of the main object stage, and the horizontal moving distance is 100mm. And pressing a button of the Y-axis moving object stage, and horizontally moving the moving object stage up and down along the Y axis of the main object stage, wherein the horizontal moving distance is 100mm. The knob of the Z-axis moving frame is rotated, so that the moving object stage can vertically move along the Z axis of the main object stage, and the vertical moving distance is 150mm. Rotating the 360 degree rotary stage knob, the rotary stage can perform 360 degree rotation along the main stage level. The rotary reset button is pressed down, the rotary objective table of the main objective table can be controlled to reset, and the reset precision is +/-0.001 mm. And the XY-axis reset button is pressed down, so that the XY object stage can be controlled to reset, and the reset accuracy is +/-0.001 mm. The Z-axis reset button is pressed down, so that the Z-stage reset can be controlled, and the reset accuracy is +/-0.003 mm. The light source switch button is used for switching on or switching off the light source device of the main object stage. The brightness of the main stage light source device can be controlled by the light source knob (namely the brightness adjusting knob).

In an embodiment, the system further comprises an upper computer, and the controller processing module is connected with the upper computer through control signals.

Specifically, the controller processing module can encode the control movement functions of all control keys in the controller by using an instruction encoding mode, and the instruction encoding is a process of converting the operation of the control keys into numbers or code forms which can be recognized and processed by a computer. The encoded control key operation information can be displayed on a PC display. When the user operates the control keys, corresponding operation information (in coded form) is displayed on the PC display. The user can intuitively see what information is encoded into the current operation of the control key through the PC display. The PC display can not only display the coded information of the control key operation, but also be used for controlling the device. The user can input instructions or select corresponding codes through an interface or related operations on the PC display to realize the control of the device.

It is to be understood that the terms "center," "longitudinal," "transverse," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counter-clockwise," "axial," "radial," "circumferential," and the like are directional or positional relationships as indicated based on the drawings, merely to facilitate describing the invention and to simplify the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore should not be construed as limiting the invention.

In the present invention, unless expressly stated or limited otherwise, a first feature "up" or "down" a second feature may be the first and second features in direct contact, or the first and second features in indirect contact via an intervening medium. Moreover, a first feature being "above," "over" and "on" a second feature may be a first feature being directly above or obliquely above the second feature, or simply indicating that the first feature is level higher than the second feature. The first feature being "under", "below" and "beneath" the second feature may be the first feature being directly under or obliquely below the second feature, or simply indicating that the first feature is less level than the second feature.

In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.

The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement the same, and are not intended to limit the scope of the present invention. All equivalent changes or modifications made in accordance with the essence of the present invention are intended to be included within the scope of the present invention.

Claims

1. A super-depth-of-field synthesis intelligent identification microscope device, characterized by comprising:

The microscope unit includes a stage assembly, a Z-axis movable frame, and a barrel assembly. The stage assembly is used to place a sample. The Z-axis movable frame supports and connects the barrel assembly and the stage assembly. The barrel assembly is connected to an eyepiece and an objective lens.

A high-definition camera unit, comprising an imaging camera with a variable-focus lens;

The super depth of field program control unit includes a controller processing module, an image data processing module and an intelligent identification module, wherein:

The controller processing module is used to drive the imaging camera to capture the overall structure and detailed structure of the sample to obtain image data of the sample, wherein the image data includes multiple pictures or multiple videos of the sample at different focal planes;

The image data processing module is configured to generate, based on the image data, composite image data of the sample using a preset multi-focal plane adaptive stacking algorithm, wherein the composite image data includes multiple composite images of the sample;

The intelligent identification and recognition module is used to obtain an intelligent identification result by comparing the synthetic image data with known sample data in a pre-established digital specimen library.

2. The super-depth synthesis intelligent identification microscope device according to claim 1, wherein driving the imaging camera to capture image data of the sample as a whole and its detailed structure specifically comprises:

The controller processing module sends a shooting instruction for controlling the shooting distance and shooting speed of the imaging camera with a variable-focus lens to the high-definition camera unit;

The imaging camera captures image data of the overall structure and detailed structure of the sample according to the received shooting instruction.

3. The super-depth synthetic intelligent identification microscope device according to claim 1, wherein generating synthetic image data of the sample using a preset multi-focal plane adaptive stacking algorithm based on the image data specifically comprises:

The preset displacement window is slid across the image row by row or column by column according to the set step size. At each window position, the feature information of the image area within the window is extracted, and the extracted feature information is classified and integrated to obtain a hierarchical feature map.

Quantitatively analyze the hierarchical feature maps, calculate their complexity and size, and obtain the details and global information of the hierarchical feature maps. Based on the details and global information of the hierarchical feature maps, use the preset algorithm and the set high-definition image standards to automatically determine whether each hierarchical feature map is a high-definition image;

Adaptively stack the hierarchical feature maps belonging to high-definition images to obtain a composite image. When the matching degree between the current hierarchical feature map and the previous hierarchical feature map is less than a preset threshold, the current hierarchical feature map is deleted and a new photo is taken. When the matching degree between the current hierarchical feature map and the previous hierarchical feature map is higher than a preset threshold, the current hierarchical feature map is superimposed to gradually obtain a composite image.

All composite images are subjected to spot removal and noise reduction processing to obtain composite image data.

4. The super-depth-of-field synthesis intelligent identification microscope device according to claim 1, wherein the known sample data of the digitized specimen library is established by the following steps:

Using photographic equipment, comprehensive image acquisition is performed on each known physical sample in the digital specimen library. A single folder is created for each sample. Multiple hierarchical feature maps with different angles and different feature structures are collected in the folder. The hierarchical feature map of each known physical sample is used as the known sample data, and the known sample data is stored in the image information library.

5. The super-depth-of-field synthesis intelligent identification microscope device according to claim 4 is characterized in that, based on the synthesized image data, by comparing it with known sample data in a pre-established digital specimen library, an intelligent identification result is obtained, specifically comprising:

Based on the hierarchical feature map of the synthetic image data, it is compared with the hierarchical feature map of the known sample data in the image information library of the pre-established digital specimen library, and the comparison results are listed and used as the intelligent identification results.

6. The super-depth-of-field synthesis intelligent identification microscope device according to claim 1 is characterized in that the super-depth-of-field program control unit also includes a USB data processing module, which is used to convert the data transmitted between the digital specimen library and the intelligent identification and recognition module according to a set format.

7. The super-depth-of-field synthetic intelligent identification microscope device according to claim 1, wherein the stage assembly comprises:

main stage;

A rotating stage is arranged on the main stage, and an X-axis moving stage and a Y-axis moving stage are arranged on the rotating stage.

8. The super-depth-of-field synthesis intelligent identification microscope device according to claim 7 is characterized in that a light source device is provided near the objective lens, and the controller processing module is connected to the control signal of the light source device.

9. The super-depth-of-field synthesis intelligent identification microscope device according to claim 8 is characterized in that it also includes a manipulator, and the manipulator is connected to the controller control signal of the rotating stage, X-axis moving stage, Y-axis moving stage, Z-axis moving frame, and light source device.

10. The super-depth-of-field synthesis intelligent identification microscope device according to claim 1 is characterized in that it also includes a host computer, and the controller processing module is connected to the host computer control signal.