Nanoscale metal tip smart detector and methods of use and analysis

Abstract

The application provides a nano-scale metal tip intelligent detector and a use and analysis method thereof. The device integrates a high-precision imaging and positioning module, an intelligent analysis module, an intelligent control interaction module and an automatic execution mechanism. The imaging and positioning module is composed of an optical microscope, a high-resolution CCD camera and a three-axis precision stage, supports nanoscale topographic feature capture and micrometer-level positioning accuracy. The intelligent analysis module can automatically extract key parameters such as curvature radius, cone angle and surface roughness. The control interaction module is coupled with the automatic execution mechanism to realize automatic control of the detection process and remote storage and optimization of process data. The device can automatically complete the detection and intelligent analysis of the nano-scale metal tip morphology, is compatible with single-nanopore / pore channel preparation, scanning probe tip morphology characterization and nanodevice detection and other application scenarios, and significantly improves the work efficiency while maintaining high-precision detection.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of nanomaterial detection, and specifically relates to a metal tip intelligent detector for nanoscale and a use and analysis method thereof, and particularly applies to automatic detection and analysis of metal tip morphology in single nanopore / channel preparation process. BACKGROUND

[0002] Preparation and characterization of nanoscale metal tips are core technologies for constructing single nanopores / channels, and the morphology characteristics and physical properties thereof directly determine the accuracy of ion detection, DNA sequencing, molecular sensing and nanofluidic devices. At present, as a common method for preparing single nanopores / channels, the template method realizes the construction of nanopore structure through precise coupling of a metal needle tip and a glass tube, but still faces three technical bottlenecks in process amplification:

[0003] 1. Lack of evaluation system: the existing detection means mainly rely on manual interpretation of tip scanning electron microscope images, lack of quantifiable three-dimensional morphology evaluation standard, and the standardization degree is insufficient, and the characterization accuracy is easily affected by subjective factors;

[0004] 2. Lack of intelligent detection: the whole process automation degree from image acquisition to feature extraction is low, the traditional equipment lacks intelligent analysis system based on machine vision, the operation process needs professional training and has poor repeatability, and it is difficult to realize high-throughput detection;

[0005] 3. High cost of precise detection: the dependence of nanoscale metal tip detection on high-precision SEM equipment is strong, resulting in high detection cost, complicated sample preparation process, and irreversible damage to the sample in the detection process.

[0006] Therefore, it is urgent to develop a detection device integrating standardized detection and intelligent analysis functions to improve the efficiency and accuracy of nanoscale metal tip morphology detection in single nanopore / channel preparation process, and to provide technical support for scanning probe tip morphology detection and nanodevice manufacturing and other extended fields. SUMMARY

[0007] The present application provides a metal tip intelligent detector for nanoscale and a use and analysis method thereof, which can automatically complete the detection and intelligent analysis of nanoscale metal tip morphology, is compatible with single nanopore / channel preparation, scanning probe tip morphology characterization and nanodevice detection and other application scenarios, significantly improves the work efficiency while maintaining high-precision detection, and effectively reduces the demand for manual intervention.

[0008] The technical scheme of the present application is: a metal tip intelligent detector for nanometer scale, comprising a base (12), one side of the base (12) is provided with a microscope support (2) and a microscope (1) arranged on the microscope support (2); a light source (3) for illuminating a sample is arranged in the middle of the base (12);

[0009] The other side of the base (12) is provided with a stage moving device, which has an XYZ three-axis moving mechanism; the X-axis and the Y-axis control the translational movement of the nanometer metal tip in the microscope, and the Z-axis controls the distance between the sample and the microscope, that is, the image focusing; a placing stage is arranged on the stage moving device; the placing stage comprises a placing platform (4), a connecting support (10) and a sample clamping mechanism; the connecting support (10) is fixedly connected with the stage moving device, the placing platform (4) is arranged on the connecting support (10) and is fixedly connected with the connecting support (10); a wire clamping shell (5) passes through the placing platform (4) and the sample clamping mechanism from top to bottom, and the lower end of the wire clamping shell (5) is located in the lens direction of the microscope (1).

[0010] An outer shell (7) is arranged above the aforementioned stage moving device, and a control module is arranged in the outer shell (7).

[0011] Alternatively, in view of the detection efficiency bottleneck, measurement accuracy limitation and excessive dependence on manual intervention in the prior art, the present application provides a nanometer metal tip intelligent detection and analysis system, which integrates four core components of high-precision imaging and positioning module, intelligent analysis module, intelligent control interaction module and automatic execution mechanism.

[0012] The high-precision imaging and positioning module comprises a microscopic imaging unit, an automatic stage and a light source system, wherein:

[0013] The microscopic imaging unit is composed of a microscope and a high-resolution CCD camera, which supports nanoscale feature capture; the automatic stage is equipped with a three-axis precision motion platform, which realizes micron-level positioning accuracy through the cooperation of a stepping motor drive and a ball screw transmission system, wherein the X / Y dual-axis linkage regulates the displacement trajectory of the nanometer metal tip in the microscope image plane, and the Z-axis controls the distance between the objective lens and the sample, that is, the dynamic focusing control; the light source system selects a back-illuminated LED light source to eliminate metal surface glare interference.

[0014] The intelligent analysis module is equipped with a Yolov10 real-time detection framework and supports morphological processing and image recognition fusion technology optimization to realize nanotip target recognition and micron-level positioning; the improved U-Net architecture is used for extracting tip curvature radius, cone angle, roughness and stability and other contour shape parameters to improve detection accuracy.

[0015] The intelligent control interaction module comprises a master control unit, a man-machine interaction interface and a cloud data management system, wherein:

[0016] The master control unit adopts an STM32F103RCT6 microcontroller, integrates a motor drive and a wireless communication module, and supports collaborative control of multiple devices; the man-machine interaction interface integrates a parameter visual configuration matrix and a real-time panoramic image stream; and the cloud data management system realizes synchronous storage of detection data and remote process iteration optimization.

[0017] The automatic execution mechanism comprises a vertical sample groove and a limiting packaging structure, a mechanical clamping mechanism driven by a steering engine, and a rod-shaped metal sample clamping shell, supports rapid loading and unloading of batch samples, and is compatible with material spectra such as platinum, tungsten and platinum-iridium alloy, and detection scales cover nanometers to millimeters.

[0018] In some embodiments, the control and interaction module is built-in with an ESP8266 Wi-Fi module, can realize remote control through an Internet of Things MQTT protocol, and uploads detection data to the cloud in real time.

[0019] In some embodiments, the light source system can adjust the light source brightness through a microcontroller to adapt to various metals and clamping shells.

[0020] In some embodiments, the intelligent analysis module integrates an abnormality processing mechanism, including tip detection failure retry, focusing failure retry and profile abnormality prompt.

[0021] In some embodiments, the control module comprises a microscope, transmits the captured metal tip image to a computer, displays the tip morphology image in real time on the man-machine interaction interface, uses image algorithms combined with deep learning to perform tip target detection and morphology analysis, generates a detection result containing curvature radius, cone angle, roughness and stability parameters, and pushes the detection result to a cloud database.

[0022] Compared with prior art solutions, the present application realizes breakthrough innovation through deep integration of a multi-dimensional technical system: relying on a full-process automated detection platform and an intelligent morphology analysis algorithm, a high-precision, non-human intervention detection closed loop is constructed; a multi-precision mapping model is used to break through the performance limitations of traditional microscopes and realize standardized quantitative evaluation of nanoscale morphology; through a reconfigurable modular architecture and a multi-scenario adaptation strategy, single-nanopore / pore channel preparation, scanning probe tip morphology detection and nanodevice manufacturing full-process can be compatible, forming a cross-field, high-robustness intelligent detection solution. BRIEF DESCRIPTION OF DRAWINGS

[0023] The present application will be further described below in conjunction with the drawings and examples.

[0024] Figure 1: The overall structure of the detection analyzer shows the layout of the microscopic imaging unit, the automatic stage and the control module.

[0025] Figure 2 : The schematic diagram of the top view of the clamping mechanism.

[0026] Figure 3 : The fitting result graph of the third-order function of the tip edge.

[0027] Figure 4 : The function block diagram of the control system describes the logical relationship between data acquisition, algorithm processing and device control.

[0028] Figure 5 : The schematic diagram of the scanning path of the microscope.

[0029] Figure 6 : The flowchart of the intelligent analysis algorithm illustrates the serial architecture of the U-Net topographic parameter calculation.

[0030] Reference signs

[0031] 1, microscope; 2, microscope support; 3, LED light source; 4, object platform; 5, metal wire clamping shell; 6, Y-axis screw sliding table; 7, shell; 8, X-axis screw sliding table; 9, Z-axis screw sliding table; 10, connecting support; 11, steering engine; 12, base; 13, clamping platform; 13-1, convex rib; 14, steering engine connecting rod; 14-1, center connecting rod; 14-2, side rod; 15, clamping connecting rod; 16, clamping jaw. DETAILED DESCRIPTION

[0032] The technical solutions of the present application will be described in detail below in combination with the drawings, and the implementation details and technical advantages of the present application will be further illustrated through specific examples, which do not constitute a limitation on the scope of protection.

[0033] Example 1: Hardware structure configuration

[0034] As shown in Figure 1 and Figure 2 , the present application is a metal tip intelligent detector for nanoscale, which includes a base 12. One side (the front half side in Figure 1 ) of the base 12 is provided with a microscope support 2 and a microscope 1 arranged on the microscope support 2. The lens of the microscope 1 is arranged from front to back.

[0035] The middle (center in Figure 1 ) of the base 12 is provided with an LED light source 3, which includes an LED support and an LED lamp bead, perpendicular to the lens direction of the microscope 1.

[0036] The other side (in Figure 1The rear half of the middle section is equipped with a platform moving device, which consists of three axes: X, Y, and Z. The Z-axis is composed of two parallel front-to-back directions ( Figure 1 The system consists of three slides: an X-axis slide 8, which is placed horizontally on top of the Z-axis slide 9 and can move back and forth along the slide 9; and a Y-axis slide 6, which is placed vertically and connected to the slider of the X-axis slide 8 and can move horizontally along the X-axis slide 8. A platform is connected to the front of the slider of the Y-axis slide 6, and the platform can move vertically along the Y-axis slide 6. All three axes (X, Y, and Z) are driven by motors or can be manually adjusted.

[0037] The storage platform consists of a storage platform 4, a connecting bracket 10, a servo motor 11, a clamping platform 13, a servo motor linkage 14, a clamping linkage 15, and a gripper 16 (e.g., ...). Figure 2 (As shown). The connecting bracket 10 is fixedly mounted on the front of the Y-axis slide 6, and the placement platform 4 is fixed to the upper half of the front of the connecting bracket 10, arranged horizontally. The clamping platform 13 is parallel to the lower part of the placement platform 4 and is fixedly connected to the connecting bracket 10. The servo motor 11 passes through the clamping platform 13, and its output shaft is located above the clamping platform 13 and connected to the servo motor linkage 14.

[0038] The servo linkage 14 consists of a central linkage 14-1 and two side linkages 14-2 pivotally connected to both ends of the central linkage 14-1. The output shaft of the servo 11 is fixedly connected to the middle portion of the central linkage 14-1, thereby allowing the central linkage 14-1 to rotate clockwise or counterclockwise. The other end of each side linkage 14-2 is pivotally connected to one end of a clamping linkage 15. The other end of the clamping linkage 15 is fixedly connected to a gripper 16. There is a pair of clamping linkages 15 and grippers 16.

[0039] The edge of clamping platform 13 ( Figure 1 On the left side, Figure 2 The lower side of the clamping link 15 has a raised ridge 13-1, and the back of the clamping link 15 has a groove corresponding to the raised ridge 13-1. The raised ridge 13-1 and the groove cooperate so that the clamping link 15 can only slide along the raised ridge 13-1. Figure 1 For sliding in the forward and backward direction, Figure 2 (Slide left or right).

[0040] In use, the servo motor 11 rotates, causing the central connecting rod 14-1 to rotate around its center point, which in turn drives the two side rods 14-2 and the clamping connecting rod 15 to move. Due to the restriction of the protruding ridge 13-1, in Figure 2 In this configuration, the two clamping links can only move closer or further apart relative to each other along the left-right direction of the convex ridge 13-1. Consequently, the pair of grippers 16 also move closer to each other to complete clamping or further apart to complete disengagement.

[0041] The metal wire clamping shell 5 (where the metal wire clamp to be tested is clamped) passes through the placement platform 4 from top to bottom and is inserted between the clamping positions of the two clamping jaws 16. After the clamping jaws 16 are closed and clamped and fixed, their movement is limited. The lower end of the metal wire clamping shell 5 is located in the lens direction of the microscope 1.

[0042] An outer shell 7 is mounted above the aforementioned object table moving device to protect the lead screw sliding table and place the control module (microcontroller).

[0043] Further, in the above structure:

[0044] 1. Automatic stage module

[0045] (1) Three-axis displacement mechanism:

[0046] A high-strength aluminum alloy base is adopted, the X / Y axis stroke is 0-100 mm, the Z axis stroke is 0-50 mm, and a high-precision ball screw (lead 1 mm) is provided;

[0047] The driving unit selects a 42-step motor (step angle 1.8°), matched with a 16-subdivision driver, and the repeated positioning error is ≤0.2 mm;

[0048] Motion control logic: adjust the motor speed by sending pulse signals (frequency 1-10 kHz) through the STM32 microcontroller, which can achieve variable speed scanning of 0.1-5 mm / s.

[0049] (2) Limiting packaging structure:

[0050] The mechanical clamping device is driven by a single-axis servo motor, and the maximum clamping force can reach 2 N·m, which is suitable for metal wire clamping shells with a diameter of 5-20 mm.

[0051] 2. Microscopic imaging module

[0052] The optical microscope is configured with a 1x-4x objective lens, and the CCD camera has a resolution of 1920x1080 pixels and a frame rate of 30 fps;

[0053] Light source system: integrate LED light source (brightness adjustable) at the back of the sample and opposite the microscope lens to ensure that the image presents large contrast and is used to eliminate metal reflection interference.

[0054] 3. Control and communication module

[0055] Main control chip: STM32F103RCT6 microcontroller, mounted on a four-layer PCB circuit board, integrated with motor drive chip and ESP8266 Wi-Fi module;

[0056] Power management: input voltage 12V / 5A, output 5V / 3A through a step-down chip, peak power consumption ≤15W;

[0057] Communication protocol: support MQTT protocol, real-time upload of detection data to the cloud (Ali Cloud IoT platform), transmission delay <200 ms.

[0058] Example 2: Manual acquisition of tip morphology for model training

[0059] Step 1: Image acquisition. Manually adjust the position of the metal wire, use the microscope / SEM to acquire the image of the metal wire tip, and save it as the target image after cropping and binaryzation preprocessing; wherein the metal wire position can be adjusted in XYZ axis, and combined with the height of the clamping position of the metal wire to adjust it to the appropriate field of view of the microscope.

[0060] Step 2: Edge extraction. For the image acquired by the microscope, according to the magnification, extract the front 5 pixel points and the edge of the front 10 μm (currently 50 pixel points) of the two edges of the metal tip, and the front 5 pixel points of the tip are used for morphology feature calculation; for the image acquired by SEM, according to the magnification, extract the front 5 pixel points and the edge of the front 1 μm (currently 150 pixel points) of the two edges of the metal tip, and use them for morphology feature calculation;

[0061] Step 3: Turning point determination. The pixel points of the upper and lower edges of the tip are fitted by a third-order function, and the curvature turning point is found, as shown in Figure 3 The calculation formula is as follows:

[0062] f(x)=ax 3 +bx 2 +cx+d

[0063] f″(x)=6ax+2b

[0064] f″(x)=0

[0065] Solve the point where the second derivative is 0, which is the curvature change point. Fit all the pixel points between the tip and the turning point by a first-order function, and the calculation formula is as follows:

[0066] f(x)=ax+b

[0067] Where f(x) is the fitting function, a, b, c, d are the fitting parameter values of the function. If there is no curvature turning point, it means that the edge is a straight line, and the midpoint of the edge is taken as the curvature turning point.

[0068] Step 4: Morphology feature calculation. Calculate the included angle of the first-order function fitting line of the two edges as the metal tip included angle, calculate the radius of the fitting circle of the front 5 pixel points of the two edges as the metal tip curvature radius, and the first-order function fitting degree R 2 of the above pixel points as the roughness, and the variation coefficient of the above features represents the etching stability, and the calculation formula is as follows:

[0069]

[0070] Stability = 0.5 x anglecv + 0.3 x radiuscv + 0.2 x (1 - R 2 )

[0071] where std is the standard deviation of multiple experiments, n is the sample number, yi is the true observed value, the average of true observed values, is the predicted value, cv is the coefficient of variation, anglecv is the coefficient of variation of the tip angle, radiuscv is the coefficient of variation of the radius of curvature, R 2 is the function fitting coefficient, and Stability is the comprehensive stability index.

[0072] Step 5: Deep learning model training. The acquired images and topographic features are used to train the deep learning model.

[0073] Example 3: Detection control flow

[0074] Step 1: System initialization. The stage performs origin reset, and the X / Y / Z axes successively touch the limit switches and then retreat to the origin; the microscope switches to a 3x low-power objective, the mechanical gripper automatically opens to the maximum opening (20 mm), and waits for sample loading;

[0075] Step 2: Sample loading. The clamping shell holding the metal wire tip (diameter 25 μm) is vertically inserted into the circular slot of the sample plate, and the half-moon slot gripper is closed by manually triggering through the interactive interface or automatically triggering after the clamping shell is inserted using an automated device.

[0076] Step 3: Scanning parameter setting. The scanning parameters are set through the human-machine interface as global positioning + local fine feature extraction, the X-axis scanning step is 500 μm, the Y-axis scanning step is 800 μm, the Z-axis automatic focusing range is ±2 mm, and the LED light source brightness is set to 15% brightness.

[0077] Step 4: Local pre-scanning. The X / Y axes are scanned in a snake-like path ( Figure 5 ), images are collected and real-time binarization processing is performed; the area of the black region in each frame of image is calculated, and the corresponding motion ID and area value are recorded; after local pre-scanning is completed, the stage is moved to the ID corresponding to the maximum area, and the automatic focusing algorithm is started. The Z-axis is moved forward and backward with a step size of 100 μm, the definition evaluation function (sobel operator, laplacian operator, and Canny operator mixed definition) of each frame of image is calculated, and when the definition reaches the peak value, the process is stopped and the last position of pre-scanning is returned to.

[0078] Step 5: Metal tip target detection. A three-step positioning strategy is adopted Figure 4

[0079] (1) Cluster denoising: K-means clustering (K=3) is performed on the high magnification image, and the largest connected domain is retained. The sum of squares of the distances of all sample points in a cluster to the centroid is calculated by Euclidean distance, and the solution of the minimum overall sum of squares is solved, and the center of each cluster is recalculated, and the clustering result is iteratively optimized, so that the objects in each cluster are as close as possible, and the objects between different clusters are as far apart as possible. The formula is as follows:

[0080]

[0081] Where x is the sample point in the cluster, ci is the index of the cluster to which the data point xi belongs, μj is the centroid of the jth cluster, K is the number of clusters, Sj is the data point set of the jth cluster, and |Sj| is the number of data points in the set.

[0082] (2) Target recognition: morphological calculation and Yolov10 model (input size 512x512, confidence threshold 0.5) are used to locate the tip area, output the boundary box coordinates, and the confidence calculation formula is as follows:

[0083]

[0084] Confidence=P(Object)*IoU(A,B)

[0085] Where A is the target frame, B is the predicted frame, P is the probability of the existence of objects in the predicted frame, IoU is the intersection over union, Confidence is the confidence, and is used to calculate the ratio of the intersection and union of the predicted frame and the real frame area.

[0086] (3) Precise positioning: according to the center coordinates of the boundary frame, the ΔX and ΔY of the moving stage are calculated, the tip is moved to the center of the field of view, the second focusing is performed and the image is collected, and the calculation formula is as follows:

[0087] △X=Xbox_centroid-Xpic_centroid+Xmove

[0088] △Y=Ybox_centroid-Ypic_centroid+Ymove

[0089] Where Xbox_centroid and Ybox_centroid are the center point coordinates of the predicted frame, Xpic_centroid and Ypic_centroid are the center point coordinates of the current acquired image, Xmove and Ymove are the acquisition positions corresponding to the image, and the positions are calculated by the corresponding motion ID.​

[0090] Step 6: Morphology output. After image preprocessing, the improved U-Net model is input to output the tip morphology characteristics, including the radius of curvature, cone angle, roughness and stability, and upload the data to the cloud. Figure 6

[0091] Example 4: Batch detection and abnormal processing

[0092] Batch loading: 12 samples are sequentially grabbed by a mechanical arm into a vertical sample slot, and each sample detection period is ≤30 seconds;

[0093] Tip detection failure retry: if the morphological and Yolo algorithms do not detect a metal tip, re-detect, up to 3 retries;

[0094] Focus failure retry: when the focus clarity peak is < threshold, the Z-axis scanning range is expanded to ±3mm, up to 3 retries;

[0095] Profile abnormality prompt: if the detected tip is classified as unqualified, prompt and record the sample as an unqualified sample, and exclude it in subsequent work.

[0096] Implementation effect:

[0097] Tested 20 groups of metal wire (platinum wire) metal tips, compared with SEM measurement results:

[0098]

[0099] The results show that in this embodiment, the application maintains a high fitting degree at a low precision, and the efficiency is more than 10 times higher than that of the traditional SEM. It also supports continuous placement and clamping of metal tips by a mechanical arm, realizes batch automatic operation, and is suitable for single nanometer hole / channel preparation, scanning probe morphology detection and other scenes, significantly reduces production cost and improves process consistency.

[0100] The above is only a preferred embodiment of the application, and therefore cannot limit the scope of the application. Equivalent changes and modifications made in accordance with the scope and content of the application should still be within the scope of the application.​

Claims

1. A nano-scale metal tip smart detector, characterized in that: The base (12) is provided with a microscope support (2) on one side and a microscope (1) on the microscope support (2); a light source (3) for illuminating the sample is arranged in the middle of the base (12); The other side of the base (12) is provided with a stage moving device, which has an XYZ three-axis moving mechanism; the X-axis and Y-axis control the translational motion of the nanometer metal tip in the field of view of the microscope, and the Z-axis controls the distance between the sample and the microscope, that is, the image focusing; the stage moving device is provided with a placing table; the placing table includes a placing platform (4), a connecting support (10), and a sample clamping mechanism; the connecting support (10) is fixedly connected with the stage moving device, and the placing platform (4) is arranged on the connecting support (10) and fixedly connected with the connecting support (10); a wire clamping shell (5) passes through the placing platform (4) and the sample clamping mechanism from top to bottom, and the lower end of the wire clamping shell (5) is located in the lens direction of the microscope (1); A housing (7) is arranged above the stage moving device, and a control module is arranged in the housing (7); The sample clamping mechanism includes a steering wheel (11), a clamping platform (13), a steering wheel connecting rod (14), a clamping connecting rod (15), and a clamping jaw (16); the clamping platform (13) is fixedly connected with the connecting support (10), the steering wheel (11) is arranged on the clamping platform (13), and the output shaft of the steering wheel (11) is located above the clamping platform (13) and connected with the steering wheel connecting rod (14); The steering wheel connecting rod (14) includes a central connecting rod (14-1) in the middle and two side rods (14-2) pivotally connected at both ends of the central connecting rod (14-1); the output shaft of the steering wheel (11) is fixedly connected with the middle part of the central connecting rod (14-1), so that the central connecting rod (14-1) rotates clockwise or counterclockwise; the other end of each side rod (14-2) is pivotally connected with one end of the clamping connecting rod (15); the other end of the clamping connecting rod (15) is fixedly connected with the clamping jaw (16); A limiting structure is arranged between the clamping platform (13) and the clamping connecting rod (15) to enable the two clamping connecting rods to slide relatively in one direction to approach or separate.

2. A nanoscale metal tip smart detector as claimed in claim 1, wherein: The Z-axis of the stage moving device includes two Z-axis sliding tables (9) arranged in parallel in the front and back directions; an X-axis sliding table (8) is arranged transversely in the left and right directions on the Z-axis sliding table (9) and can move forward and backward along the Z-axis sliding table (9); a Y-axis sliding table (6) is arranged vertically and connected with the sliding block of the X-axis sliding table (8), and the Y-axis sliding table (6) can move left and right along the X-axis sliding table (8); the sliding block of the Y-axis sliding table (6) is connected with the placing table on the front side, and the placing table can move up and down along the Y-axis sliding table (6).

3. The nanometer-scale metal tip intelligent detector according to claim 1, characterized in that: The edge of the clamping platform (13) is provided with a convex rib (13-1) upward, the back of the clamping connecting rod (15) is provided with a groove corresponding to the convex rib (13-1), and the convex rib (13-1) and the groove cooperate to enable the clamping connecting rod (15) to slide only along the convex rib (13-1).

4. A nanoscale metal tip smart detector as claimed in claim 1, wherein: Also include intelligent analysis module, the intelligent analysis module is based on deep learning algorithm, adopts Yolov10 model to realize cutting-edge target detection, extracts cutting-edge curvature radius, angle and contour shape parameter in combination with improved U-Net architecture, and supports morphological processing and image fusion technology optimization feature extraction.

5. A nanoscale metal tip smart detector as claimed in claim 1, wherein The microscope (1) is composed of a lens and a high-resolution CCD camera supporting nanoscale image acquisition.

6. A nanoscale metal tip smart detector as claimed in claim 1, wherein The XYZ three-axis moving mechanism is driven by a stepping motor.

7. A method for using and analyzing a nanoscale metal tip intelligent detector according to any one of claims 1 to 6, comprising the following steps: Step 1: system initialization, the stage performs origin reset, and the X / Y / Z axes touch the limit switches in turn and then retreat to the origin; the microscope switches to a 3x low-power objective, the clamping jaw is automatically opened to the maximum opening, and the sample is loaded; Step 2: sample loading, the metal wire clamping shell clamping the metal wire tip is vertically inserted into the circular slot of the sample platform, and the clamping jaw is closed manually through the interactive interface or automatically after the metal wire clamping shell is inserted using an automatic device; Step 3: scanning parameter setting, the scanning parameters are set to global positioning and local fine feature extraction through the human-computer interface, the X-axis scanning step is 500 μm, the Y-axis scanning step is 800 μm, the Z-axis automatic focusing range is ±2 mm, and the LED light source brightness is set to 15% brightness; Step 4: local pre-scanning, the X / Y axes are scanned in a snake-shaped path, images are collected and real-time binarization processing is performed; the area of the black region in each frame of image is calculated, and the corresponding motion ID and area value are recorded; after the local pre-scanning is completed, the stage is moved to the ID corresponding to the maximum area, and the automatic focusing algorithm is started; the Z-axis moves forward and backward with a step of 100 μm, the definition evaluation function of each frame of image is calculated, and when the definition reaches the peak value, the process is stopped and the last position of the pre-scanning is returned; Step 5: metal tip target detection; a three-step positioning strategy is adopted: (1) clustering denoising: K-means clustering is performed on the high-power lens image, K=3, and the largest connected domain is retained; the sum of squares of the distances of all sample points in a cluster to the center is calculated by Euclidean distance, the solution of the minimum overall sum of squares is solved, and the center of each cluster is recalculated, and the clustering result is continuously iteratively optimized, so that the objects in each cluster are close, and the objects in different clusters are separated, and the formula is as follows: Wherein, x is a sample point in the cluster, c i is the data point x i is the index of the cluster to which x belongs, μ j is the centroid of the jth cluster, K is the number of clusters, S j is the set of data points of the jth cluster, |S j is the number of data points in this set; (2) target recognition: morphological calculation and Yolov10 model are used to jointly locate the tip area, and the boundary box coordinates are output, and the confidence calculation formula is as follows: Confidence=P(Object)*IoU(A,B) Wherein, A is the target boundary box, B is the predicted boundary box, P is the probability of the existence of the object in the predicted box, IoU is the intersection over union, and Confidence is the confidence, which is used to calculate the ratio of the intersection to the union of the areas of the predicted box and the real box; (3) accurate positioning: according to the center coordinates of the boundary box, ΔX and ΔY required for the movement of the stage are calculated, the tip is moved to the center of the field of view, secondary focusing is performed, and an image is collected, and the calculation formula is as follows: ΔX = X box_centroid - X pic_centroid + X move ΔY = Y box_centroid - Y pic_centroid + Y move wherein X box_centroid and Y box_centroid are the center point coordinates of the predicted bounding box, X pic_centroid and Y pic_centroid are the center point coordinates of the current acquired image, and X move and Y move are the acquisition position corresponding to the image, which is calculated according to the corresponding motion ID. Step 6: Morphology output; Morphology includes radius of curvature, cone angle, roughness and stability.

8. A method of using and analyzing a nanoscale metal tip smart detector as claimed in claim 7, wherein, Also includes the following batch detection and abnormal handling steps: Batch loading: multiple samples are grabbed by the mechanical arm to the vertical sample slot one by one, and each sample detection period is less than or equal to 30 seconds; Tip detection failure retry: if the morphology and Yolo algorithm do not detect the metal tip, re-detect, up to 3 retries; Focus failure retry: when the focus clarity peak is less than the threshold, the Z-axis scanning range is expanded to ±3mm, up to 3 retries; Profile abnormal prompt: if the detected tip is classified as unqualified, prompt and record the sample as unqualified sample, and exclude it in subsequent work.

Images