Micro-target detecting and positioning method based on laser radar

Abstract

The invention discloses a micro-target detection positioning method based on a laser radar, and relates to the technical field of radar micro-target detection, and the method comprises the steps: obtaining first data of a to-be-detected micro-target, and carrying out the preprocessing of the first data; based on the preprocessed first data, constructing a multi-branch deep learning model and performing feature extraction on the first data; performing detection and material classification according to a feature extraction result; and presetting a dynamic feature library, and carrying out online learning on the unknown material according to a feature extraction result. According to the method, the multi-branch deep learning model is constructed to fuse geometric and spectral features, the micro-target detection precision and the material classification accuracy are remarkably improved, automatic recognition of unknown materials is achieved by means of a dynamic feature library and an online learning mechanism, special feature extraction branches are designed for multi-wavelength laser radar data, and the detection accuracy is improved. The material specificity information is fully excavated, and a micro-target detection positioning solution with high precision, high adaptability and low maintenance cost is provided.

Description

Technical Field

[0001] The present invention relates to the technical field of micro-target detection, and particularly to a micro-target detection and positioning method based on lidar. Background Art

[0002] With the development of industrial automation, intelligent equipment and remote sensing detection technologies, the demand for micro-target detection and positioning in fields such as high-precision manufacturing, autonomous driving, and drone inspection is increasing day by day. As a high-precision active remote sensing device, lidar can obtain the three-dimensional coordinate information of targets and multi-wavelength reflection intensity data, becoming one of the core technologies for micro-target detection. By analyzing the spatial structure features and spectral reflection characteristics in the point cloud data, the existence judgment, material classification, and precise positioning of micro-targets can be realized.

[0003] However, the common solutions currently available have many drawbacks, including: traditional methods mostly use a single-branch network to process point cloud data, and cannot simultaneously fuse the spatial structure features and spectral material features of micro-targets. When only relying on geometric features, it is easy to miss micro-targets with extremely small point cloud numbers, and it is impossible to distinguish targets with similar materials but similar geometric structures; when only relying on spectral features, it is difficult to locate the spatial position of the target, and it is significantly affected by lidar noise. Summary of the Invention

[0004] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this part, as well as in the abstract and title of the specification of this application, to avoid obscuring the purpose of this part, the abstract, and the title. However, such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the problems existing in the above-mentioned existing lidar-based micro-target detection and positioning methods, the present invention is proposed.

[0006] Therefore, the purpose of the present invention is to provide a lidar-based micro-target detection and positioning method, which is suitable for solving the problems that traditional point cloud analysis only extracts geometric information such as three-dimensional coordinates and point cloud density, there is a risk of missing micro-targets with sparse point clouds or complex spatial structures, and the multi-wavelength reflection intensity data is not effectively utilized, resulting in low material classification accuracy. A small number of multi-modal fusion methods use fixed weights or simple concatenation, and do not adaptively learn the complementary relationship between geometric and spectral features according to the characteristics of micro-targets, making it difficult to meet the complex detection requirements of multi-material and multi-scale targets.

[0007] To solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, an embodiment of the present invention provides a method for detecting and positioning micro-targets based on lidar, which includes obtaining first data of the micro-targets to be measured and preprocessing the first data; based on the preprocessed first data, constructing a multi-branch deep learning model and extracting features from the first data; performing detection and material classification according to the feature extraction results; presetting a dynamic feature library and performing online learning on unknown materials according to the feature extraction results.

[0008] As a preferred solution of the method for detecting and positioning micro-targets based on lidar according to the present invention, wherein: the first data is point cloud data including multi-wavelength reflection intensity and three-dimensional coordinates; the first data is obtained by a multi-wavelength lidar; the multi-branch deep learning model is used to extract features from the first data; the dynamic feature library includes known material templates and an unknown sample temporary storage area.

[0009] As a preferred solution of the method for detecting and positioning micro-targets based on lidar according to the present invention, wherein: constructing a multi-branch deep learning model based on the preprocessed first data includes the following steps: dividing the geometric branch and the spectral branch according to data characteristics and clarifying the functions of each branch; the geometric branch uses the PointNet point cloud network to extract geometric features; the spectral branch uses CNN to extract spectral material features; a multi-branch deep learning model is constructed by fusing the features output by the geometric branch and the spectral branch.

[0010] As a preferred solution of the method for detecting and positioning micro-targets based on lidar according to the present invention, wherein: the specific formula of the multi-branch deep learning model is as follows: , wherein, is the fused feature vector; is the weight coefficient of the geometric feature; is the center point of the geometric feature vector; is the spectral feature vector.

[0011] As a preferred solution of the method for detecting and positioning micro-targets based on lidar according to the present invention, wherein: the micro-target detection according to the feature extraction results includes the following steps: calculating the modulus length of the fused feature vector; the micro-target detection according to the feature extraction results includes the following steps: when the modulus length of the fused feature vector is greater than the first threshold, it indicates that the micro-target exists, and the target has significant geometric structure features and spectral anomaly features, record the position of the area where the target is located, and trigger the material classification process; when the modulus length of the fused feature vector is less than or equal to the first threshold, it indicates that the target does not exist, record the current area as the background, do not trigger the material classification process, move the detection window to the next area, and repeat the feature extraction and threshold judgment.

[0012] As a preferred solution of the lidar-based micro-target detection and positioning method of the present invention, the specific steps of the material classification process are as follows: Calculate the similarity between the spectral feature vector and the known material templates in the dynamic feature library; if the maximum similarity is greater than the second threshold, the material of the micro-target is a known material category, and the material classification result is directly output; if the maximum similarity is less than or equal to the second threshold, it indicates that the spectral features of the micro-target are not sufficiently matched with all known material templates in the dynamic feature library and cannot be accurately classified into any known material, indicating that the material is an unknown material, and the unknown material processing process is performed.

[0013] As a preferred solution of the lidar-based micro-target detection and positioning method of the present invention, the specific steps of the unknown material processing process are as follows: Extract the spectral feature vector and geometric features of the micro-target and mark them as unknown samples; Add the unknown samples to the dynamic feature library and store them separately from the known material templates; Prompt the operator to manually confirm the unknown samples and label the material tags for the unknown samples; Store the spectral feature vector corresponding to the labeled unknown samples and the material tags in the known material template area of the dynamic feature library.

[0014] In a second aspect, to further solve the above technical problems, an embodiment of the present invention provides a lidar-based micro-target detection and positioning system, which includes: a data acquisition module for acquiring the first data of the target to be measured and performing preprocessing; a model construction module for constructing a multi-branch deep learning model and extracting features from the first data; a result processing module for performing detection and material classification according to the feature extraction results; a model update module for performing online learning on unknown materials according to the feature extraction results.

[0015] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: When the computer program is executed by the processor, it implements any step of the lidar-based micro-target detection and positioning method as described in the first aspect of the present invention.

[0016] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and: When the computer program is executed by the processor, it implements any step of the lidar-based micro-target detection and positioning method as described in the first aspect of the present invention.

[0017] The beneficial effects of the present invention are as follows: By constructing a multi-branch deep learning model to fuse geometric and spectral features, the present invention significantly improves the micro-target detection accuracy and material classification accuracy, and solves the problem of incomplete description of single-modal features; With the help of a dynamic feature library and an online learning mechanism, automatic recognition of unknown materials is realized, significantly reducing the manual annotation cost and shortening the adaptation period of new materials; By fusing feature threshold linkage detection and classification processes, the real-time processing efficiency is optimized and the robustness in complex environments is enhanced; At the same time, a dedicated feature extraction branch is designed for multi-wavelength lidar data to fully exploit material-specific information, providing a micro-target detection and positioning solution with high precision, high adaptability, and low maintenance cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. Among them: Figure 1 It is a flowchart for the implementation of the present invention in Embodiment 1.

[0019] Figure 2 It is a dynamic adjustment diagram of the multi-branch deep learning model in Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will make a detailed description of the specific embodiments of the present invention with reference to the accompanying drawings of the specification.

[0021] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0022] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that excludes other embodiments.

[0023] Embodiment 1 Referring to Figure 1 and Figure 2 , this is the first embodiment of the present invention. This embodiment provides a micro-target detection and positioning method based on lidar, including the following steps: S1: Obtain the first data of the micro-target to be measured, and preprocess the first data.

[0024] Specifically, the first data is point cloud data containing multi-wavelength reflection intensity and three-dimensional coordinate information.

[0025] Furthermore, the first data is obtained by a multi-wavelength lidar.

[0026] Furthermore, the three-dimensional coordinate information supports the spatial positioning and existence judgment of the target, and the multi-wavelength reflection intensity supports material classification and feature differentiation extraction.

[0027] In the embodiment of the present application, the preprocessing of the multi-wavelength reflection intensity is to eliminate the influence of noise and equipment differences and enhance the relative reflection characteristics of the material to different wavelengths; the preprocessing of the three-dimensional coordinate information is spatial structure noise reduction and positioning accuracy optimization.

[0028] Preferably, by performing spatial structure noise reduction on the three-dimensional coordinate information, eliminating noise and equipment difference effects on the multi-wavelength reflection intensity, reducing interference factors in the data, making the point cloud data more accurately reflect the micro-target characteristics, the preprocessing of the three-dimensional coordinates optimizes the positioning accuracy, supports the spatial positioning and existence judgment of the target; the preprocessing of the multi-wavelength reflection intensity strengthens the relative reflection characteristics of the material to different wavelengths, improves the reliability of material classification and feature differentiation extraction, provides a high-quality data basis for subsequent micro-target detection, recognition and other tasks, and the preprocessing operation makes the data format more standardized, facilitating efficient processing by subsequent algorithms or models, enhancing data compatibility and task processing efficiency.

[0029] S2: Based on the preprocessed first data, construct a multi-branch deep learning model and perform feature extraction on the first data.

[0030] It should be noted that the multi-branch deep learning model is used to perform feature extraction on the first data. Feature extraction.

[0031] Specifically, based on the preprocessed first data, constructing a multi-branch deep learning model includes the following steps: dividing the geometric branch and the spectral branch according to the data characteristics, and clarifying the functions of each branch.

[0032] The geometric branch uses the PointNet point cloud network to extract geometric features. The specific formula is as follows: , In the formula, is the geometric feature vector of the center point , describing the local spatial structure information; is the three-dimensional coordinate of the center point, used to calculate the relative position relationship with the neighboring points; is The coordinates of the j-th nearest neighbor point, providing the spatial distribution information of the local neighborhood; k is the number of nearest neighbors in the nearest neighbor algorithm, used to control the size of the local neighborhood; is the nearest neighbor algorithm, the set of k nearest neighbor points of.

[0033] The spectral branch uses CNN to extract spectral material features, and the specific formula is as follows: , In the formula, is the spectral feature vector, used to distinguish materials; M is the number of spectral bands, determining the dimension of the spectral features; is the weight of the m-th band, adjusting the contribution of each band to the final feature; is the reflectance of the m-th band, representing the reflection ability of the object at the wavelength.

[0034] By fusing the features output by the geometric branch and the spectral branch, a multi-branch deep learning model is constructed.

[0035] Specifically, the specific formula of the multi-branch deep learning model is as follows: , In the formula, is the fused feature vector, concatenating all dimensions of the geometric and spectral features; is the weight coefficient of the geometric feature, controlling the relative importance of the two types of features, automatically optimized by model training; is the center point of the geometric feature vector; is the spectral feature vector.

[0036] It should be noted that if the geometric feature is more critical in the detection task, the model will automatically increase ; if the material classification is the focus, the model will automatically decrease , enhancing the weight of the spectral feature.

[0037] It should be noted that PointNet is a deep learning network suitable for point cloud data, mainly used to extract geometric features, and CNN is a convolutional neural network, mainly used to extract spectral material features.

[0038] In the embodiments of the present application, the point cloud data has disorder and sparsity. PointNet can effectively capture the three-dimensional geometric structure by aligning the relative coordinates of the nearest neighbor points. The spectral data has inter-band correlation and spatial locality. The convolutional operation of CNN can extract local spectral patterns through a sliding window, and the weights can adaptively learn the importance of different bands.

[0039] S3: Perform micro-target detection and material classification according to the feature extraction results.

[0040] Preferably, micro-target detection based on the feature extraction results includes the following steps: Calculate the norm of the fused feature vector; When the norm of the fused feature vector is greater than the first threshold, it indicates the existence of a micro-target, and the target has significant geometric and spectral features. Record the position of the target area and trigger the material classification process.

[0041] When the norm of the fused feature vector is less than or equal to the first threshold, it indicates the non-existence of the target. Record the current area as the background, do not trigger the material classification process, move the detection window to the next area, and repeat feature extraction and threshold judgment.

[0042] Specifically, the dynamic adjustment steps of the first threshold are as follows: Continuously collect the fused feature data of the most recent time, calculate the average value and the fluctuation range of the fused feature data, and avoid the failure of the fixed threshold due to environmental changes or changes in target characteristics by dynamically counting real-time data.

[0043] Based on the average value and the fluctuation range in the real-time data, dynamically generate the detection threshold for the current frame. When the data fluctuates greatly, the threshold is automatically increased to filter noise; when the data is stable, the threshold is kept moderate to balance the detection sensitivity.

[0044] If noise is misjudged as a target for 5 consecutive frames, it means the threshold is too low. Automatically increase the threshold to strict detection standards and reduce misjudgment.

[0045] If a real target is missed for 5 consecutive frames, it means the threshold is too high. Automatically lower the threshold to relax the detection standards and avoid missed detection.

[0046] It should be noted that limit the amplitude of each threshold adjustment to avoid frequent oscillation of the detection results.

[0047] Specifically, the specific steps of the material classification process are as follows: Calculate the similarity between the spectral feature vector and the known material templates in the dynamic feature library. The specific formula is as follows: , In the formula, is the similarity between the current micro-target spectral feature and the i-th known material template; is the spectral feature vector of the micro-target, extracted by the CNN network of the spectral branch; is the template vector of the i-th known material in the dynamic feature library.

[0048] If the maximum similarity is greater than the second threshold, the material of the micro-target is the known material category, and directly output the material classification result.

[0049] If the maximum similarity If it is less than or equal to the second threshold, it indicates that the spectral characteristics of the micro-target do not match well with all known material templates in the dynamic feature library and cannot be accurately classified as any known material, indicating that the material is an unknown material, and the unknown material processing process is carried out.

[0050] It should be noted that the second threshold is obtained by setting based on the similarity characteristics of known material templates and combining the accuracy requirements of the classification task.

[0051] Preferably, micro-target detection is carried out by fusing geometric and spectral features, combined with a dynamic threshold adjustment mechanism to adapt to environmental changes, effectively improving the detection robustness and reducing false positives and false negatives; material classification uses spectral features as the dominant weight and combines geometric auxiliary judgment to improve the material recognition accuracy.

[0052] S4: Preset a dynamic feature library and perform online learning on the unknown material according to the feature extraction results.

[0053] Preferably, the specific steps of the unknown material processing process are as follows: Extract the spectral feature vector and geometric features of the micro-target and label it as an unknown sample; Add the unknown sample to the dynamic feature library and store it separately from the known material templates; Prompt the operator to manually confirm the unknown sample and label the material label for the unknown sample; Store the spectral feature vector corresponding to the labeled unknown sample and the material label in the known material template area of the dynamic feature library.

[0054] Preferably, automatically store the spectral and geometric features of the unknown sample and manage them in partitions. Incorporate new materials into the known templates through manual annotation and incremental training, enabling the model to adapt to the detection requirements of new materials without offline retraining. This mechanism significantly reduces the manual annotation cost, shortens the adaptation cycle of new materials, and improves the model generalization ability by continuously expanding the feature library, especially suitable for industrial detection scenarios where the types of materials are frequently updated.

[0055] Exemplarily, point cloud data containing three-dimensional coordinates and dual-band reflection intensity is obtained by a multi-wavelength lidar. After preprocessing such as three-dimensional noise reduction and spectral reflectance enhancement, it is input into a multi-branch model composed of a PointNet geometric branch and a CNN spectral branch to fuse and generate a comprehensive feature vector. Micro-scale metal particles are accurately detected by dynamically adjusting the first threshold, and material classification is achieved based on cosine similarity. When unknown copper-nickel alloy particles are detected, features are automatically stored, manually labeled, and then the dynamic feature library is updated, triggering online learning of the model, shortening the adaptation period of new materials from 2 weeks in the traditional method to 2 hours, and improving the particle detection rate and material classification accuracy to 96% and 95% respectively, significantly enhancing the defect detection and process traceability efficiency in semiconductor manufacturing.

[0056] In summary, the present invention constructs a multi-branch deep learning model to fuse geometric and spectral features, significantly improving the micro-target detection accuracy and material classification accuracy, and solving the problem of incomplete description of single-modal features; with the help of a dynamic feature library and an online learning mechanism, it realizes the automatic recognition of unknown materials, greatly reducing the manual labeling cost and shortening the adaptation period of new materials; by fusing the feature threshold linkage detection and classification processes, it optimizes the real-time processing efficiency and enhances the robustness in complex environments; at the same time, a dedicated feature extraction branch is designed for multi-wavelength lidar data to fully explore the material-specific information, providing a micro-target detection and positioning solution with high precision, high adaptability, and low maintenance cost.

[0057] Embodiment 2 This embodiment also provides a micro-target detection and positioning system based on lidar, including: a data acquisition module for acquiring first data of a target to be measured and performing preprocessing; a model construction module for constructing a multi-branch deep learning model and extracting features from the first data; a result processing module for performing detection and material classification according to the feature extraction results; and a model update module for performing online learning on unknown materials according to the feature extraction results.

[0058] Embodiment 3 This embodiment provides a computer device applicable to the case of a micro-target detection and positioning method based on lidar, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the micro-target detection and positioning method based on lidar proposed in the above embodiment.

[0059] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0060] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for micro-target detection and positioning based on lidar as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM for short), Electrically Erasable Programmable Read-Only Memory (EEPROM for short), Erasable Programmable Read-Only Memory (EPROM for short), Programmable Read-Only Memory (PROM for short), Read-Only Memory (ROM for short), magnetic memory, flash memory, a magnetic disk, or an optical disc.

[0061] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A method for detecting and positioning micro-targets based on lidar, characterized in that: Including: Obtain the first data of the micro-target to be measured, and preprocess the first data; Based on the preprocessed first data, construct a multi-branch deep learning model and extract features from the first data; Perform detection and material classification according to the feature extraction results; Preset a dynamic feature library, and perform online learning on unknown materials according to the feature extraction results.

2. The method for micro-target detection and positioning based on lidar according to claim 1, wherein: The first data is point cloud data containing multi-wavelength reflection intensity and three-dimensional coordinates; The first data is obtained by a multi-wavelength lidar; The multi-branch deep learning model is used to extract features from the first data; The dynamic feature library includes known material templates and an unknown sample temporary storage area.

3. The method for micro-target detection and positioning based on lidar according to claim 2, wherein: Based on the preprocessed first data, constructing a multi-branch deep learning model includes the following steps: Divide the geometric branch and the spectral branch according to data characteristics, and clarify the functions of each branch; The geometric branch uses the PointNet point cloud network to extract geometric features; The spectral branch uses a CNN to extract spectral material features; Construct a multi-branch deep learning model by fusing the features output by the geometric branch and the spectral branch.

4. The method for micro-target detection and positioning based on lidar according to claim 3, wherein: The specific formula of the multi-branch deep learning model is as follows: ， In the formula, is the fused feature vector; is the weight coefficient of the geometric feature; is the center point of the geometric feature vector; is the spectral feature vector.

5. The method for micro-target detection and positioning based on lidar according to claim 1, characterized in that: Performing micro-target detection according to the feature extraction results includes the following steps: Calculate the modulus of the fused feature vector; When the modulus of the fused feature vector is greater than the first threshold, it indicates that the micro-target exists, and the target has significant geometric structure features and spectral anomaly features. Record the position of the target area and trigger the material classification process; When the modulus of the fused feature vector is less than or equal to the first threshold, it indicates that the target does not exist. Record the current area as the background, do not trigger the material classification process, move the detection window to the next area, and repeat feature extraction and threshold judgment.

6. The method for detecting and positioning micro-targets based on lidar according to claim 5, wherein: The specific steps of the material classification process are as follows: Calculate the similarity between the spectral feature vector and the known material templates in the dynamic feature library; If the maximum similarity is greater than the second threshold, the material of the micro-target is the known material category, and directly output the material classification result; If the maximum similarity is less than or equal to the second threshold, it indicates that the spectral features of the micro-target do not match well with all known material templates in the dynamic feature library and cannot be accurately classified as any known material, indicating that the material is an unknown material, and perform the unknown material processing process.

7. The method for micro-target detection and positioning based on lidar according to claim 6, characterized in that: The specific steps of the unknown material processing process are as follows: Extract the spectral feature vector and geometric features of the micro-target and mark them as unknown samples; Add the unknown samples to the dynamic feature library and store them separately from the known material templates; Prompt the operator to manually confirm the unknown samples and label the material labels for the unknown samples; Store the spectral feature vector and material label corresponding to the labeled unknown samples in the known material template area of the dynamic feature library.

8. A lidar-based micro-target detection and positioning system, based on the lidar-based micro-target detection and positioning method according to any one of claims 1 to 7, characterized in that: Including, A data acquisition module, which is used to acquire the first data of the target to be measured and perform preprocessing; A model construction module, which is used to construct a multi-branch deep learning model and extract features from the first data; A result processing module, which is used to perform detection and material classification according to the feature extraction results; A model update module, which is used to perform online learning on unknown materials according to the feature extraction results.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, the steps of the lidar-based micro-target detection and positioning method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the lidar-based micro-target detection and positioning method according to any one of claims 1 to 7 are implemented.

Images