Anti-shake performance verification method and system for displacement sensor video corner tracking

Abstract

The invention relates to an anti-shake performance verification method and system for video corner tracking of a displacement sensor. The method comprises the following steps: acquiring a to-be-tested video file; feature point detection is carried out on each frame of video in the to-be-detected video file; performing optical flow tracking on the feature points between adjacent frames to obtain position information of the feature points in each frame; calculating the displacement difference of the feature points according to the position information of the feature points in each frame; and calculating the standardized displacement difference of each feature point in the adjacent frames according to the displacement difference of the feature points. The method has the beneficial effects that the calculation efficiency is improved, a unified evaluation standard is provided, and evaluation results obtained by using different methods and equipment have comparability and comprehensiveness.

Description

Technical Field

[0001] The present invention relates to the field of anti-shake performance verification, and in particular to an anti-shake performance verification method and system for video corner point tracking using a displacement sensor. Background Art

[0002] When shooting videos or images, hand shaking often results in blurred images. Optical image stabilization (Sensorshift) technology reduces the impact of shaking on image quality by introducing a compensation mechanism in the lens or sensor. However, the current sensor shift performance test methods are diverse and lack standardization.

[0003] Problems and defects of the existing technology:

[0004] 1. Insufficient accuracy: The existing sensor shift performance verification method may have errors in feature point detection and tracking, resulting in inaccurate displacement differences in the final calculation, thus affecting the evaluation results of the sensor shift performance.

[0005] 2. High computational complexity: Many existing methods require complex calculations and a lot of processing time, especially when dealing with high-resolution videos and a large number of feature points. This increases the demand and cost of computing resources.

[0006] 3. Low level of automation: Existing technologies usually require a lot of manual intervention, resulting in inefficiency and increasing the risk of human error.

[0007] 4. Poor adaptability: Some existing methods have poor adaptability to different types of devices and different shooting conditions (such as light changes, motion blur, etc.), which limits their application scope.

[0008] 5. Poor real-time performance: Many existing SENSOR SHIFT performance verification methods cannot process video data in real time, resulting in the inability to obtain SENSOR SHIFT performance evaluation results in a timely manner.

[0009] 6. Inconvenience in data recording and management: The existing technology is inconvenient in recording and managing feature point displacement data, and lacks systematic recording and management tools, which makes data analysis and comparison difficult.

[0010] 7. The result display is not intuitive: Some existing methods are not intuitive in displaying the SENSOR SHIFT performance evaluation results, and cannot clearly convey the pros and cons of SENSOR SHIFT performance, affecting users' understanding and decision-making of the results.

[0011] 8. Lack of standardization: The existing SENSOR SHIFT performance verification methods lack unified standards, which makes it difficult to compare and integrate the evaluation results of different methods and equipment. Summary of the invention

[0012] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method and system for verifying the anti-shake performance of displacement sensor video corner tracking, which improves the calculation efficiency and provides a unified evaluation standard, so that the evaluation results using different methods and devices are comparable and comprehensive.

[0013] The objective of the present invention is achieved through the following technical solutions:

[0014] The anti-shake performance verification method of the displacement sensor video corner tracking includes:

[0015] Get the video file to be tested;

[0016] Perform feature point detection on each frame of the video file to be tested;

[0017] Perform optical flow tracking on the feature points between adjacent frames to obtain the position information of the feature points in each frame. During the optical flow tracking process, if there are undetected feature points in one frame, the quality threshold parameter can be adjusted to ensure that the number of feature points in each frame is consistent.

[0018] According to the position information of the feature points in each frame, the displacement difference of the feature points is calculated;

[0019] Calculate the normalized displacement difference of each feature point in adjacent frames according to the displacement difference of the feature points;

[0020] Based on the normalized displacement difference, the SR value is calculated, the anti-shake performance is determined according to the SR value, and an evaluation result is generated;

[0021] The feature points are corner points. Corner points are selected as feature points to enhance the stability and accuracy of tracking.

[0022] Further optional, feature point detection method and parameter setting: adopting a corner point detection algorithm and an adaptive quality threshold to ensure high-precision feature point extraction; preferably, the corner point detection method can adopt the Harris corner point detection method.

[0023] Feature point tracking algorithm and its improvement measures: By adjusting the window size, number of pyramid layers and termination criteria, the stability and accuracy of feature point tracking can be enhanced.

[0024] Furthermore, the anti-shake performance verification method also includes:

[0025] According to the standardized displacement difference of each feature point in adjacent frames, the maximum standardized displacement difference of each feature point is calculated, and the maximum standardized displacement difference of each feature point is stored and recorded in a table. The maximum standardized displacement difference is obtained to evaluate the performance of SENSOR SHIFT.

[0026] Furthermore, the evaluation result includes a maximum SR value, a minimum SR value and an average SR value among the SR values ​​corresponding to all frames as the OIS performance evaluation result.

[0027] The maximum SR value, minimum SR value, and average SR value are related values ​​of multiple feature points. For example, there are three feature points in total, and each feature point has a maximum displacement difference. Then take the average value or the maximum and minimum values ​​of these three points. This helps to eliminate the influence of extreme points and reflect the stability of the overall image.

[0028] Shake Reduction (SR) refers to the degree to which image shake is reduced when the imaging device turns on optical image stabilization compared to when it does not turn on optical image stabilization in an actual environment with a certain vibration frequency.

[0029] Furthermore, the calculation formula of the standardized displacement difference is:

[0030]

[0031] Furthermore, according to the standardized displacement difference of the feature points, the specific process of calculating the SR value of each frame is as follows:

[0032] The weighted average displacement difference of each frame is calculated based on the standardized displacement difference of the feature points, and then the SR value is calculated based on the weighted average displacement difference;

[0033] The formula for calculating the SR value is:

[0034]

[0035] Wherein, j=2, 3...m-1, the total number of video frames is m frames;

[0036] The formula for calculating the weighted average displacement difference is:

[0037]

[0038] Among them, ωi is the weight of the i-th feature point, di is the standardized displacement difference of the i-th feature point, and the total number of feature points is n.

[0039] Furthermore, the anti-shake performance verification method also includes:

[0040] Calculate the displacement difference change rate according to the position information of the feature point in each frame, and store and record the displacement difference change rate in a table;

[0041] The calculation formula of the displacement difference change rate is:

[0042]

[0043] Wherein, ΔX = |displacement difference of the current frame in the X-axis direction - displacement difference of the previous frame in the X-axis direction|;

[0044] Δy=|displacement difference of the current frame on the y-axis - displacement difference of the previous frame on the y-axis|.

[0045] By calculating the displacement difference change rate of each feature point between consecutive frames, the performance of the OIS system in a dynamic environment is identified.

[0046] Furthermore, in the anti-shake performance verification method, the goodFeaturesToTrack function in the OpenCV library is used to detect feature points in each frame of the video file.

[0047] Furthermore, in the anti-shake performance verification method, the calcOpticalFlowPyrLK function is used to track the position changes of feature points in adjacent frames.

[0048] The present disclosure also provides an anti-shake performance verification system for video corner tracking using a displacement sensor, comprising:

[0049] File import module, used to obtain the video file to be tested;

[0050] A feature point detection module is used to perform feature point detection on each frame of the video file to be tested;

[0051] The optical flow tracking module is used to perform optical flow tracking on feature points between adjacent frames to obtain the position information of the feature points in each frame;

[0052] A displacement difference calculation module, used to calculate the displacement difference of the feature point according to the position information of the feature point in each frame;

[0053] A standardized displacement difference calculation module is used to calculate the standardized displacement difference of each feature point in adjacent frames according to the displacement difference of the feature points;

[0054] An SR calculation module, used to calculate the SR value of each frame based on the normalized displacement difference;

[0055] A result generation module, used to determine the anti-shake performance according to the SR value and generate an evaluation result;

[0056] Wherein, the feature point is a corner point.

[0057] Furthermore, the SR calculation module is used to calculate the SR value of each frame of all frames based on the standardized displacement difference, calculate the average SR value, and output the maximum SR value, the minimum SR value, and the average SR value.

[0058] Furthermore, the result generation module is used to display the maximum SR value, the minimum SR value, and the average SR value based on the calculated maximum SR value, the minimum SR value, and the average SR value.

[0059] Preferably, in the anti-shake performance verification system, the result display interface of the result generation module is implemented through the Tkinter library.

[0060] The present disclosure also provides an electronic device, comprising a processor and a memory; wherein the memory is used to store a computer program, and the computer program is loaded and executed by the processor to implement the anti-shake performance verification method as described above.

[0061] The present disclosure also provides a computer-readable storage medium for storing a computer program; wherein the computer program implements the anti-shake performance verification method as described above when executed by a processor.

[0062] The present disclosure also provides a computer program product, which includes a computer program, and the computer program implements the above method when executed by a processor.

[0063] The beneficial effects of the present invention are:

[0064] The present invention proposes a method and system for verifying the anti-shake performance of video corner tracking using a displacement sensor, which can improve the accuracy of displacement difference calculation, improve calculation efficiency, and is highly automated, adaptable, and has real-time processing capabilities. It has systematic data recording and management, and the performance evaluation results are more intuitive, which is convenient for users to understand and make decisions. A unified evaluation standard is provided, so that the evaluation results using different methods and devices are comparable and comprehensive. Specifically:

[0065] (1) The present disclosure uses advanced computer vision algorithms (such as optical flow) to detect and track feature points, ensuring high accuracy under different shooting conditions (such as light changes, motion blur, etc.); and enhances the stability and accuracy of feature point tracking by selecting the best feature points (such as corner points);

[0066] (2) The present disclosure records and analyzes the position information of feature points to accurately calculate the displacement difference of each feature point; and adopts innovative calculation methods (such as maximum displacement difference (maximum SR value) calculation) to improve the accuracy of SENSOR SHIFT performance evaluation;

[0067] (3) The present disclosure eliminates the scale differences between different videos by standardizing the displacement differences, making the evaluation results comparable; by performing weighted average calculation on the displacement differences of feature points, the influence of noise is reduced and the robustness of the evaluation results is improved; by calculating the rate of change of the displacement differences between frames, the performance of the OIS system in a dynamic environment can be better reflected;

[0068] (4) The present disclosure realizes real-time processing and analysis of video data by optimizing algorithms and data processing procedures, and provides real-time SENSOR SHIFT performance evaluation results to meet the needs of rapid evaluation;

[0069] (5) The present disclosure provides a stabilization performance verification system for displacement sensor video corner tracking for Sensor shift 8K video, which integrates highly systematic data recording and management functions; the system can automatically collect and output key data such as location information, motion trajectory and displacement difference of feature points, and save them to EXCEL tables to achieve efficient data storage, retrieval and in-depth analysis; through a strict recording mechanism, the integrity, accuracy and traceability of the data are ensured, thereby supporting the system to make effective data utilization in subsequent performance analysis, visualization and stabilization algorithm optimization; this solution not only improves the automation and intelligence level of the stabilization verification process, but also greatly improves the accuracy and efficiency of data processing, ensuring the reliability and scientificity of the verification results;

[0070] (6) The SR value is calculated by standardizing the displacement difference and weighted average displacement difference, which provides a unified evaluation standard and method, making the evaluation results of different methods and equipment comparable and comprehensive, ensuring the accuracy and consistency of the evaluation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 It is a schematic diagram of the system architecture diagram of the present invention;

[0072] Figure 2 It is a flowchart of feature point detection and tracking;

[0073] Figure 3 is a schematic diagram of the calculation flow chart of displacement difference;

[0074] Figure 4 The following is a schematic diagram of the result display interface. DETAILED DESCRIPTION

[0075] Reference will now be made to embodiments of the present general inventive concept, examples of which are illustrated in the drawings. These embodiments are described below in order to explain the present general inventive concept by referring to the drawings.

[0076] Example 1

[0077] refer to Figure 1 As shown, a stabilization performance verification system for displacement sensor video corner tracking of Sensor shift 8K video includes:

[0078] File import module, used to obtain the video file to be tested;

[0079] A feature point detection module is used to perform feature point detection on each frame of the video file to be tested;

[0080] The optical flow tracking module is used to perform optical flow tracking on feature points between adjacent frames to obtain the position information of the feature points in each frame;

[0081] A displacement difference calculation module, used to calculate the displacement difference of the feature point according to the position information of the feature point in each frame;

[0082] A standardized displacement difference calculation module is used to calculate the standardized displacement difference of each feature point in adjacent frames according to the displacement difference of the feature points;

[0083] An SR calculation module, used to calculate the SR value of each frame based on the normalized displacement difference;

[0084] A result generation module, used to determine the anti-shake performance according to the SR value and generate an evaluation result;

[0085] Wherein, the feature point is a corner point.

[0086] Furthermore, the SR calculation module is used to calculate the SR value of each frame based on the standardized displacement difference, and calculate the average SR value.

[0087] Furthermore, the result generation module is used to display the maximum SR value, the minimum SR value, and the average SR value based on the calculated maximum SR value, the minimum SR value, and the average SR value to show the SENSOR SHIFT performance result.

[0088] Example 2

[0089] refer to Figure 1 As shown, a stabilization performance verification system for displacement sensor video corner tracking of Sensor shift 8K video, the overall architecture of which includes the following main modules: file import module, feature point detection module, optical flow tracking module, data recording module, displacement difference calculation module, standardized displacement difference calculation module, SR calculation module, result generation module;

[0090] Among them, the file import module is used to import the video file to be tested;

[0091] A feature point detection module is used to detect feature points in each frame of the video file to be tested;

[0092] An optical flow tracking module is used to use the detected feature points to perform optical flow tracking of feature points between adjacent frames and output the position information of the feature points in the adjacent frames;

[0093] A data recording module is used to record the position information of feature points in each frame;

[0094] A displacement difference calculation module, used to calculate the displacement difference of the feature point according to the position information of the feature point in each frame;

[0095] A standardized displacement difference calculation module is used to calculate the standardized displacement difference of each feature point in adjacent frames according to the displacement difference of the feature points;

[0096] An SR calculation module, used to calculate the SR value of each frame based on the normalized displacement difference;

[0097] A result generation module, used to determine the anti-shake performance according to the SR value and generate an evaluation result;

[0098] Wherein, the feature point is a corner point.

[0099] Furthermore, the SR calculation module is specifically used to calculate the SR value of each frame based on the standardized displacement difference, calculate the average SR value, and output the maximum SR value, the minimum SR value, and the average SR value.

[0100] Furthermore, the result generation module is used to display the maximum SR value, the minimum SR value, and the average SR value based on the calculated maximum SR value, the minimum SR value, and the average SR value to show the SENSOR SHIFT performance result.

[0101] Example 3

[0102] The anti-shake performance verification method of displacement sensor video corner tracking includes S100 to S700.

[0103] S100. Obtain a video file to be tested.

[0104] For example, input the video file to be tested into the system and read the video frame through the cv2.VideoCapture function.

[0105] S200. Perform feature point detection on each frame of the video file to be tested, wherein the feature points are corner points.

[0106] For example, Harris corner detection is used to detect feature points in each frame of video, and cv2.goodFeaturesToTrack function in OpenCV is used to detect corner points. If there are undetected feature points, the quality threshold parameter is adjusted to ensure the consistency of the number of feature points.

[0107] Sort the detected corner points to ensure the stability of corner point detection.

[0108] S300. Perform optical flow tracking on feature points between adjacent frames to obtain position information of the feature points in each frame.

[0109] For example, between consecutive frames, optical flow tracking is performed through the cv2.calcOpticalFlowPyrLK function (integrated function) to obtain the motion trajectory of each feature point (if there are undetected feature points in one frame, the quality threshold parameter can be adjusted to ensure that the number of feature points in each frame is consistent). If the optical flow tracking is successful, the new position of the feature point is recorded; otherwise, the feature point is discarded.

[0110] S400. Calculate the displacement difference of the feature point according to the position information of the feature point in each frame.

[0111] For example, suppose that in the first frame, the position coordinates of a feature point are (x1, y1), and in the second frame, the position coordinates of the same feature point are (x2, y2). The displacement difference of the feature point can be calculated as:

[0112]

[0113] S500. Calculate the normalized displacement difference of each feature point in adjacent frames according to the displacement difference of the feature points.

[0114] For example, the normalized displacement difference is calculated as:

[0115]

[0116] S600. Calculate the SR value of each frame based on the normalized displacement difference.

[0117] For example, the SR value of the jth frame can be calculated by the following formula:

[0118]

[0119] Wherein, j=2, 3, ..., m-1, and the total number of video frames is m frames.

[0120] S700. Determine the anti-shake performance according to the SR value and generate an evaluation result.

[0121] For example, if the maximum value, minimum value, and average value are all greater than 25dB, the anti-shake performance is qualified, otherwise it is unqualified.

[0122] Furthermore, in step S600, the formula for calculating the weighted average displacement difference of each frame is:

[0123]

[0124] Among them, ωi is the weight of the i-th feature point, di is the standardized displacement difference of the i-th feature point, and the total number of feature points is n.

[0125] This implementation specifically includes functional processes such as feature point detection and tracking process, standardized displacement difference calculation, real-time processing and result display, data recording and management, and standardized evaluation.

[0126] 1. Feature Point Detection and Tracking Process

[0127] The following steps are involved:

[0128] Step 1: Video Input

[0129] Input the video file to be tested into the system and read the video frame through the cv2.VideoCapture function.

[0130] Step 2: Feature point detection

[0131] Harris corner detection method is used to detect feature points in each frame of image, and cv2.goodFeaturesToTrack function in OpenCV is used to detect corner points. If there are undetected feature points, the quality threshold parameter is adjusted to ensure the consistency of the number of feature points.

[0132] Sort the detected corner points to ensure the stability of corner point detection.

[0133] Step 3: Optical flow tracking

[0134] Between consecutive frames, optical flow tracking is performed through the cv2.calcOpticalFlowPyrLK function (integrated function) to obtain the motion trajectory of each feature point.

[0135] If the optical flow tracking is successful, the new position of the feature point is recorded; otherwise, the feature point is discarded.

[0136] Step 4: Feature point location recording

[0137] The position of each feature point in each frame is recorded in an Excel table, and the OpenPyXL library is used for data recording and management.

[0138] Flowchart as Figure 2 shown.

[0139] 2. Calculation of Standardized Displacement Difference

[0140] (1) The displacement difference calculation includes the following steps:

[0141] Step S1: Read feature point position data

[0142] Read the recorded feature point location data from the Excel table.

[0143] Step S2: Calculate the normalized displacement difference

[0144] For each feature point, calculate its displacement difference in the entire video, and calculate the standardized displacement difference through the displacement difference;

[0145] The normalized displacement difference is calculated as:

[0146]

[0147] Step S3: Calculate the maximum normalized displacement difference

[0148] Find the value with the largest normalized displacement difference among all feature points as the key indicator for OIS performance evaluation.

[0149] Flowchart as Figure 3 shown.

[0150] Further optimized, based on the displacement difference calculation result in step S2, step three of this implementation further includes: calculating the displacement difference change rate, and storing and tabulating the displacement difference change rate;

[0151] The calculation formula of the displacement difference change rate is:

[0152]

[0153] Wherein, ΔX = |displacement difference of the current frame in the X-axis direction - displacement difference of the previous frame in the X-axis direction|;

[0154] Δy=|displacement difference of the current frame on the y-axis - displacement difference of the previous frame on the y-axis|.

[0155] 3. Real-time processing and result display

[0156] The following steps are involved:

[0157] Step A1: Real-time processing

[0158] Optimize feature point detection and optical flow tracking algorithms to achieve real-time processing of video data.

[0159] While processing the video frames, the feature point positions are recorded in real time and the displacement differences are calculated.

[0160] Step A2: Result display

[0161] The user interface is created through the Tkinter library to display the OIS performance evaluation results in real time.

[0162] The display module can provide multiple display modes, such as displaying the maximum SR value, minimum SR value and average SR value separately according to the needs. An optional result display interface is as follows Figure 4 As shown, the result displayed is the maximum SR value in db.

[0163] 4. Data Recording and Management

[0164] The following steps are involved:

[0165] Step B1: Data Recording

[0166] Use the OpenPyXL library to record the feature point location data in an Excel table, with one worksheet for each feature point.

[0167] Record the coordinates of the feature points in each frame and their differences from the initial positions in the worksheet.

[0168] Step B2: Data Management

[0169] Provide systematic data management tools to facilitate data storage, retrieval and analysis.

[0170] Ensure data integrity and traceability through standardized data formats.

[0171] 5. Standardized Assessment

[0172] The standardized assessment methodology includes the following steps:

[0173] Step C1: Standardize the evaluation criteria

[0174] Establish unified evaluation standards to ensure that the evaluation results of different equipment and methods are comparable.

[0175] Step C2: Consistency of results

[0176] Ensure the accuracy and consistency of evaluation results through consistent evaluation methods and calculation formulas.

[0177] 1. Read feature point data from Excel file

[0178] Read the location data and difference data of feature points from the Excel file generated by each video processing.

[0179] 2. Calculate SR value

[0180] Calculate the SR value of each feature point based on the data read from the Excel file.

[0181] ① Calculate the weighted average displacement difference of each frame based on the standardized displacement difference of the feature points.

[0182] ② Calculate the SR value based on the weighted average displacement difference.

[0183] 3. Comprehensive calculation of SR value

[0184] According to the SR value of each feature point, the overall SR value is calculated. The maximum value, minimum value or average value can be used as the final evaluation result.

[0185] The above are only preferred embodiments of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the present invention through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art do not deviate from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.

Claims

1. A method for verifying the anti-shake performance of a displacement sensor video corner tracking, characterized in that: include: Get the video file to be tested; Perform feature point detection on each frame of the video file to be tested; Perform optical flow tracking on feature points between adjacent frames to obtain the position information of feature points in each frame; According to the position information of the feature points in each frame, the displacement difference of the feature points is calculated; Calculate the normalized displacement difference of each feature point in adjacent frames according to the displacement difference of the feature points; Based on the normalized displacement difference, the SR value of each frame is calculated; Determine the anti-shake performance according to the SR value and generate an evaluation result; Wherein, the feature point is a corner point.

2. The anti-shake performance verification method according to claim 1, characterized in that: The anti-shake performance verification method also includes: According to the standardized displacement difference of each feature point in adjacent frames, the maximum standardized displacement difference of each feature point is calculated, and the maximum standardized displacement difference of each feature point is stored and recorded in a table.

3. The anti-shake performance verification method according to claim 1, characterized in that: The evaluation result includes a maximum SR value, a minimum SR value and an average SR value among the SR values ​​corresponding to all frames.

4. The anti-shake performance verification method according to claim 1, characterized in that: The normalized displacement difference is calculated as:

5. The anti-shake performance verification method according to claim 4, characterized in that: Based on the normalized displacement difference, the SR value of each frame is calculated, including: The weighted average displacement difference of each frame is calculated based on the standardized displacement difference of the feature points, and then the SR value of each frame is calculated based on the weighted average displacement difference; The calculation formula of SR value is: Wherein, j=2, 3...m-1, the total number of video frames is m frames; The formula for the weighted mean displacement difference is: Among them, ωi is the weight of the i-th feature point, di is the standardized displacement difference of the i-th feature point, and the total number of feature points is n.

6. The anti-shake performance verification method according to claim 1, characterized in that: The anti-shake performance verification method also includes: Calculate the displacement difference change rate according to the position information of the feature point in each frame, and store and record the displacement difference change rate in a table; The calculation formula of the displacement difference change rate is: Wherein, ΔX = |displacement difference of the current frame in the X-axis direction - displacement difference of the previous frame in the X-axis direction|; Δy=|displacement difference of the current frame on the y-axis - displacement difference of the previous frame on the y-axis|.

7. The anti-shake performance verification system of the displacement sensor video corner tracking is characterized by: include: File import module, used to obtain the video file to be tested; A feature point detection module is used to perform feature point detection on each frame of the video file to be tested; The optical flow tracking module is used to perform optical flow tracking on feature points between adjacent frames to obtain the position information of the feature points in each frame; A displacement difference calculation module, used to calculate the displacement difference of the feature point according to the position information of the feature point in each frame; A standardized displacement difference calculation module is used to calculate the standardized displacement difference of each feature point in adjacent frames according to the displacement difference of the feature points; An SR calculation module, used to calculate the SR value of each frame based on the normalized displacement difference; A result generation module, used to determine the anti-shake performance according to the SR value and generate an evaluation result; Wherein, the feature point is a corner point.

8. The anti-shake performance verification system according to claim 7, characterized in that: SR calculation module: used to calculate the SR value of each frame based on the standardized displacement difference, and calculate the average SR value of all frames.

9. The anti-shake performance verification system according to claim 8, characterized in that: Result generation module: used to display the maximum SR value, minimum SR value, and average SR value based on the calculated maximum SR value, minimum SR value, and average SR value.

10. An electronic device, characterized in that: The electronic device comprises a processor and a memory; wherein the memory is used to store a computer program, and the computer program is loaded and executed by the processor to implement the anti-shake performance verification method according to any one of claims 1 to 6.

Images