Target detection method and system for DNA nanostructure in AFM scene

Abstract

The invention discloses a DNA nanostructure-oriented target detection method and system in an AFM scene, and belongs to the technical field of image recognition. The target detection method comprises the following steps: acquiring an image data set of a DNA nanostructure in an AFM scene; performing data annotation and enhancement on a target AFM image in the image data set to form a target AFM image data set; based on a YOLOv5 model, establishing a fusion target detection unit; training a fusion target detection unit through the target AFM image data set to obtain different target detection models; and selecting a corresponding trained target detection model as an optimal model according to the current scene and demand, and inputting a to-be-detected AFM image into the target detection model for image target recognition to obtain a target detection result. Therefore, the target object in the image can be accurately recognized, the target detection time is greatly shortened, the flexibility of system selection is enhanced, and the recognition accuracy is remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the field of image recognition, and in particular relates to a target detection method and system for DNA nanostructures in an AFM scenario. Background Art

[0002] DNA origami is an efficient bottom-up technique for building arbitrary predefined two-dimensional (2D) and three-dimensional (3D) structures. DNA origami involves a long backbone chain that is assembled into a pre-designed shape with the help of hundreds of shorter chains. A single origami structure can be connected with other structures to form a larger and complex compound, just like bricks connected together. The size of origami ranges from nanometers (single origami) to micrometers (origami compounds).

[0003] In 1986, Binning, Gerber, and Quate first proposed atomic force microscopy (AFM) based on scanning tunneling microscopy (STM). This method can be used to observe the morphology of nanostructures and detect the mechanical properties of samples. AFM "contacts" the surface of origami with a sensitive probe to characterize its shape and mechanical properties. With the development of atomic force microscopy and DNA nanotechnology, this method has become the standard for DNA detection. To identify DNA origami, it should be tightly attached to a flat substrate (such as a mica plate) in a suitable buffer solution. Appropriate scanning probes, AFM holders, and scanning procedures are essential for origami imaging. After nearly 30 years of development of atomic force microscopy, various DNA structures have been observed. There are now a large number of AFM origami images that can be used as training data for deep learning origami detection and characterization.

[0004] In the field of nanotechnology. Although deep learning methods have been used to detect nanostructures, they are still rarely used to detect DNA origami because the environment of origami detection is complex. In view of this, a target detection method for DNA nanostructures in an AFM scenario is proposed. Summary of the invention

[0005] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a target detection method and system for DNA nanostructures in an AFM scenario, which solves the problems in the prior art.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A target detection method for DNA nanostructures in an AFM scenario includes the following steps:

[0008] Acquire image datasets of DNA nanostructures in AFM scenarios;

[0009] Performing data annotation and enhancement on the target AFM image in the image data set to form a target AFM image data set;

[0010] Based on the YOLOv5 model, a fusion target detection unit is established;

[0011] The fusion target detection unit is trained by using the target AFM image data set to obtain different target detection models;

[0012] According to the current scene and requirements, a corresponding trained target detection model is selected as the best model, and the AFM image to be detected is input into the target detection model for image target recognition to obtain the target detection result.

[0013] Furthermore, the step of obtaining the image data set is:

[0014] S11, determining a plurality of AFM images with DNA nanostructures that need to be collected by presetting required data samples;

[0015] S12, preprocessing a target number of AFM images to obtain an image data set.

[0016] Furthermore, the step of preprocessing the AFM image includes:

[0017] S31, reading image data from a data set;

[0018] S32, scaling the original image to a specified size according to the input requirements of the model;

[0019] S33, image data augmentation operations, including random cropping, random flipping, and random rotation;

[0020] S34, normalizing the image.

[0021] Furthermore, the network structure of the YOLOv5 model includes: a backbone network, a feature pyramid network and multiple prediction heads;

[0022] The backbone network adopts the CSPDarknet structure, which includes multiple CSP modules, each of which is composed of multiple convolutional layers and residual connections, and can effectively extract low-level and mid-level feature information of the image;

[0023] The feature pyramid network is set after the backbone network to fuse feature information of different scales; FPN upsamples and fuses feature maps of different resolutions output by the backbone network;

[0024] The prediction head sets multiple prediction heads corresponding to feature maps of different scales to predict the category probability of the target, the position and size information of the bounding box; each prediction head maps the feature map to the category and position space of the target through convolutional layer and fully connected layer operations to obtain the final detection result.

[0025] Furthermore, the specific steps of training the fusion target detection unit are:

[0026] S41, dividing the target AFM image dataset into a training set, a test set, and a validation set;

[0027] S42, configuring model training parameters in the fusion target detection unit;

[0028] S43, during the training process, the target detection model and the complete target detection model are trained and iteratively optimized according to the configured model training parameters;

[0029] The target detection model realizes the detection and positioning functions of the target, which is implemented using the YOLOv5 model; the complete target detection model is a target detection system with complete functions and processes, which includes not only the core fusion target detection unit, but also data preprocessing, model evaluation, optimization and integration with actual application scenarios;

[0030] S44, by monitoring the training process, inputting the validation set for validation, and measuring the performance of the model on the validation set by calculating the evaluation index, observing the performance of the object detection model and the complete object detection model on the validation set;

[0031] S45, after model training and tuning are completed, the final performance of the model is evaluated using the test set data;

[0032] S46, when the target detection model and the complete target detection model achieve satisfactory performance, save the corresponding trained target detection model and the complete target detection model.

[0033] Furthermore, the target detection result includes a bounding box, a category and a confidence level of the detected target object.

[0034] A target detection system for DNA nanostructures in an AFM scenario, comprising:

[0035] Image acquisition module: acquires image datasets of DNA nanostructures in AFM scenarios;

[0036] Image annotation module: annotate and enhance the target AFM image in the image dataset to form a target AFM image dataset;

[0037] Detection unit building module: Based on the YOLOv5 model, a fusion target detection unit is established;

[0038] Detection unit training module: training the fused target detection unit through the target AFM image data set to obtain different target detection models;

[0039] And, target detection module: select a corresponding trained target detection model as the best model according to the current scene and requirements, and input the AFM image to be detected into the target detection model for image target recognition to obtain the target detection result.

[0040] A computer storage medium stores a readable program, which can execute the above-mentioned target detection method for DNA nanostructures in an AFM scenario when the program is running.

[0041] An electronic device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other through the communication bus;

[0042] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the target detection method for DNA nanostructures in the above-mentioned AFM scenario.

[0043] A computer program product includes computer instructions, wherein the computer instructions instruct a computing device to execute operations corresponding to the above-mentioned target detection method for DNA nanostructures in an AFM scenario.

[0044] Beneficial effects of the present invention:

[0045] The present invention established a high-quality AFM image dataset by scanning a variety of DNA nanostructure pictures with different site information in AFM image scenes, and successfully constructed a feature extraction and recognition model for DNA nanostructures in AFM image scenes based on YOLOv5, so as to achieve good recognition performance in the recognition task of DNA nanostructures in AFM image scenes. The present invention combines deep learning with machine vision technology to construct a new target detection method for DNA nanostructures in AFM image scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0047] Figure 1It is an execution flow chart of the target detection method for DNA nanostructure in the AFM image scene of the present invention;

[0048] Figure 2 Schematic diagram of AFM images of DNA nanostructures with different site information of the present invention;

[0049] Figure 3 is a schematic diagram of an AFM image in the target detection method of the present invention;

[0050] Figure 4 It is a marking schematic diagram in the target detection method of the present invention;

[0051] Figure 5 It is a schematic diagram of the file content annotated in the target detection method of the present invention;

[0052] Figure 6 Schematic diagram of the training process curve of the target detection method of the present invention;

[0053] Figure 7 It is a schematic diagram of the detection results of the target detection method of the present invention. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] Example 1

[0056] like Figure 1 As shown, a target detection method for DNA nanostructures in an AFM scenario includes the following steps:

[0057] S1, obtain the image dataset of DNA nanostructure in AFM scene;

[0058] The steps to obtain the image dataset include:

[0059] S11, determining a plurality of AFM images with DNA nanostructures that need to be collected by presetting required data samples;

[0060] S12, preprocessing a target number of AFM images to obtain an image data set;

[0061] Among them, the process of AFM image preprocessing includes:

[0062] 1) Image reading: reading image data from the dataset.

[0063] 2) Image scaling: According to the input requirements of the model, the original image is scaled to the specified size.

[0064] 3) Data enhancement: In order to increase the diversity of data and improve the generalization ability of the model, a series of data enhancement operations are performed on the images, including random cropping, random flipping, random rotation, etc.

[0065] 4) Normalization: In order to make the distribution of AFM image data more consistent with the training requirements of the model and accelerate the convergence of the model, the image is normalized.

[0066] S2, annotating and enhancing the target AFM image in the image data set to form a target AFM image data set;

[0067] S3, based on the YOLOv5 model, establish a fusion target detection unit;

[0068] The fusion target detection unit is established based on the YOLOv5 model; the network structure of the YOLOv5 model includes: a backbone network, a feature pyramid network and multiple prediction heads;

[0069] The backbone network adopts the CSPDarknet structure, which includes multiple cross-stage local modules (CSP), each of which is composed of multiple convolutional layers and residual connections, and can effectively extract low-level and mid-level feature information of the image, providing a rich semantic basis for subsequent target detection;

[0070] The feature pyramid network uses a feature pyramid network (FPN) after the backbone network to fuse feature information of different scales; the FPN upsamples and fuses feature maps of different resolutions output by the backbone network, so that the generated feature pyramid contains both rich semantic information and high-resolution detail information, thereby better detecting objects of different sizes;

[0071] The prediction head sets multiple prediction heads corresponding to feature maps of different scales, which are used to predict information such as the category probability of the target, the position and size of the bounding box; each prediction head maps the feature map to the category and position space of the target through operations such as convolutional layers and fully connected layers to obtain the final detection result.

[0072] S4, training the fusion target detection unit by using the target AFM image data set to obtain different target detection models;

[0073] The specific steps for training the fusion target detection unit are:

[0074] S41, dividing the target AFM image dataset into a training set, a test set, and a validation set according to a set ratio (8:1:1);

[0075] S42, configuring the YOLOv5 model training parameters in the fusion target detection unit, including the pre-training weight path, the model configuration file path, the parameter file path corresponding to the data set, the number of training rounds, the amount of input data for each batch, etc.;

[0076] S43, during the training process, the target detection model and the complete target detection model are trained and iteratively optimized according to the configured model training parameters;

[0077] The target detection model realizes the detection and positioning functions of the target, and is implemented using the YOLOv5 model. It can find the area where the target object is located in the input image and determine which type of object it belongs to. The complete target detection model is a target detection system with complete functions and processes, which not only includes the core fusion target detection unit, but also covers data preprocessing, model evaluation, optimization, and integration with actual application scenarios. The iterative optimization specifically uses data enhancement optimization, increasing training rounds and data volume to optimize the model.

[0078] S44, by monitoring the training process, inputting the validation set for validation, measuring the performance of the model on the validation set by calculating various evaluation indicators (such as accuracy, recall, mean average precision, confusion matrix, etc.), and observing the performance of the object detection model and the complete object detection model on the validation set;

[0079] S45, the final performance evaluation is achieved through the test set test. The test set is a data set used to objectively and impartially evaluate the final performance of the model after model training and tuning. The evaluation indicators of the validation set are similar to those of the test set. By calculating various evaluation indicators (such as accuracy, recall, average precision mean, intersection-over-union ratio, etc.) on the test set, the quality of the model can be accurately evaluated, providing a basis for the application and promotion of the model.

[0080] S46, when the target detection model and the complete target detection model achieve satisfactory performance, save the corresponding trained target detection model and the complete target detection model.

[0081] S5, selecting a corresponding trained target detection model as the best model according to the current scene and requirements, and inputting the AFM image to be detected into the target detection model for image target recognition to obtain a target detection result;

[0082] The target detection result includes the bounding box, category and confidence level of the detected target object.

[0083] Based on similar inventive concepts, an embodiment of the present invention further provides a computer storage medium storing a readable program, which, when running, can execute the above-mentioned target detection method for DNA nanostructures in an AFM scenario.

[0084] Based on similar inventive concepts, an embodiment of the present invention provides an electronic device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other through the communication bus;

[0085] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the target detection method for DNA nanostructures in the above-mentioned AFM scenario.

[0086] Based on similar inventive concepts, an embodiment of the present invention further provides a computer program product, including computer instructions, which instruct a computing device to execute operations corresponding to the above-mentioned target detection method for DNA nanostructures in an AFM scenario.

[0087] Example 2

[0088] In this embodiment, a target detection method proposed by the present invention is described through a specific example; the method comprises the following steps:

[0089] 1. Establishment of data set: The present invention uses AFM to scan a large number of high-quality photos of DNA nanostructures with different site information, annotates the data through labelme software, and produces four data sets of DNA nanostructures with different site information.

[0090] 1) Image acquisition.

[0091] After screening and eliminating incomplete and unclear images, 123 AFM images with a total of about 2,100 DNA nanostructure samples were finally selected to construct a dataset containing DNA nanostructures with different site information, such as Figure 2 , Figure 3 shown.

[0092] 2) Data labeling.

[0093] Labelme software was used to label the target category data. During labeling, the smallest circumscribed rectangular frame of the DNA nanostructure with different site information was selected as the true frame to reduce the background pixels within the true frame to ensure the consistency of the labeling.

[0094] In this dataset, DNA nanostructures with different site information are uniformly annotated as classes, “0”, “1”, “2”, and “3”, and the classes are named “00”, “01”, “10”, and “11”. The annotation diagram and annotation file are as follows Figure 4 and Figure 5 shown.

[0095] 2. Construction of a DNA nanostructure recognition model in AFM image scenarios based on YOLOv5L.

[0096] YOLOv5L was used to successfully build a feature extraction and recognition model for DNA nanostructures with different site information. Experimental results show that the optimized model achieved higher precision, recall and mean average precision (mAP@0.5) on DNA nanostructures with different site information, significantly improving the recognition performance. Figure 6 As shown, it can be seen that the improvement trajectory of the model performance during the training process is shown. With the increase of the training cycle, the training loss gradually decreases, and shows a similar downward trend on the validation set, which indicates that the model is constantly learning and gradually improving its prediction ability, achieving an effective convergence model; by comparing and analyzing the values ​​of each loss function on the training set and the validation set, it can be found that the loss curves of the two always maintain a similar trend during the training process, without significant separation, which indicates that the simulation has not been overfitted.

[0097] 1) Model construction.

[0098] YOLO stands for you only look once, which means that you only need to browse once to identify the category and location of the object in the image. The YOLO series of algorithms adopts an end-to-end processing idea, which combines the two steps of candidate box prediction and target classification into one, thus cleverly transforming the target detection problem into a regression problem.

[0099] The present invention uses the YOLOv5L model, which has an efficient backbone network: it adopts the CSPDarknet structure and effectively reduces the amount of calculation through the design of a cross-stage local network, while enhancing the network's feature extraction capability, which can better capture low-level and intermediate feature information in the image, and provide a richer semantic basis for subsequent target detection. And it has better performance in multi-scale feature fusion: its feature pyramid network (FPN) can fuse feature information of different scales, generate feature pyramids with rich semantic information and different receptive fields, thereby improving the detection accuracy of targets of different sizes. Compared with some traditional target detection models, YOLOv5L can more comprehensively utilize the multi-scale information of the image and better detect small, medium and large targets.

[0100] 2) Model optimization.

[0101] Configure the model training parameters in the target detection system and input the training set into the fusion target detection unit.

[0102] During the training process, the target detection model and the complete target detection model are trained and iteratively optimized according to the configured model training parameters. The training process is monitored, and the validation set is input for verification to observe the performance of the target detection model and the complete target detection model on the validation set.

[0103] 3. Testing of target detection methods for DNA nanostructures in AFM image scenarios.

[0104] The model is tested using a test set containing 18 images. The ideal detection effect is as follows Figure 7 shown.

[0105] The present invention scanned a variety of DNA nanostructure images with different site information in AFM image scenes, and annotated them using Labelme software to establish a high-quality AFM image dataset; based on YOLOv5L, a feature extraction and recognition model for DNA nanostructures in AFM image scenes was successfully constructed. The experimental results show that the model has achieved good recognition performance in the recognition task of DNA nanostructures in AFM image scenes, with an average accuracy of 98.5%.

[0106] The present invention combines deep learning with machine vision technology to construct a new target detection method for DNA nanostructures in AFM image scenarios.

[0107] Example 3

[0108] Based on the target detection method for DNA nanostructures in an AFM scenario proposed in Example 1, in this embodiment, a target detection system for DNA nanostructures in an AFM scenario is proposed, including:

[0109] Image acquisition module: acquires image datasets of DNA nanostructures in AFM scenarios;

[0110] Image annotation module: annotate and enhance the target AFM image in the image dataset to form a target AFM image dataset;

[0111] Detection unit building module: Based on the YOLOv5 model, a fusion target detection unit is established;

[0112] Detection unit training module: training the fused target detection unit through the target AFM image data set to obtain different target detection models;

[0113] And, target detection module: select a corresponding trained target detection model as the best model according to the current scene and requirements, and input the AFM image to be detected into the target detection model for image target recognition to obtain the target detection result.

[0114] The method of the present invention may be implemented in hardware, firmware, or as software or computer code that may be stored in a recording medium (such as a CDROM, RAM, floppy disk, hard disk, or magneto-optical disk), or as computer code that is originally stored in a remote recording medium or a non-temporary machine-readable medium downloaded over a network and will be stored in a local recording medium, so that the method described herein may be stored in such software processing on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It is understood that a computer, processor, microprocessor controller, or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by a computer, processor, or hardware, the method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the method shown herein, the execution of the code converts the general-purpose computer into a dedicated computer for executing the method shown herein.

[0115] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A target detection method for DNA nanostructures in an AFM scenario, characterized in that: The following steps are involved: Acquire image datasets of DNA nanostructures in AFM scenarios; Performing data annotation and enhancement on the target AFM image in the image data set to form a target AFM image data set; Based on the YOLOv5 model, a fusion target detection unit is established; The fusion target detection unit is trained by using the target AFM image data set to obtain different target detection models; According to the current scene and requirements, a corresponding trained target detection model is selected as the best model, and the AFM image to be detected is input into the target detection model for image target recognition to obtain the target detection result.

2. The target detection method for DNA nanostructures in an AFM scenario according to claim 1, characterized in that: The steps of obtaining the image data set are: S11, determining a plurality of AFM images with DNA nanostructures that need to be collected by presetting required data samples; S12, preprocessing a target number of AFM images to obtain an image data set.

3. The target detection method for DNA nanostructures in an AFM scenario according to claim 2, characterized in that: The steps of preprocessing the AFM image include: S31, reading image data from a data set; S32, scaling the original image to a specified size according to the input requirements of the model; S33, image data augmentation operations, including random cropping, random flipping, and random rotation; S34, normalizing the image.

4. The target detection method for DNA nanostructures in an AFM scenario according to claim 1, characterized in that: The network structure of the YOLOv5 model includes: a backbone network, a feature pyramid network and multiple prediction heads; The backbone network adopts the CSPDarknet structure, which includes multiple CSP modules, each of which is composed of multiple convolutional layers and residual connections, and can effectively extract low-level and mid-level feature information of the image; The feature pyramid network is set after the backbone network to fuse feature information of different scales; FPN upsamples and fuses feature maps of different resolutions output by the backbone network; The prediction head sets multiple prediction heads corresponding to feature maps of different scales to predict the category probability of the target, the position and size information of the bounding box; each prediction head maps the feature map to the category and position space of the target through convolutional layer and fully connected layer operations to obtain the final detection result.

5. The target detection method for DNA nanostructures in an AFM scenario according to claim 1, characterized in that: The specific steps of training the fusion target detection unit are: S41, dividing the target AFM image dataset into a training set, a test set, and a validation set; S42, configuring model training parameters in the fusion target detection unit; S43, during the training process, the target detection model and the complete target detection model are trained and iteratively optimized according to the configured model training parameters; The target detection model realizes the detection and positioning functions of the target, which is implemented using the YOLOv5 model; the complete target detection model is a target detection system with complete functions and processes, which includes not only the core fusion target detection unit, but also data preprocessing, model evaluation, optimization and integration with actual application scenarios; S44, by monitoring the training process, inputting the validation set for validation, and measuring the performance of the model on the validation set by calculating the evaluation index, observing the performance of the object detection model and the complete object detection model on the validation set; S45, after model training and tuning are completed, the final performance of the model is evaluated using the test set data; S46, when the target detection model and the complete target detection model achieve satisfactory performance, save the corresponding trained target detection model and the complete target detection model.

6. The target detection method for DNA nanostructures in an AFM scenario according to claim 1, characterized in that: The target detection result includes a bounding box, a category, and a confidence level of the detected target object.

7. A target detection system for DNA nanostructures in an AFM scenario, characterized in that: include: Image acquisition module: acquires image datasets of DNA nanostructures in AFM scenarios; Image annotation module: annotate and enhance the target AFM image in the image dataset to form a target AFM image dataset; Detection unit building module: Based on the YOLOv5 model, a fusion target detection unit is established; Detection unit training module: training the fused target detection unit through the target AFM image data set to obtain different target detection models; And, target detection module: select a corresponding trained target detection model as the best model according to the current scene and requirements, and input the AFM image to be detected into the target detection model for image target recognition to obtain the target detection result.

8. A computer storage medium storing a readable program, characterized in that: When the program is running, it can execute the target detection method for DNA nanostructure in an AFM scenario as described in any one of claims 1 to 6.

9. An electronic device, characterized in that: include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the target detection method for DNA nanostructures in an AFM scenario as described in any one of claims 1-6.

10. A computer program product comprising computer instructions, characterized in that: The computer instructions instruct the computing device to execute operations corresponding to the target detection method for DNA nanostructures in an AFM scenario as described in any one of claims 1-6.

Images