An image device displacement anomaly early warning system and method based on feature point detection

Abstract

The present invention provides an image device displacement anomaly early warning system and method based on feature point detection, belonging to the technical field of anomaly early warning. Specifically, it includes: using the ORB algorithm to perform feature point detection and descriptor calculation on the reference image and the currently acquired image, and using the feature descriptors of the reference image and the current image to match to obtain the image comparison image feature points. Using a distance function to determine the feature point similarity coefficient of different comparison image feature points, and using the feature point similarity coefficient to perform matching feature points of different comparison image feature points. Based on the matching feature points, a homography matrix is constructed, and based on the homography matrix, geometric transformation information is extracted to obtain the rotation angle, scaling ratio, and translation amount, and combined with a preset threshold to determine whether the image device has shifted and issue an early warning signal, improving the reliability of monitoring and processing.

Description

Technical Field

[0001] The present invention belongs to the technical field of anomaly warning, and particularly relates to an image device displacement anomaly warning system and method based on feature point detection. Background Art

[0002] With the popularization of intelligent devices and Internet of Things technologies, image acquisition devices (such as cameras) are widely used in fields such as monitoring, navigation, and intelligent manufacturing. However, during operation, the acquisition device may be displaced from its installation position due to external interference (such as vibration, collision, or manual adjustment), affecting the accuracy of image acquisition and even causing system anomalies.

[0003] Therefore, in order to dynamically monitor the installation state of the image acquisition device and issue a warning in a timely manner when an offset anomaly occurs, in the invention patent application CN202120809098.0 "Video Monitoring Device for Tourist Scenic Areas", when in use, the locking mechanism is clamped on the tree trunk or the crossbar of the lamp post through lateral movement, and the tension spring controls the locking mechanism to close to prevent it from falling off, then it can be powered on and used. The locking mechanism prevents the monitoring device from sliding and causing the monitoring position to shift. However, through analysis, the following technical problems exist:

[0004] The prior art solutions do not disclose using the analysis results of the monitoring images to perform anomaly detection on the abnormal displacement of the installation position of the image acquisition device, so it is impossible to accurately identify and process the abnormal changes in the installation position of the image device.

[0005] Specifically, the present invention proposes an image device displacement anomaly warning system and method based on feature point detection. By analyzing the geometric transformation relationship (rotation angle, scaling ratio, and translation vector) between the images of the initial installation state and the current operating state of the image acquisition device, it is determined whether the device has shifted. Summary of the Invention

[0006] To achieve the object of the present invention, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present application provides an image device displacement anomaly warning system based on feature point detection, specifically including:

[0008] A feature point screening module, a matrix calculation and information extraction module, and a displacement recognition and warning module;

[0009] Among them, the feature point screening module is responsible for using the ORB algorithm to perform feature point detection and descriptor calculation on the reference image and the currently acquired image, and using the feature descriptors of the reference image and the current image to match to obtain the image comparison image feature points;

[0010] The matrix calculation and information extraction module is responsible for determining the feature point similarity coefficients of different comparison image feature points using a distance function, and using the feature point similarity coefficients to match the feature points of different comparison image feature points, and constructing a homography matrix based on the matched feature points;

[0011] The displacement recognition and warning module is responsible for extracting geometric transformation information based on the homography matrix to obtain the rotation angle, scaling ratio, and translation amount, and combining a preset threshold to determine whether the image device has shifted and issue a warning signal.

[0012] A further technical solution is that the feature point similarity coefficient is determined according to the Hamming distance function.

[0013] A further technical solution is that when any one of the rotation angle, scaling ratio, and translation amount is greater than a preset threshold, it is determined that the image device has shifted and a warning signal is issued.

[0014] In a second aspect, the present application provides a method for warning of abnormal displacement of an image device based on feature point detection, which is applied to the above-mentioned system for warning of abnormal displacement of an image device based on feature point detection, and specifically includes:

[0015] S1 Based on the analysis result of the monitoring image of the image device, determine the change situation of the light data at different times on different dates, and use the change situation to determine the feature comparison period in the period;

[0016] S2 Obtain the change situation of different ORB image feature points and initial image feature points in the monitoring image during the feature comparison period, and use the change situation of the image feature points to determine the comparison image feature points in the feature comparison period and the credibility coefficients of different comparison image feature points;

[0017] S3 Obtain the change situation of the comparison image feature points and the initial image feature points during the feature comparison period, and combine the credibility coefficients of different comparison image feature points to determine that when the image device has a suspected change, proceed to the next step;

[0018] S4 Determine whether to issue a warning signal based on the change situation of the comparison image feature points in different periods.

[0019] The beneficial effects of the present invention are as follows:

[0020] Based on the change situation of the comparison image feature points and the initial image feature points in the feature comparison period and the credibility coefficients of different comparison image feature points, it is determined whether the image device has a suspected change, thus realizing the accurate evaluation of the displacement change probability of the image feature points from the change situation and the credibility coefficients of the comparison image feature points in the feature comparison period, and also laying a foundation for the differential recognition and processing of the displacement of the image feature points.

[0021] Based on the change situation of the comparison image feature points in different periods to determine whether to send out a warning signal, it avoids the technical problem of insufficient accuracy of recognition and processing caused by only considering the change situation of the comparison image feature points in a single period, realizes the recognition and processing of the warning signal from the change data of the comparison image feature points in multiple periods, and improves the reliability and accuracy of the output processing of the warning signal for the abnormal change of the image device.

[0022] A further technical solution is that the change situation of the light data is determined according to the brightness of the monitoring image of the image device in the period.

[0023] A further technical solution is that the light data is determined according to the preset light brightness corresponding to the brightness of the monitoring image.

[0024] A further technical solution is that the method for determining the feature comparison period in the period is as follows:

[0025] Determine the light brightness of the period in different dates for the light data in different dates, and determine the reference light brightness with the average value of the light brightness of the period in different dates;

[0026] Determine the light change date in the date according to the deviation amount between the light brightness in different dates and the reference light brightness;

[0027] Determine whether the period is a feature comparison period by the proportion of the number of the light change dates.

[0028] A further technical solution is that when the proportion of the number of the light change dates is less than the preset proportion, it is determined that the period is a feature comparison period.

[0029] A further technical solution is that the light change date is the date when the deviation amount from the reference light brightness is not within the preset deviation amount range.

[0030] A further technical solution is that determining whether to send out a warning signal specifically includes:

[0031] Using the matching results of the comparison image feature points in different time periods, determining the feature point similarity coefficients of different comparison image feature points by using a distance function, and using the feature point similarity coefficients to perform matching feature points of different comparison image feature points;

[0032] Constructing a homography matrix based on the matching feature points, extracting geometric transformation information including rotation angle, scaling ratio, and translation amount based on the homography matrix, and determining whether the image device is offset in combination with a preset threshold.

[0033] A further technical solution is that when any one of the rotation angle, scaling ratio, and translation amount in multiple time periods does not meet the requirements of the preset threshold, it is determined that the image device has an offset, and a warning signal is output.

[0034] Other feature points and advantages will be described in the subsequent specification. The objectives and other advantages of the present invention are achieved and obtained by the structures specifically pointed out in the specification and the drawings.

[0035] To make the above objectives, feature points, and advantages of the present invention more obvious and understandable, the following specifically enumerates preferred embodiments and, in conjunction with the accompanying drawings, provides a detailed description as follows. Description of the Drawings

[0036] By referring to the drawings and describing its exemplary embodiments in detail, the above and other feature points and advantages of the present invention will become more obvious.

[0037] Figure 1 is a framework diagram of an image device displacement anomaly warning system based on feature point detection;

[0038] Figure 2 is a flowchart of a method for warning of image device displacement anomaly based on feature point detection;

[0039] Figure 3 is a flowchart of a method for determining the feature comparison time period in a time period;

[0040] Figure 4 is a flowchart of a method for determining comparison image feature points in a feature comparison time period;

[0041] Figure 5 is a flowchart for determining that the image device has a suspected change. Detailed Embodiments

[0042] To enable those skilled in the art to better understand the technical solutions in this specification, the following will clearly and completely describe the technical solutions in the embodiments of this specification in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all the embodiments. Based on the embodiments of this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this specification.

[0043] With the popularization of intelligent devices and Internet of Things technologies, image acquisition devices (such as cameras) are widely used in fields such as monitoring, navigation, and intelligent manufacturing. However, during the operation of the acquisition device, the installation position may shift due to external interference (such as vibration, collision, or manual adjustment), affecting the accuracy of image acquisition and even causing system anomalies. Therefore, there is an urgent need for an efficient and reliable method to dynamically monitor the installation status of image acquisition devices and issue early warnings in case of abnormal offsets. In the prior art, there is still a lack of a simple, easy-to-use, and highly robust solution.

[0044] The present invention proposes a method for monitoring the installation displacement and abnormal warning of image devices based on ORB feature detection. By analyzing the geometric transformation relationship (rotation angle, scaling ratio, and translation vector) between the initial installation state image and the current running state image of the image acquisition device, it is determined whether the device has shifted. When the transformation information exceeds the set threshold, an abnormal warning message is generated.

[0045] The implementation steps of this method include image preprocessing, feature extraction and matching, geometric transformation calculation, offset threshold judgment, and warning output. By superimposing offset information text on the matching image, the device status is visually displayed, and at the same time, automatic alarm output is supported to meet the intelligent operation and maintenance requirements of the device.

[0046] 1. Initial image acquisition and preprocessing

[0047] After the device is installed, use the camera to collect its initial state image and store it as a reference image.

[0048] Perform grayscale processing on the subsequently collected images to improve the calculation efficiency and robustness of ORB feature extraction and matching.

[0049] Grayscale processing: Convert the color image to a grayscale image. By reducing color information, subsequent calculations are simplified. The calculation formula is as follows: Where:

[0050] Igray: The converted grayscale value.

[0051] R, G, B: The red, green, and blue channel values of each pixel point in the original image.

[0052] Weights 0.299, 0.587, 0.114: Weighting values that simulate the sensitivity of the human eye to different colors.

[0053] Feature point detection and descriptor calculation

[0054] Use the ORB algorithm to perform feature point detection and descriptor calculation on the reference image and the currently captured image. The ORB algorithm combines the advantages of FAST corner detection and BRIEF descriptor calculation, has high real-time performance and stability, and is suitable for dynamic monitoring scenarios.

[0055] The ORB feature detection process consists of two parts:

[0056] FAST corner detection: Quickly identify corners in the image.

[0057] FAST (Features from Accelerated Segment Test) is a fast corner detection algorithm, and the detection process is as follows:

[0058] Set the pixel point p and its neighborhood with a radius of r.

[0059] There are 16 pixel points in the neighborhood: If at least N pixel points have significantly higher (or lower) brightness than the center point p, then p is identified as a corner.

[0060] Judgment formula for N pixel points: Satisfy the condition I(q)>I(p)+δ or I(q)<I(p)−δ

[0061] Among them:

[0062] I(p): Gray value of the center point.

[0063] I(q): Gray value of the neighborhood pixel point.

[0064] δ: Set threshold.

[0065] BRIEF descriptor calculation: Generate a short but unique binary descriptor to identify corners. BRIEF (Binary Robust Independent Elementary Features) is a binary descriptor that uses simple pixel intensity comparison for feature description. The generation process is as follows:

[0066] Define a fixed-size image patch (usually an S×S area) around the feature point

[0067] Select random pairs (pi, qi): Generate a binary string, and the formula is as follows: Among them:

[0068] pi, qi: Randomly selected pixel pairs.

[0069] I(pi), I(qi): Grayscale values of pixel points.

[0070] Binary descriptor: A string composed of 0s and 1s, serving as the feature vector of the feature point.

[0071] ORB adds rotational invariance during feature point detection, enabling it to maintain a high detection accuracy even when there is an angular offset.

[0072] Feature point matching and optimization

[0073] Use BFMatcher (Brute-Force Matcher) to match the feature descriptors of the reference image and the current image. The matching results are sorted according to the distance, and the top 50 optimal matching points are selected to reduce noise interference and improve computational stability.

[0074] BFMatcher is a brute-force matching algorithm used for matching feature point descriptors. The matching process is as follows:

[0075] Calculate the distance between two descriptors.

[0076] Find the matching pair with the minimum distance as the matching result of the feature point. The distance is calculated using the Hamming distance between descriptors, and its calculation formula is as follows: Where:

[0077] d1, d2: Two binary descriptors.

[0078] n: The length of the descriptor (usually 256 bits).

[0079] ⊕: Bitwise exclusive OR operation, calculating whether each bit is different.

[0080] The calculation of the matching distance uses the Hamming distance to measure the similarity of descriptors.

[0081] The Hamming distance is used to compare the differences between two binary strings, and the formula is as follows: Where:

[0082] A, B: Two binary strings.

[0083] n: The number of bits of the binary string.

[0084] ⊕: Bitwise exclusive OR operation.

[0085] Homography matrix calculation

[0086] Based on the matched feature points, the homography matrix between images is calculated by the RANSAC (Random Sample Consensus) algorithm. During the calculation, RANSAC randomly selects a set of matched points, solves the homography matrix, and improves the robustness of the calculation by excluding outliers.

[0087] The homography matrix H represents the perspective transformation relationship between two images and is defined as: Where:

[0088] H: After matrix decomposition, it can be used to extract rotation, scaling, and translation information;

[0089] x′, y′: Transformed image coordinates (coordinates of points on the target image);

[0090] x, y: Image coordinates before transformation (coordinates of points on the reference image);

[0091] w′: Normalization factor used to convert homogeneous coordinates to Cartesian coordinates.

[0092] Geometric transformation information extraction

[0093] Extract the first two rows and two columns of the homography matrix to form the rotation matrix R:

[0094] Where:

[0095] R: Rotation matrix, describing the rotation state of the image.

[0096] hij: Components related to rotation in the homography matrix.

[0097] Rotation angle: Calculate the rotation angle θ through the rotation matrix: Where:

[0098] θ: Rotation angle, in radians (can be converted to degrees later).

[0099] atan2(h21, h11): Arctangent function used to calculate the angle, considering the angle calculation formula in different quadrants.

[0100] Scaling ratio: Obtain Sx and Sy through the row vector norms of the rotation matrix: Where:

[0101] sx: Scaling factor in the x direction

[0102] sy: Scaling factor in the y direction.

[0103] Translation vector: Extract the translation vector T from the homography matrix: Where:

[0104] T: The translation vector, representing the translation state of the image.

[0105] h 13 ,h 23 : The parameter describing translation in the homography matrix.

[0106] 6. Offset Threshold Judgment

[0107] Set a predefined threshold range:

[0108] Rotation angle threshold: For example, within 5 degrees is the normal range.

[0109] Scaling ratio range: For example, from 0.95 to 1.05 indicates no significant scaling.

[0110] Translation threshold: For example, a translation amount within 10 pixels is normal.

[0111] According to the extracted information, compare the current calculated value with the threshold to determine whether it exceeds the set range to detect whether the device is offset.

[0112] Alarm and Visual Output

[0113] If the device status is abnormal, generate a warning message, output detailed offset parameters, such as rotation angle, scaling ratio, and translation vector. Overlay transformation information text on the matching image to visually display the offset situation. Real-time notify the operation and maintenance personnel through methods such as printing alarm prompts on the console, sound reminders, or message pushes.

[0114] Embodiment 1 To solve the above problems, according to one aspect of the present invention, as Figure 1 shown, a displacement anomaly warning system for an image device based on feature point detection is provided, specifically including:

[0115] Feature point screening module, matrix calculation and information extraction module, displacement recognition and warning module;

[0116] Among them, the feature point screening module is responsible for using the ORB algorithm to perform feature point detection and descriptor calculation on the reference image and the currently acquired image, and using the feature descriptors of the reference image and the current image to match to obtain the image comparison image feature points;

[0117] The matrix calculation and information extraction module is responsible for using the distance function to determine the feature point similarity coefficients of different comparison image feature points, and using the feature point similarity coefficients to match the feature points of different comparison image feature points, and constructing a homography matrix based on the matching feature points;

[0118] The displacement recognition and early warning module is responsible for extracting geometric transformation information based on the homography matrix to obtain the rotation angle, scaling ratio, and translation amount, and determining whether the image device has shifted and sending out an early warning signal in combination with a preset threshold.

[0119] Further, the feature point similarity coefficient is determined according to the Hamming distance function.

[0120] Specifically, when any one of the rotation angle, scaling ratio, and translation amount is greater than the preset threshold, it is determined that the image device has shifted and an early warning signal is sent out.

[0121] Embodiment 2 Second aspect, as Figure 2 shown, the present application provides an image device displacement anomaly early warning method based on feature point detection, which is applied to the above-mentioned image device displacement anomaly early warning system based on feature point detection, and specifically includes:

[0122] S1 Based on the analysis result of the monitoring image of the image device, determine the change situation of the light data at different times on different dates, and use the change situation to determine the feature comparison time period in the time period;

[0123] Further, the change situation of the light data is determined according to the brightness of the monitoring image of the image device in the time period.

[0124] Specifically, the light data is determined according to the preset light brightness corresponding to the brightness of the monitoring image.

[0125] Specifically, as Figure 3 shown, the method for determining the feature comparison time period in the time period is:

[0126] Determine the light brightness of the time period on different dates from the light data of the time period on different dates, and determine the reference light brightness with the average value of the light brightness of the time period on different dates;

[0127] Determine the light change date in the date according to the deviation amount between the light brightness on different dates and the reference light brightness;

[0128] Determine whether the time period is a feature comparison time period through the proportion of the number of light change dates.

[0129] Further, when the proportion of the number of light change dates is less than the preset proportion, it is determined that the time period is a feature comparison time period.

[0130] It can be understood that the light change date is the date when the deviation amount from the reference light brightness is not within the preset deviation amount range.

[0131] Optionally, the method for determining the feature comparison period in the time period is as follows:

[0132] Determine the light brightness of the time period on different dates from the light data of the time period on different dates, and determine the reference light brightness with the average value of the light brightness of the time period on different dates;

[0133] According to the deviation amount between the light brightness on different dates and the reference light brightness, calculate the average value of the deviation amounts on different dates;

[0134] Determine whether the time period is a feature comparison period through the average value of the deviation amounts on different dates.

[0135] Specifically, when the average value of the deviation amounts on different dates does not meet the requirements, it is determined that the time period is a feature comparison period.

[0136] In another embodiment, the method for determining the feature comparison period in the time period is as follows:

[0137] S11 Determine the light brightness of the time period on different dates from the light data of the time period on different dates, and determine the reference light brightness with the average value of the light brightness of the time period on different dates;

[0138] S12 Determine the light change date in the date according to the deviation amount between the light brightness on different dates and the reference light brightness;

[0139] S13 Determine the feature point comparison coefficient of the time period through the proportion of the number of light change dates and the reference light brightness, and use the feature point comparison coefficient to determine whether the time period is a feature comparison period.

[0140] Specifically, the method for determining the feature point comparison coefficient of the time period is as follows:

[0141] Based on the reference light brightness, determine the preset comparison coefficient corresponding to the reference light brightness;

[0142] Determine the feature point ratio coefficient of the time period according to the product of the proportion of the number of light change dates and the preset comparison coefficient.

[0143] Further, the value range of the feature point comparison coefficient is between 0 and 1. When the feature point comparison coefficient of the time period is less than the preset comparison coefficient threshold, it is determined that the time period is a feature comparison period.

[0144] Optionally, the above step S11 includes the following content:

[0145] S111 Determine the light brightness of the time period in different dates from the light data of the time period in different dates. Determine the reference light brightness based on the average value of the light brightness of the time period in different dates. When the reference light brightness is less than the preset brightness threshold, it is determined that the time period does not belong to the feature comparison time period. When the reference light brightness is not less than the preset brightness threshold, go to step S112;

[0146] S112 Based on the light brightness of the time period in different dates, when it is determined that there is a date with a light brightness less than the preset brightness threshold, go to step S112. When there is no date with a light brightness less than the preset brightness threshold, go to step S12;

[0147] S113 When the number of dates with a light brightness less than the preset brightness threshold is greater than the preset date number, it is determined that the time period does not belong to the feature comparison time period. When the number of dates with a light brightness less than the preset brightness threshold is not greater than the preset date number, go to step S12.

[0148] Optionally, the following content is included in the above step S12:

[0149] S121 Based on the deviation amount between the light brightness in different dates and the reference light brightness, when it is determined that there is no date with light change in the date, go to S122. When there is a date with light change in the date, go to step S123;

[0150] S122 When the average value of the deviation amount between the light brightness in different dates and the reference light brightness is less than the preset deviation amount threshold, it is determined that the time period belongs to the feature comparison time period. When the average value of the deviation amount between the light brightness in different dates and the reference light brightness is not less than the preset deviation amount threshold, it is determined that the time period does not belong to the feature comparison time period;

[0151] S123 Obtain the number of light change dates. When the number of light change dates is greater than the preset number of change dates, it is determined that the time period does not belong to the feature comparison time period. When the number of light change dates is not greater than the preset number of change dates, go to step S124;

[0152] S124 Determine the comprehensive deviation coefficient based on the proportion of the number of light change dates and the deviation amount of different light change dates. When the comprehensive deviation coefficient does not meet the requirements, it is determined that the time period does not belong to the feature comparison time period. When the comprehensive deviation coefficient meets the requirements, go to step S13.

[0153] S2 Obtain the change situation of different ORB image feature points in the monitoring image during the feature comparison period and the initial image feature points, and use the change situation of the image feature points to determine the comparison image feature points in the feature comparison period and the credibility coefficients of different comparison image feature points;

[0154] Further, the change situation of the ORB image feature points and the initial image feature points is determined according to the deviation amount between the ORB image feature points and the initial image feature points in the monitoring images corresponding to different dates.

[0155] Specifically, the initial image feature points are the image feature points of the reference image corresponding to the image device during installation.

[0156] Specifically, as Figure 4 shown, the method for determining the comparison image feature points in the feature comparison period is:

[0157] Determine the similarity coefficient between the ORB feature points and the initial image feature points on different dates according to the change situation of the ORB image feature points and the initial image feature points;

[0158] Based on the similarity coefficient with the initial image feature points, determine the average value of the similarity coefficients on different dates;

[0159] Determine the credibility coefficient of the ORB image feature points according to the average value of the similarity coefficients on different dates, and use the credibility coefficient to determine whether the ORB feature points are comparison image feature points.

[0160] Further, using the credibility coefficient to determine whether the ORB feature points are comparison image feature points specifically includes:

[0161] Take the ORB feature points whose credibility coefficients meet the requirements as comparison image feature points.

[0162] In another embodiment, the method for determining the comparison image feature points in the feature comparison period is:

[0163] Determine the similarity coefficient between the ORB feature points and the initial image feature points on different dates according to the change situation of the ORB image feature points and the initial image feature points;

[0164] Based on the similarity coefficient with the initial image feature points, determine the similar dates in these dates;

[0165] Determine the credibility coefficient of the ORB image feature points according to the proportion of the number of similar dates, and use the credibility coefficient to determine whether the ORB feature points are comparison image feature points.

[0166] Further, the similar date is the date with a similarity coefficient greater than the preset similarity coefficient to the initial image feature points.

[0167] Optionally, the method for determining the comparison image feature points in the feature comparison period is as follows:

[0168] S11 Determine the similarity coefficients of the ORB feature points on different dates to the initial image feature points based on the change situation between the ORB image feature points and the initial image feature points;

[0169] S12 Determine the similar dates in the dates based on the similarity coefficients to the initial image feature points, and determine the weight coefficients of different similar dates using the similarity coefficients of different similar dates;

[0170] S13 Determine the credibility coefficient of the ORB image feature points according to the proportion of the number of similar dates and the average value of the weight coefficients of different similar dates, and determine whether the ORB feature points are comparison image feature points using the credibility coefficient.

[0171] Specifically, the method for determining the credibility coefficient of the ORB image feature points is as follows:

[0172] Determine the average weight coefficient using the average value of the weight coefficients of different similar dates;

[0173] Determine the credibility coefficient of the ORB image feature points based on the product of the proportion of the number of similar dates and the average weight coefficient.

[0174] Optionally, the following content is included in step S11 above:

[0175] S111 Determine the similarity coefficients of the ORB feature points on different dates to the initial image feature points based on the change situation between the ORB image feature points and the initial image feature points. When there are dates with similarity coefficients not meeting the requirements, it is determined that the ORB image feature points do not belong to the comparison image feature points. When there are no dates with similarity coefficients not meeting the requirements, proceed to step S112;

[0176] S112 Determine the average value of the similarity coefficients of different dates using the similarity coefficients of the ORB feature points on different dates to the initial image feature points. When the average value of the similarity coefficients of different dates is less than the preset similarity coefficient threshold, it is determined that the ORB image feature points do not belong to the comparison image feature points. When the average value of the similarity coefficients of different dates is not less than the preset similarity coefficient threshold, proceed to step S113;

[0177] S113 uses the dates with similarity coefficients within a preset similarity coefficient range as the screened deviation dates. When the proportion of the number of the screened deviation dates does not meet the requirement, it is determined that the ORB image feature points do not belong to the comparison image feature points. When the average value of the similarity coefficients of different dates is not less than the preset similarity coefficient threshold and the proportion of the number of the screened deviation dates meets the requirement, it proceeds to step S12.

[0178] Optionally, the above step S12 includes the following content:

[0179] Based on the similarity coefficients with the initial image feature points, determine the similar dates in the dates, and use the similarity coefficients of different similar dates to determine the weight coefficients of different similar dates

[0180] S121 Based on the similarity coefficients with the initial image feature points, when it is determined that there are no similar dates in the dates, it is determined that the ORB image feature points do not belong to the comparison image feature points. When there are similar dates in the dates, it proceeds to step S122;

[0181] S122 Obtain the proportion of the number of similar dates in the dates. When the proportion of the number of similar dates is less than the preset proportion, it is determined that the ORB image feature points do not belong to the comparison image feature points. When the proportion of the number of similar dates is not less than the preset proportion, it proceeds to step S123;

[0182] S123 Use the similarity coefficients of different similar dates to determine the weight coefficients of different similar dates. When the sum of the weight coefficients of different similar dates is less than the preset weight coefficient threshold, it is determined that the ORB image feature points do not belong to the comparison image feature points. When the sum of the weight coefficients of different similar dates is not less than the preset weight coefficient threshold, it proceeds to step S13.

[0183] S3 Obtain the change situation of the comparison image feature points and the initial image feature points in the feature comparison period, and combine the credibility coefficients of different comparison image feature points to determine that when the image device has a suspected change, enter the next step;

[0184] Specifically, as Figure 5 shown, determining that the image device has a suspected change specifically includes:

[0185] Based on the change situation of the comparison image feature points and the initial image feature points, determine the deviation amounts of different comparison image feature points and the initial image feature points, and use the deviation amounts to determine the deviation coefficients of the comparison image feature points;

[0186] Determine the sum of deviation coefficients according to the sum of the products of the deviation coefficients and the credibility coefficients of different comparison image feature points;

[0187] Based on the deviation coefficient, determine whether the image device has a suspected change.

[0188] Furthermore, the deviation coefficient of the comparison image feature points is determined according to the product of the deviation amount and a preset proportionality factor.

[0189] It should be noted that when the sum of the deviation coefficients is greater than a preset deviation coefficient threshold, it is determined that the image device has a suspected change.

[0190] Optionally, determining that the image device has a suspected change specifically includes:

[0191] Based on the change situation between the comparison image feature points and the initial image feature points, determine the deviation amounts of different comparison image feature points from the initial image feature points. When the deviation amounts of different comparison image feature points from the initial image feature points are all within a preset deviation amount range, it is determined that the image device has not had a suspected change;

[0192] When there are comparison image feature points whose deviation amounts from the initial image feature points are not within the preset deviation amount range:

[0193] Take the comparison image feature points whose deviation amounts from the initial image feature points are not within the preset deviation amount range as deviation image feature points. When the number of the deviation image feature points is less than a preset number threshold:

[0194] Obtain the average value of the deviation amounts of different comparison image feature points from the initial image feature points. When the average value of the deviation amounts of different comparison image feature points from the initial image feature points meets the requirements, it is determined that the image device has a suspected change;

[0195] When the average value of the deviation amounts of different comparison image feature points from the initial image feature points does not meet the requirements or the number of the deviation image feature points is not less than the preset number threshold:

[0196] When the number of the deviation image feature points is greater than a preset feature point number threshold, it is determined that the image device has a suspected change;

[0197] When the number of the deviation image feature points is not greater than the preset feature point number threshold:

[0198] Based on the credibility coefficients of different deviation image feature points, determine the sum of the credibility coefficients. When the sum of the credibility coefficients is greater than a preset credibility coefficient threshold, it is determined that the image device has a suspected change;

[0199] When the sum of the credibility coefficients is not greater than the preset credibility coefficient threshold:

[0200] Determine the deviation coefficient of the comparison image feature points using the deviation amount, and determine the corrected deviation coefficients of different comparison image feature points according to the products of the deviation coefficients and credibility coefficients of different comparison image feature points. When there are no comparison image feature points with corrected deviation coefficients not meeting the requirements, it is determined that the image device has not undergone a suspected change;

[0201] When there are comparison image feature points with corrected deviation coefficients not meeting the requirements:

[0202] When the number of comparison image feature points with corrected deviation coefficients not meeting the requirements is greater than the preset deviation feature point quantity threshold, it is determined that the image device has undergone a suspected change;

[0203] When the number of comparison image feature points with corrected deviation coefficients not meeting the requirements is not greater than the preset deviation feature point quantity threshold:

[0204] Determine the change probability based on the corrected deviation coefficients of different comparison image feature points, and use the change probability to determine whether the image device has undergone a suspected change.

[0205] S4 Determine whether to issue a warning signal based on the change situations of comparison image feature points in different time periods.

[0206] Specifically, determining whether to issue a warning signal specifically includes:

[0207] Using the matching results of comparison image feature points in different time periods, determining the feature point similarity coefficients of different comparison image feature points using a distance function, and using the feature point similarity coefficients to match the feature points of different comparison image feature points;

[0208] Construct a homography matrix based on the matching feature points, extract geometric transformation information including rotation angle, scaling ratio, and translation amount from the homography matrix, and determine whether the image device is offset in combination with a preset threshold.

[0209] Furthermore, when any one of the rotation angle, scaling ratio, and translation amount in multiple time periods does not meet the requirements of the preset threshold, it is determined that the image device is offset, and a warning signal is output.

[0210] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments. In particular, for the device, equipment, and non - volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments.

[0211] The above description has been made of specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the acts or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the figures do not necessarily require the particular order or sequential order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0212] The above is only one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, various changes and modifications can be made to one or more embodiments of this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of one or more embodiments of this specification shall be included within the scope of the claims of this specification.

Claims

1. A method for early warning of abnormal displacement of image equipment based on feature point detection, characterized in that: Specifically include: Determine the change of light data in different time periods on different dates based on the analysis results of the monitoring images of the imaging device, and use the change to determine the feature comparison period in the time period; Using the ORB algorithm to perform feature point detection on the reference image and the monitoring image in the feature comparison period, obtaining changes in different ORB image feature points and reference image feature points in the monitoring image in the feature comparison period, and using the changes in the image feature points to determine the credibility coefficients of the comparison image feature points and different comparison image feature points in the feature comparison period; Obtaining the changes of the feature points of the comparison image and the feature points of the reference image during the feature comparison period, and combining the credibility coefficients of different feature points of the comparison image to determine that the image device has a suspected change, and then proceeding to the next step; Determine whether to issue a warning signal based on the changes in the feature points of the compared images in different time periods.

2. The method for warning abnormal displacement of an image device based on feature point detection as described in claim 1, wherein the change in the light data is determined based on the brightness of the monitoring image of the image device during the time period.

3. The method for warning abnormal displacement of an image device based on feature point detection as described in claim 1, wherein the light data is determined based on a preset light brightness corresponding to the brightness of the monitored image.

4. The method for early warning of abnormal displacement of an image device based on feature point detection according to claim 1, wherein the method for determining the feature comparison period in the period is: Determine the light brightness of the time periods on different dates by using the light data of the time periods on different dates, and determine the reference light brightness by using the average light brightness of the time periods on different dates; Determining the light change date among the dates according to the deviation between the light brightness on different dates and the reference light brightness; Whether the time period is a feature comparison time period is determined by the number ratio of the light change dates.

5. The method for warning abnormal displacement of an image device based on feature point detection as described in claim 4, wherein when the proportion of the number of light change dates is less than the preset proportion, the time period is determined to be a feature comparison period.

6. The method for early warning of abnormal displacement of an image device based on feature point detection according to claim 1, wherein the method comprises: determining whether to issue an early warning signal; and Using the matching results of the feature points of the comparison images in different time periods, using the distance function to determine the feature point similarity coefficients of different feature points of the comparison images, and using the feature point similarity coefficients to match the feature points of different feature points of the comparison images; A homography matrix is ​​constructed based on the matching feature points, geometric transformation information is extracted based on the homography matrix to obtain a rotation angle, a scaling ratio, and a translation amount, and a preset threshold is combined to determine whether the image device is offset.

7. The method for warning abnormal displacement of an image device based on feature point detection as described in claim 6 has the characteristic point that when any one of the rotation angle, zoom ratio and translation amount in multiple time periods does not meet the requirements of a preset threshold, it is determined that the image device is offset and an early warning signal is output.

Images