High-speed trace enhancement extraction method based on fixation scanning

By employing gaze scanning technology and a trace enhancement method optimized for the FPGA platform, the problems of high computational complexity and noise interference in existing technologies have been solved, achieving efficient and accurate trace detection.

CN118982483AInactive Publication Date: 2024-11-19CHANGCHUN UNIV OF SCI & TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411068934.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-06
Publication Date
2024-11-19
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing trace detection technologies require significant computational resources and time, and are susceptible to noise interference in complex environments, resulting in low detection accuracy and efficiency.

Method used

A high-speed trace enhancement extraction method based on gaze scanning is adopted. Multi-angle images are acquired by stepper motor and industrial camera. Histogram equalization enhancement, median filtering denoising, 12-direction Sobel edge detection and RANSAC stitching algorithm are combined and hardware acceleration and optimization are performed using FPGA platform.

Benefits of technology

It enables high-quality and rapid extraction of trace images in complex backgrounds, reducing trace loss and information blurring, and improving detection efficiency and accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118982483A_ABST
    Figure CN118982483A_ABST
Patent Text Reader

Abstract

The invention belongs to the field of criminal technology science, and particularly relates to a high-speed trace enhancement extraction method based on gaze scanning, which comprises the following steps: S1, acquiring four images with traces at different angles based on a gaze scanning technology; s2, histogram equalization enhancement and median filtering denoising are carried out on the four collected trace images; the method improves the subjectivity and time consumption of manual operation, greatly improves the trace detection efficiency and accuracy, and provides powerful technical support for the detection of criminal cases. By combining an image processing algorithm with trace detection, the advantages of computer vision and an image processing technology can be brought into full play, and the feasibility and reliability of trace detection are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of criminal technology science, and in particular to a high-speed trace enhancement extraction method based on gaze scanning. Background Art

[0002] The core task of criminal technology is to extract physical evidence from the scene and ensure its clarity, accuracy, real-time and objectivity. In this process, it is crucial to process the interference components of the evidence image. Current trace detection technology mainly detects physical evidence traces through machine vision and chemical analysis. Traces are detected and identified by training machine learning models. However, this method requires a large amount of labeled data for training, and the generalization ability of the model may be limited. The chemical analysis-based method detects and identifies traces by analyzing the chemical composition in the traces. However, this method requires a deep understanding of the chemical composition of the traces, and the analysis process may be complicated and time-consuming.

[0003] Trace detection usually requires high-resolution images or data to accurately detect and identify traces. However, existing trace detection algorithms usually require a lot of calculations and processing, which may lead to high computational complexity and consume a lot of computing resources and time. In practical applications, it may be difficult to obtain high-resolution images or data due to environmental factors, equipment limitations, etc. It will also be affected by noise, such as background noise, environmental noise, etc. These noises may affect the accuracy and reliability of trace detection, making it difficult to extract complete traces in complex backgrounds, and may cause problems such as false detection and missed detection. At the same time, the existing detection methods have a slow processing speed for trace images and require a long processing time, which greatly affects the efficiency of trace detection.

[0004] Therefore, we propose a high-speed trace enhancement extraction method based on gaze scanning to solve the above problems. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In view of the deficiencies of the prior art, the present invention provides a high-speed trace enhancement extraction method based on gaze scanning, which solves the problems raised in the above background technology.

[0007] (II) Technical solution

[0008] In order to achieve the above-mentioned purpose, the present invention specifically adopts the following technical solutions:

[0009] A high-speed trace enhancement extraction method based on gaze scanning comprises the following steps:

[0010] S1: Four images with traces at different angles collected based on gaze scanning technology;

[0011] S2: Perform histogram equalization enhancement and median filtering denoising on the four collected trace images;

[0012] S3: Use the 12-direction Sobel edge detection algorithm to extract the edge structure of the trace and obtain the image edge;

[0013] S4: Use the RANSAC stitching algorithm to stitch the preprocessed images to form a complete trace image;

[0014] S5: The above stitching method is hardware accelerated and optimized in the FPGA platform to obtain high-speed and real-time trace detection results.

[0015] Furthermore, in S1, the image acquisition system used for acquiring trace images based on the gaze scanning technology includes:

[0016] Stepper motor: The stepper motor accurately controls the camera to collect horizontal and vertical images, retaining certain identical parts in two adjacent pictures, which can retain more feature values ​​and increase the accuracy of image stitching;

[0017] Industrial camera: The industrial camera is located in front of the stepper motor and performs a global scan of the target trace to obtain a series of high-pixel detail images for subsequent image noise reduction, stitching, and other processing;

[0018] During the image acquisition process, the FPGA sends a trigger signal to the stepper motor to control the camera to collect trace images at different angles, so as to achieve precise control of the shooting angle.

[0019] Furthermore, in S2, the four collected trace images are subjected to histogram equalization enhancement and median filtering denoising, and a global histogram equalization algorithm is used to achieve the purpose of image enhancement by counting the brightness of the pixels of the image and stretching the histogram. The algorithm is simple and effective. First, the total number of gray levels of the original image is defined as L, and the gray level is K∈[0,L-1]; secondly, the number of each pixel in the gray level in the entire image is counted; then, the probability distribution of each gray level in the image is calculated, and the probability of the k-th gray value is defined as k=0,1,2...L-1, where n is the total number of pixel values, and the cumulative distribution function is calculated as:

[0020]

[0021] Finally, the grayscale value after equalization is calculated and mapped back to the pixel value of the original pixel coordinate;

[0022] The median filtering method uses a two-dimensional sliding template of a certain structure to sort the pixels in the board according to their numerical values ​​to form a sequence of two-dimensional data that is monotonically decreasing (or increasing). The output of the two-dimensional median filter is expressed as:

[0023] g(x,y)=med{f(xk,yl),(k,l∈W)}.

[0024] Furthermore, in S3, the edge structure of the trace is extracted using the 12-direction Sobel edge detection algorithm, and then the edge points are identified by threshold value to obtain the image edge, g n is the direction of the center point, G d is the gradient amplitude of the center point after processing, and the formula is expressed as:

[0025]

[0026] Then set the threshold Th, and compare each pixel value in the newly obtained gradient image with the threshold Th. The formula is expressed as:

[0027]

[0028] If the gray value of a pixel exceeds the threshold Th, it is marked as an edge point G; if the gray value is lower than the threshold Th, the pixel is excluded from the edge points and its value is set to 0; in this way, the edge points detected by the Sobel operator are connected to form the edge of the image.

[0029] Further, in S4, the preprocessed images are stitched using the RANSAC stitching algorithm to form a complete trace image, n pairs of (n, l) feature matching points are randomly selected, and the corresponding H matrix is ​​calculated; the obtained homography matrix H is used to calculate all data in the feature matching point set to obtain the matching error between the true value of the matching pair and the calculated value of the matching pair, and the maximum threshold of the matching error is set; if the matching error is less than the threshold, then the pair of feature matching points is regarded as an inner point, otherwise, it is regarded as an outer point and should be eliminated, and finally a set of inner points can be obtained; the maximum number of iterations is set, and the first two steps are repeated until the largest set of inner points is found, and the corresponding homography matrix is ​​the optimal homography matrix; the homography matrix H can map points on a plane (represented by three-dimensional homogeneous coordinates, such as (u, v, 1)) to points on another plane, where H is a 3X3 matrix, and the representation formula is:

[0030]

[0031] Furthermore, in S5, after the whole system is powered on, the FPGA completes the control of the camera through the IIC control signal, collects the original image information of the trace image after the camera is initialized, and transmits the original image data from the image source to the FPGA through the interface for subsequent image processing and analysis; the image data containing the trace information is processed by the splicing core algorithm module to execute the edge structure extraction and splicing algorithm of the image; the clock module provides a clock signal to synchronize the operation of each module, and the processed image data is stored in the image cache module to ensure that the data will not be lost and can be accessed at an appropriate time. Finally, the complete trace image passes through the image output module, and the image data processed by the trace enhancement is transmitted to the output interface for users or other devices to observe, save or further process; the hardware parallel architecture is used inside the FPGA to perform fast trace extraction and detection processing;

[0032] The above stitching method is hardware accelerated and optimized in the FPGA platform, and the hardware parallel architecture is used inside the FPGA to perform fast stitching processing, wherein a lookup table (LUT) is introduced into the hardware parallel architecture to realize parallel processing of data and reduce the transmission delay of data between processors.

[0033] Furthermore, in the process of FPGA hardware implementation, the structure is optimized by using the parallelization method of the lookup table (LUT). The parallelization process of the lookup table (LUT) is as follows:

[0034] Ⅰ. Generate a lookup table offline. Perform offline calculations on the random sample selection stage in the RANSAC algorithm and generate a lookup table to quickly obtain random samples.

[0035] II. Design a VHDL module to implement the lookup table on the FPGA and perform other computational steps of the RANSAC algorithm. In the design process, the lookup table is used as a resource to store pre-computed results to achieve efficient data access and processing.

[0036] III. Integrate the lookup table module with other computational steps of the RANSAC algorithm to form a complete FPGA implementation, ensure that the lookup table module and the algorithm module are correctly connected, and adjust the clock and data paths to achieve smooth data transmission and computation flow.

[0037] Furthermore, the FPGA platform mainly includes the following modules:

[0038] Image acquisition module, used to transmit the raw image data from the image source to the FPGA through the interface for subsequent image processing and analysis;

[0039] The clock module is used to generate and distribute periodic clock signals to ensure that all modules in the entire system are in a synchronized state in order to coordinate data transmission and processing;

[0040] Image cache module, used to temporarily store intermediate results or data that needs to be processed later, to ensure that the data is not lost and can be accessed at the appropriate time;

[0041] The image output module is used to transmit the image data after the trace enhancement processing to the output interface so that the user or other devices can observe, save or further process it.

[0042] (III) Beneficial effects

[0043] Compared with the prior art, the present invention provides a high-speed trace enhancement extraction method based on gaze scanning, which has the following beneficial effects:

[0044] 1. The present invention improves the limitations of traditional manual collection of trace evidence. Based on the general characteristics of traces, a method of achieving high-quality extraction by scanning multiple trace images is proposed, which can completely and clearly extract traces in complex backgrounds.

[0045] 2. Based on the detection algorithm implemented on the FPGA platform, the speed is optimized through the lookup table method to achieve the purpose of quickly generating complete and coherent trace images, reduce possible trace loss and information ambiguity, improve quality and speed up, and better cope with the processing needs of large-scale trace evidence. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 Schematic diagram of the process of the trace extraction method used in the embodiment of the present invention;

[0047] Figure 2 A detection flow chart of a high-speed trace enhancement extraction method based on gaze scanning in an embodiment of the present invention;

[0048] Figure 3 The figure is an overall hardware block diagram of the trace extraction method in the embodiment of the present invention. DETAILED DESCRIPTION

[0049] 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.

[0050] Example

[0051] See also Figure 1-3An embodiment of the present invention provides a high-speed trace enhancement extraction method based on gaze scanning, and the method specifically includes steps S1-S5:

[0052] S1: Four trace images at different angles are collected based on gaze scanning technology;

[0053] The image acquisition system used to collect trace images based on gaze scanning technology includes:

[0054] Stepper motor: The stepper motor accurately controls the camera to capture horizontal and vertical images, retaining certain identical parts in two adjacent pictures, which can retain more feature values ​​and increase the accuracy of image stitching.

[0055] Industrial camera: The industrial camera is located in front of the stepper motor and performs a global scan of the target trace to obtain a series of high-pixel detail images for subsequent image denoising, stitching, and other processing.

[0056] During the image acquisition process, the FPGA sends a trigger signal to the stepper motor to control the camera to acquire trace images at different angles, so as to achieve precise control of the shooting angle, thereby effectively enhancing the acquisition effect of the trace image at different angles.

[0057] S2: Perform histogram equalization enhancement and median filtering denoising on the four collected trace images.

[0058] After the camera completes the acquisition of the trace image, in order to ensure the effectiveness of trace extraction, the acquired image containing traces is subjected to preprocessing steps such as image enhancement and denoising to eliminate the influence of irrelevant information in the image on the subsequent defect detection process.

[0059] The four trace images collected were enhanced by histogram equalization and denoised by median filtering. The global histogram equalization algorithm is used to achieve the purpose of image enhancement by counting the brightness of the pixels in the image and stretching the histogram. The algorithm is simple and effective. First, the total number of gray levels of the original image is defined as L, and the gray level is, K∈[0,L-1]; secondly, the number of each pixel in the gray level in the entire image is counted; then the probability distribution of each gray level in the image is calculated, and the probability of the k-th gray value is defined as k=0,1,2...L-1, where n is the total number of pixel values, and the cumulative distribution function is calculated as:

[0060]

[0061] Finally, the grayscale value after equalization is calculated and mapped back to the pixel value of the original pixel coordinates.

[0062] The median filtering method uses a two-dimensional sliding template of a certain structure to sort the pixels in the board according to their numerical values ​​to form a sequence of two-dimensional data that is monotonically decreasing (or increasing). The output of the two-dimensional median filter is expressed as:

[0063] g(x,y)=med{f(xk,yl),(k,l∈W)}.

[0064] S3: Use the 12-direction Sobel edge detection algorithm to detect the edge structure of the denoised and enhanced image, and then use the threshold to identify the edge points to obtain the image edge, providing reliable image information for trace recognition and detection.

[0065] In S3, in order to improve the accuracy of trace extraction, the four trace images that have been denoised and enhanced are processed by Sobel edge detection. n is the direction of the center point, G d is the gradient amplitude of the center point after processing, and the formula is expressed as:

[0066]

[0067] Then set the threshold Th, and compare each pixel value in the newly obtained gradient image with the threshold Th. The formula is expressed as:

[0068]

[0069] If the gray value of a pixel exceeds the threshold Th, it is marked as an edge point G. If the gray value is lower than the threshold Th, the pixel is excluded from the edge point and its value is set to 0. In this way, the edge points detected by the Sobel operator are connected to form the edge of the image.

[0070] S4: Use the RANSAC stitching algorithm to stitch the preprocessed images to form a complete trace image.

[0071] In S4, the preprocessed images are stitched using the RANSAC stitching algorithm to obtain the optimal homography matrix. The homography matrix H can map points on a plane (expressed using three-dimensional homogeneous coordinates, such as (u, v, 1)) to points on another plane, where H is a 3X3 matrix, expressed as:

[0072]

[0073] Randomly select n pairs of (n, l) feature matching points and calculate the corresponding H matrix; use the obtained homography matrix H to calculate all the data in the feature matching point set to obtain the matching error between the true value of the matching pair and the calculated value of the matching pair, and set the maximum threshold of the matching error. If the matching error is less than the threshold, then this pair of feature matching points is regarded as an inlier, otherwise, it is regarded as an outlier and should be eliminated. Finally, a set of inliers can be obtained; set the maximum number of iterations, repeat the first two steps until the largest set of inliers is found, and the corresponding homography matrix is ​​the optimal homography matrix.

[0074] S5: The above stitching method is hardware accelerated and optimized in the FPGA platform to obtain high-speed and real-time trace detection results.

[0075] See also Figure 3 , which is a hardware block diagram of the trace extraction method in an embodiment of the present invention, and the method includes the following steps:

[0076] In S5, after the whole system is powered on, FPGA completes the control of the camera through the IIC control signal. After the camera is initialized, the original image information of the trace image is collected, and the original image data from the image source is transmitted to the FPGA through the interface for subsequent image processing and analysis. The image data containing trace information is processed by the splicing core algorithm module, and the edge structure extraction and splicing algorithm of the image is executed. The clock module provides a clock signal to synchronize the operation of each module. The processed image data is stored in the image cache module to ensure that the data will not be lost and can be accessed at the appropriate time. Finally, the complete trace image passes through the image output module, and the image data processed by the trace enhancement is transmitted to the output interface for users or other devices to observe, save or further process.

[0077] A hardware parallel architecture is used inside the FPGA to perform fast trace extraction detection processing, wherein a lookup table (LUT) is introduced into the hardware parallel architecture to achieve parallel processing of data and reduce data transmission delay between processors.

[0078] The FPGA platform mainly includes the following modules:

[0079] Image acquisition module, used to transmit the raw image data from the image source to the FPGA through the interface for subsequent image processing and analysis;

[0080] The clock module is used to generate and distribute periodic clock signals to ensure that all modules in the entire system are in a synchronized state in order to coordinate data transmission and processing;

[0081] Image cache module, used to temporarily store intermediate results or data that needs to be processed later, to ensure that the data is not lost and can be accessed at the appropriate time;

[0082] The image output module is used to transmit the image data after the trace enhancement processing to the output interface so that the user or other devices can observe, save or further process it.

[0083] During the FPGA hardware implementation, the LUT lookup table parallelization method is used to optimize the structure to achieve high-speed operation of trace extraction detection. The LUT parallelization process is as follows:

[0084] Ⅰ. Offline generation of lookup table. The random sample selection stage in the RANSAC algorithm is calculated offline and a lookup table is generated to quickly obtain random samples.

[0085] Ⅱ. Design a VHDL module to implement the lookup table on the FPGA and perform other computational steps of the RANSAC algorithm. During the design process, the lookup table is used as a resource to store pre-computed results for efficient data access and processing.

[0086] III. Integrate the lookup table module with other computational steps of the RANSAC algorithm to form a complete FPGA implementation. Ensure that the lookup table module and the algorithm module are properly connected, and adjust the clock and data paths to achieve smooth data transfer and computation flow.

[0087] The embodiment of the present invention provides a high-speed trace enhancement and extraction method based on gaze scanning. To achieve this purpose, the technical solution integrates the principles and mechanisms of gaze scanning technology, combines relevant trace features, and designs a method for enhancing and splicing trace images at different angles to obtain high-quality trace images. This technology selects histogram equalization enhancement and median filtering denoising methods to improve the visualization and recognition of traces, and uses 12-directional Sobel edge detection to detect the edge structure of traces. This method helps to accurately extract traces, thereby providing reliable image information for trace identification and detection. The preprocessed images are spliced ​​to form a complete trace image, and the method is accelerated and optimized through FPGA hardware, ultimately achieving efficient, reliable and accurate detection and extraction of traces.

[0088] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A high-speed trace enhancement extraction method based on gaze scanning, characterized in that: The following steps are included: S1: Four images with traces at different angles collected based on gaze scanning technology; S2: Perform histogram equalization enhancement and median filtering denoising on the four collected trace images; S3: Use the 12-direction Sobel edge detection algorithm to extract the edge structure of the trace and obtain the image edge; S4: Use the RANSAC stitching algorithm to stitch the preprocessed images to form a complete trace image; S5: The above stitching method is hardware accelerated and optimized in the FPGA platform to obtain high-speed and real-time trace detection results.

2. The high-speed trace enhancement extraction method based on gaze scanning according to claim 1 is characterized in that: In S1, the image acquisition system used for collecting trace images based on the gaze scanning technology includes: Stepper motor: The stepper motor accurately controls the camera to collect horizontal and vertical images, retaining certain identical parts in two adjacent pictures, which can retain more feature values ​​and increase the accuracy of image stitching; Industrial camera: The industrial camera is located in front of the stepper motor and performs a global scan of the target trace to obtain a series of high-pixel detail images for subsequent image noise reduction, stitching, and other processing; During the image acquisition process, the FPGA sends a trigger signal to the stepper motor to control the camera to collect trace images at different angles, so as to achieve precise control of the shooting angle.

3. The high-speed trace enhancement extraction method based on gaze scanning according to claim 1 is characterized in that: In S2, the four collected trace images are enhanced by histogram equalization and denoised by median filtering. The global histogram equalization algorithm is used to achieve the purpose of image enhancement by counting the brightness of the pixels in the image and stretching the histogram. The algorithm is simple and effective. First, the total number of gray levels of the original image is defined as L, and the gray level is K∈[0,L-1]; secondly, the number of each pixel in the gray level in the entire image is counted; then the probability distribution of each gray level in the image is calculated, and the probability of the k-th gray value is defined as k=0,1,2...L-1, where n is the total number of pixel values, and the cumulative distribution function is calculated as: Finally, the grayscale value after equalization is calculated and mapped back to the pixel value of the original pixel coordinate; The median filtering method uses a two-dimensional sliding template of a certain structure to sort the pixels in the board according to their numerical values ​​to form a sequence of two-dimensional data that is monotonically decreasing (or increasing). The output of the two-dimensional median filter is expressed as: g(x,y)=med{f(xk,yl),(k,l∈W)}.

4. The high-speed trace enhancement extraction method based on gaze scanning according to claim 3 is characterized in that: In S3, the 12-directional Sobel edge detection algorithm is used to extract the edge structure of the trace, and then the edge points are identified by threshold value to obtain the image edge, g n is the direction of the center point, G d is the gradient amplitude of the center point after processing, and the formula is expressed as: Then set the threshold Th, and compare each pixel value in the newly obtained gradient image with the threshold Th. The formula is expressed as: If the gray value of a pixel exceeds the threshold Th, it is marked as an edge point G; if the gray value is lower than the threshold Th, the pixel is excluded from the edge points and its value is set to 0; in this way, the edge points detected by the Sobel operator are connected to form the edge of the image.

5. The high-speed trace enhancement extraction method based on gaze scanning according to claim 4 is characterized in that: In S4, the preprocessed images are stitched using the RANSAC stitching algorithm to form a complete trace image, n pairs of (n, l) feature matching points are randomly selected, and the corresponding H matrix is ​​calculated; the obtained homography matrix H is used to calculate all data in the feature matching point set to obtain the matching error between the true value of the matching pair and the calculated value of the matching pair, and a maximum threshold of the matching error is set; if the matching error is less than the threshold, then the pair of feature matching points is regarded as an inner point, otherwise, it is regarded as an outer point and should be eliminated, and finally a set of inner points can be obtained; Set the maximum number of iterations and repeat the first two steps until the largest set of internal points is found. The corresponding homography matrix is ​​the optimal homography matrix. The homography matrix H can map points on a plane (expressed using three-dimensional homogeneous coordinates, such as (u, v, 1)) to points on another plane, where H is a 3X3 matrix, and the expression formula is:

6. The high-speed trace enhancement extraction method based on gaze scanning according to claim 5 is characterized in that: In S5, after the whole system is powered on, FPGA completes the control of the camera through the IIC control signal, collects the original image information of the trace image after the camera is initialized, and transmits the original image data from the image source to the FPGA through the interface for subsequent image processing and analysis; the image data containing the trace information is processed by the splicing core algorithm module, and the edge structure extraction and splicing algorithm of the image is executed; the clock module provides a clock signal to synchronize the operation of each module, and the processed image data is stored in the image cache module to ensure that the data will not be lost and can be accessed at an appropriate time. Finally, the complete trace image passes through the image output module, and the image data after the trace enhancement processing is transmitted to the output interface for users or other devices to observe, save or further process; the hardware parallel architecture is used inside the FPGA to perform fast trace extraction and detection processing; The above stitching method is hardware accelerated and optimized in the FPGA platform, and the hardware parallel architecture is used inside the FPGA to perform fast stitching processing, wherein a lookup table (LUT) is introduced into the hardware parallel architecture to realize parallel processing of data and reduce the transmission delay of data between processors.

7. The high-speed trace enhancement extraction method based on gaze scanning according to claim 6 is characterized in that: In the process of FPGA hardware implementation, the structure is optimized by using the parallelization method of the lookup table (LUT). The parallelization process of the lookup table (LUT) is as follows: Ⅰ. Generate a lookup table offline. Perform offline calculations on the random sample selection stage in the RANSAC algorithm and generate a lookup table to quickly obtain random samples. II. Design a VHDL module to implement the lookup table on the FPGA and perform other computational steps of the RANSAC algorithm. In the design process, the lookup table is used as a resource to store pre-computed results to achieve efficient data access and processing. III. Integrate the lookup table module with other computational steps of the RANSAC algorithm to form a complete FPGA implementation, ensure that the lookup table module and the algorithm module are correctly connected, and adjust the clock and data paths to achieve smooth data transmission and computation flow.

8. The high-speed trace enhancement extraction method based on gaze scanning according to claim 6 is characterized in that: The FPGA platform mainly includes the following modules: Image acquisition module, used to transmit the raw image data from the image source to the FPGA through the interface for subsequent image processing and analysis; The clock module is used to generate and distribute periodic clock signals to ensure that all modules in the entire system are in a synchronized state in order to coordinate data transmission and processing; Image cache module, used to temporarily store intermediate results or data that needs to be processed later, to ensure that the data is not lost and can be accessed at the appropriate time; The image output module is used to transmit the image data after the trace enhancement processing to the output interface so that the user or other devices can observe, save or further process it.