A method and device for motion detection based on feature point analysis

Abstract

The present invention relates to the field of motion detection, and specifically discloses a method and device for motion detection based on feature point analysis, including the following steps: a. Using the SIFT algorithm to extract feature points in an image; b. For each feature point, extracting pixel information within its neighborhood, including gradient, color, local contrast, and texture features, to enrich the description of the feature point; c. Fusing feature point information from different sensors or image sources, calculating the fusion weight for each feature point, and performing weighted averaging on the feature points based on these weights to achieve effective fusion of multi-source information; The present invention effectively integrates multi-source information by fusing feature point information from different sensors or image sources, calculating fusion weights and performing weighted averaging, improving the reliability and comprehensiveness of motion detection. This fusion strategy can make full use of the advantages of different sensors or image sources to make up for the deficiencies of single-source information.

Description

Technical Field

[0001] The present invention belongs to the field of motion detection, and particularly relates to a method and device for motion detection based on feature point analysis. Background Art

[0002] The motion detection method based on feature point analysis is an important research direction in the field of computer vision. It realizes the recognition, positioning, and tracking of moving targets by detecting, tracking, and analyzing feature points in images. With the continuous development of computer technology, this method has been widely applied in fields such as video surveillance, human-computer interaction, and autonomous driving.

[0003] In practical applications, motion detection often involves information from multiple sensors or image sources. However, in the prior art, the integration of multi-source information often encounters difficulties, resulting in insufficient utilization of information. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a method for motion detection based on feature point analysis to solve the problem that the integration of multi-source information in the prior art often encounters difficulties, resulting in insufficient utilization of information.

[0005] A method for motion detection based on feature point analysis includes the following steps:

[0006] a. Using the SIFT algorithm, extract feature points in the image;

[0007] b. For each feature point, extract the pixel information in its neighborhood, including gradient, color, local contrast, and texture features, to enrich the description of the feature point;

[0008] c. Fuse the feature point information from different sensors or image sources, calculate the fusion weight of each feature point, and perform weighted averaging on the feature points based on these weights to achieve effective fusion of multi-source information;

[0009] d. Based on the fused feature points and their neighborhood information, use the matching algorithm to determine the correspondence of feature points between consecutive frames;

[0010] e. On the basis of feature point matching, combine the block matching algorithm and use the pixel information in the neighborhood of the feature points for motion estimation;

[0011] f. Use the Kalman filtering algorithm to perform real-time tracking on the moving target, update the target state according to the observed value and the predicted value to improve the stability and accuracy of tracking;

[0012] g. According to the motion vector, perform fine tracking on the moving target, including target recognition, trajectory drawing, and speed measurement;

[0013] h. Objectively evaluate the tracking results through the accuracy evaluation formula to verify the accuracy of tracking;

[0014] i. Post-process the motion estimation results to improve the accuracy and integrity of motion detection.

[0015] Preferably, in step b, the extraction of the feature point neighborhood information further includes calculating the local contrast of the pixels in the neighborhood to further enhance the stability and distinctiveness of the feature points.

[0016] Preferably, in step c, the calculation formula for the fusion weight of the feature points is as follows:

[0017]

[0018] where ω i is the fusion weight of the i-th sensor;

[0019] d i is the distance between the target detected by the i-th sensor and the sensor;

[0020] μ is the average value of the target distances detected by all sensors;

[0021] σ is the standard deviation of the target distances detected by all sensors;

[0022] SNR i is the signal-to-noise ratio of the i-th sensor;

[0023] n is the total number of sensors;

[0024] The calculation formula for the target position after multi-source information fusion is as follows:

[0025]

[0026] Ρ is the estimated target position after fusion;

[0027] ω i is the fusion weight of the i-th sensor;

[0028] Ρ i is the target position detected by the i-th sensor.

[0029] Preferably, in step d, the feature point matching algorithm adopts the RANSAC algorithm, and the inliers that best fit the model are estimated from the matching point pairs through iteration to improve the accuracy and robustness of the matching and reduce the possibility of false matching.

[0030] Preferably, in step e, the selection and parameter setting of the block matching algorithm are adaptively adjusted according to the application scenario and image characteristics to achieve more refined tracking of the moving target and more accurate motion estimation.

[0031] Preferably, in step f, the update formula of the Kalman filter algorithm is as follows:

[0032] mathbfX k|k = X k|k-1 + K k (Z k - Η k X k|k-1 )

[0033]

[0034] mathbfΡ k|k =(Ι - K k Η k )Ρ k|k-1 ;

[0035] Among them, X k|k is the state estimate at time k;

[0036] X k|k-1 is the state predicted at time k - 1 for time k;

[0037] K k is the Kalman gain matrix;

[0038] Z k is the observation value at time k;

[0039] Η k is the matrix that maps the state vector to the observation vector;

[0040] R k is the covariance matrix of the observation noise;

[0041] Ρ k|k is the covariance matrix of the estimation error at time k;

[0042] Ρ k|k-1 is the error covariance matrix predicted at time k - 1 for time k;

[0043] Ι is the identity matrix.

[0044] Preferably, in step h, the accuracy evaluation formula for moving target tracking is as follows:

[0045]

[0046] Among them, Accuracy k is the tracking accuracy at time k;

[0047] X k|k is the state estimate at time k;

[0048] X trueThe true state vector of the target.

[0049] Preferably, it further includes performing trajectory analysis, speed change analysis, acceleration analysis, and interaction analysis between targets on the moving target based on the trajectory and speed measurement data obtained in step g to extract useful behavior information.

[0050] An apparatus for motion detection based on feature point analysis, comprising:

[0051] A feature point extraction module for extracting feature points in an image using the SIFT algorithm;

[0052] A feature description module for extracting pixel information within the neighborhood of each feature point, including gradient, color, local contrast, and texture features, to enrich the description of the feature point;

[0053] An information fusion module for fusing feature point information from different sensors or image sources, calculating the fusion weight of each feature point, and performing weighted averaging on the feature points based on these weights to achieve effective fusion of multi-source information;

[0054] A feature point matching module for determining the correspondence of feature points between consecutive frames based on the fused feature points and their neighborhood information using a matching algorithm;

[0055] A motion estimation module for performing motion estimation using the pixel information within the neighborhood of the feature points in combination with the block matching algorithm based on feature point matching;

[0056] A real-time tracking module for performing real-time tracking of the moving target using the Kalman filtering algorithm, updating the target state based on the observed value and the predicted value to improve the stability and accuracy of tracking;

[0057] A fine tracking module for performing fine tracking of the moving target based on the motion vector, including target recognition, trajectory drawing, and speed measurement;

[0058] An accuracy evaluation module for objectively evaluating the tracking result through an accuracy evaluation formula to verify the accuracy of tracking;

[0059] A post-processing module for post-processing the motion estimation result to improve the accuracy and integrity of motion detection;

[0060] A behavior analysis module for performing trajectory analysis, speed change analysis, acceleration analysis, and interaction analysis between targets on the moving target based on the trajectory and speed measurement data obtained in the fine tracking module to extract useful behavior information.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] By fusing the feature point information from different sensors or image sources, calculating the fusion weights and performing weighted averaging, the multi-source information is effectively integrated, improving the reliability and comprehensiveness of motion detection. This fusion strategy can make full use of the advantages of different sensors or image sources to make up for the deficiencies of single-source information;

[0063] Based on feature point matching, a block matching algorithm is combined for motion estimation, and the Kalman filtering algorithm is used to track moving targets in real time. This combined strategy not only improves the stability of tracking but also significantly enhances the tracking accuracy, making the trajectories of moving targets more accurate and continuous. Brief Description of the Drawings

[0064] Figure 1 It is a schematic flowchart of the method of the present invention;

[0065] Figure 2 It is a diagram of the algorithm execution steps of the present invention. Detailed Embodiments

[0066] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0067] As Figures 1 to 2 shown:

[0068] Embodiment 1: The present invention provides a method for motion detection based on feature point analysis, including the following steps:

[0069] a. Using the SIFT algorithm, extract feature points in the image;

[0070] b. For each feature point, extract the pixel information in its neighborhood, including gradient, color, local contrast, and texture features, to enrich the description of the feature point;

[0071] c. Fuse the feature point information from different sensors or image sources, calculate the fusion weight of each feature point, and perform weighted averaging on the feature points based on these weights to achieve effective fusion of multi-source information;

[0072] d. Based on the fused feature points and their neighborhood information, use the matching algorithm to determine the corresponding relationship of feature points between consecutive frames;

[0073] e. Based on feature point matching, combine the block matching algorithm, and use the pixel information in the neighborhood of the feature point for motion estimation;

[0074] f. Use the Kalman filter algorithm to perform real-time tracking on the moving target, update the target state according to the observed value and the predicted value, so as to improve the stability and accuracy of the tracking;

[0075] g. According to the motion vector, perform fine tracking on the moving target, including target recognition, trajectory drawing, and speed measurement;

[0076] h. Objectively evaluate the tracking result through the accuracy evaluation formula to verify the accuracy of the tracking;

[0077] i. Post-process the motion estimation result to improve the accuracy and integrity of the motion detection;

[0078] Among them, this method first uses the SIFT algorithm to extract feature points from the image, and extracts the neighborhood pixel information of each feature point in detail, such as gradient, color, local contrast, and texture features, to enhance the descriptiveness of the feature points; subsequently, it cleverly fuses the feature point information from different sensors or image sources, and through calculating the fusion weights and performing weighted averaging, realizes the effective integration of multi-source information; on the basis of feature point matching, combines the block matching algorithm for motion estimation, and uses the Kalman filter algorithm to perform real-time tracking on the moving target, improving the stability and accuracy of the tracking; finally, performs fine tracking on the moving target according to the motion vector, including target recognition, trajectory drawing, and speed measurement, and objectively evaluates the tracking result through the accuracy evaluation formula; in addition, it also includes post-processing the motion estimation result to further improve the accuracy and integrity of the motion detection.

[0079] Specifically, in step b, the extraction of the feature point neighborhood information also includes calculating the local contrast of the pixels in the neighborhood to further enhance the stability and distinctiveness of the feature points.

[0080] Specifically, in step c, the calculation formula for the fusion weight of the feature points is as follows:

[0081]

[0082] where ω i is the fusion weight of the i-th sensor;

[0083] d i is the distance between the target detected by the i-th sensor and the sensor;

[0084] μ is the average value of the target distances detected by all sensors;

[0085] σ is the standard deviation of the target distances detected by all sensors;

[0086] SNR i is the signal-to-noise ratio of the i-th sensor;

[0087] n is the total number of sensors;

[0088] The calculation formula for the target position after multi-source information fusion is as follows:

[0089]

[0090] Ρ is the estimated target position after fusion;

[0091] ω i is the fusion weight of the i-th sensor;

[0092] Ρ i is the target position detected by the i-th sensor.

[0093] As can be seen from the above, by calculating the local contrast of pixels in the neighborhood of feature points, this method not only enriches the information of feature points, but also improves the stability and distinctiveness of feature points, which is crucial for subsequent feature point matching and motion estimation; in step c, a scientific calculation formula for the fusion weight of feature points is adopted, comprehensively considering factors such as the distance between the sensor and the target, the average value and standard deviation of the target distances detected by all sensors, and the signal-to-noise ratio of the sensor, so as to be able to calculate the fusion weight of each feature point more accurately; based on these weights, weighted averaging of feature points is performed to achieve effective fusion of multi-source information and improve the estimation accuracy of the target position; therefore, the motion detection method of the present invention has been optimized and innovated in terms of feature point description and fusion, providing a more accurate and stable basis for subsequent matching, estimation, and tracking steps.

[0094] Embodiment 2: This embodiment is basically the same as the previous embodiment, the difference is that in step d, the feature point matching algorithm uses the RANSAC algorithm, and the inliers that best fit the model are estimated from the matching point pairs in an iterative manner to improve the accuracy and robustness of the matching and reduce the possibility of false matching;

[0095] Among them, the specific application process of the RANSAC algorithm is as follows:

[0096] Randomly select a subset: Randomly select a part of the points from all the matching point pairs as a subset for fitting a model (such as a homography matrix or an essential matrix).

[0097] Model fitting: Use the selected subset to fit a model that can describe the relationship between the matching point pairs.

[0098] Inlier judgment: For the remaining points, calculate their distances from the fitted model, and the points with distances less than a certain threshold are considered inliers (i.e., points that conform to the model), otherwise they are considered outliers (i.e., points that do not conform to the model).

[0099] Iteration and Selection: Repeat the above process multiple times (i.e., iterate), and each time select a different subset for model fitting and inlier judgment. Finally, select the model with the largest number of inliers as the ultimately fitted model.

[0100] Through the RANSAC algorithm, the inliers that best fit the model can be estimated from a large number of mismatched matching point pairs, thereby improving the accuracy and robustness of the matching and reducing the possibility of mismatches.

[0101] Specifically, in step e, the selection and parameter settings of the block matching algorithm are adaptively adjusted according to the application scenario and image characteristics to achieve more refined tracking of the moving target and more accurate motion estimation.

[0102] Specifically, in step f, the update formula of the Kalman filtering algorithm is as follows:

[0103] mathbfX kk = X kk-1 + K k (Z k - Η k X kk-1 )

[0104]

[0105] mathbfΡ kk =(Ι - K k Η k )Ρ kk-1 ;

[0106] Among them, X kk is the state estimate at time k;

[0107] X kk-1 is the predicted state at time k from time k - 1;

[0108] K k is the Kalman gain matrix;

[0109] Z k is the observation value at time k;

[0110] Η k is the matrix that maps the state vector to the observation vector;

[0111] R k is the covariance matrix of the observation noise;

[0112] Ρ kk is the covariance matrix of the estimation error at time k;

[0113] Ρ kk-1 is the error covariance matrix predicted at time k from time k - 1;

[0114] Ι is the identity matrix.

[0115] Specifically, in step h, the accuracy evaluation formula for moving target tracking is as follows:

[0116]

[0117] where Accuracy k is the tracking accuracy at time k;

[0118] X kk is the state estimation at time k;

[0119] X true is the true state vector of the target.

[0120] Specifically, it also includes performing trajectory analysis, speed change analysis, acceleration analysis, and interaction analysis between targets on the moving target based on the trajectory and speed measurement data obtained in step g to extract useful behavior information.

[0121] As can be seen from the above, this embodiment adopts the RANSAC algorithm, which filters out the inliers that best fit the model from a large number of matching point pairs through an iterative method, significantly improving the accuracy and robustness of the matching and effectively reducing the false matching rate; in the selection of the block matching algorithm, adaptive adjustment is achieved, and the algorithm and parameters are flexibly selected according to the actual application scenario and image characteristics, thus realizing more refined tracking and more accurate motion estimation of the moving target; in the application of the Kalman filter algorithm, the algorithm update formula is given in detail, and the real-time update of the state of the moving target is realized by comprehensively considering the observed value and the predicted value, improving the stability and accuracy of the tracking; in addition, the accuracy evaluation formula for moving target tracking is introduced, and the accuracy of the tracking result is objectively evaluated by comparing the state estimation with the true state vector; finally, steps such as trajectory analysis, speed change analysis, acceleration analysis, and interaction analysis between targets on the moving target are added to extract useful behavior information, providing strong support for the in-depth understanding and analysis of the moving target.

[0122] Embodiment 3: A device for motion detection based on feature point analysis, comprising:

[0123] A feature point extraction module for extracting feature points in an image using the SIFT algorithm;

[0124] A feature description module for extracting pixel information in the neighborhood of each feature point, including gradient, color, local contrast, and texture features, to enrich the description of the feature point;

[0125] An information fusion module, which is used to fuse the feature point information from different sensors or image sources, calculate the fusion weights of each feature point, and perform weighted averaging on the feature points based on these weights to achieve the effective fusion of multi-source information;

[0126] A feature point matching module, which is used to determine the correspondence of feature points between consecutive frames based on the fused feature points and their neighborhood information by using a matching algorithm;

[0127] A motion estimation module, which is used to perform motion estimation by combining a block matching algorithm and using the pixel information within the neighborhood of the feature points based on the feature point matching;

[0128] A real-time tracking module, which is used to perform real-time tracking on a moving target by using the Kalman filtering algorithm, and update the target state according to the observed value and the predicted value to improve the stability and accuracy of the tracking;

[0129] A fine tracking module, which is used to perform fine tracking on the moving target according to the motion vector, including the recognition of the target, the drawing of the trajectory, and the measurement of the speed;

[0130] An accuracy evaluation module, which is used to objectively evaluate the tracking result through an accuracy evaluation formula to verify the accuracy of the tracking;

[0131] A post-processing module, which is used to post-process the motion estimation result to improve the accuracy and integrity of the motion detection;

[0132] A behavior analysis module, which is used to perform trajectory analysis, speed change analysis, acceleration analysis, and interaction analysis between targets on the moving target based on the trajectory and speed measurement data obtained in the fine tracking module to extract useful behavior information.

[0133] As can be seen from the above, this embodiment introduces a motion detection device based on feature point analysis, which consists of multiple modules. Each module undertakes a specific function and collaborates together to achieve efficient and accurate motion detection. The feature point extraction module uses the SIFT algorithm to extract feature points from images, providing a basis for subsequent processing. The feature description module then describes each feature point in detail, including information such as gradients, colors, local contrasts, and texture features within its neighborhood, to enrich the representation of feature points. The information fusion module can fuse the feature point information from different sensors or image sources, and through calculating the fusion weights of each feature point and performing weighted averaging, it realizes the effective integration of multi-source information. The feature point matching module, based on the fused feature points and their neighborhood information, uses a matching algorithm to determine the corresponding relationships of feature points between consecutive frames. The motion estimation module, on the basis of feature point matching and in combination with the block matching algorithm, performs motion estimation on the pixel information within the neighborhood of feature points. The real-time tracking module uses the Kalman filtering algorithm to perform real-time tracking of moving targets, improving the stability and accuracy of tracking. The fine tracking module performs fine tracking of moving targets according to motion vectors, including target recognition, trajectory drawing, and speed measurement, etc. The accuracy evaluation module objectively evaluates the tracking results to verify the accuracy of tracking. The post-processing module further optimizes the motion estimation results to improve the accuracy and integrity of motion detection. Finally, the behavior analysis module, based on the data of the fine tracking module, analyzes the trajectories, speeds, accelerations, and interactions between targets of moving targets, and extracts useful behavior information.

[0134] The standard parts used in the present invention can all be purchased from the market. The special-shaped parts can be customized according to the descriptions in the specification and the drawings. The specific connection methods of each part all adopt conventional means such as bolts, rivets, and welding that are mature in the prior art. The machines, parts, and equipment all adopt conventional models in the prior art. Coupled with the circuit connection adopting the conventional connection method in the prior art, details are not described herein again. The content not described in detail in this specification belongs to the prior art well-known to those skilled in the art.

[0135] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. The meaning of "plurality" is two or more, unless otherwise specifically defined.

[0136] In the present invention, unless otherwise clearly defined or limited, terms such as "installed", "connected", "coupled", "fixed", etc. shall be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or integrated; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0137] In the present invention, unless otherwise clearly defined or limited, the first feature being "on" or "under" the second feature may be that the first and second features are in direct contact, or the first and second features are indirectly in contact through an intermediate medium. Moreover, the first feature being "above", "over" and "on top of" the second feature may be that the first feature is directly above or obliquely above the second feature, or merely indicates that the first feature has a higher horizontal height than the second feature. The first feature being "under", "beneath" and "underneath" the second feature may be that the first feature is directly below or obliquely below the second feature, or merely indicates that the first feature has a lower horizontal height than the second feature.

[0138] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0139] In the drawings of the disclosed embodiments of the present invention, only the structures related to the disclosed embodiments are involved, and other structures can refer to the general design. Without conflict, the same embodiment and different embodiments of the present invention can be combined with each other.

[0140] Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. 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 method for motion detection based on feature point analysis, characterized in that, It includes the following steps: a. Use the SIFT algorithm to extract feature points in the image; b. For each feature point, extract the pixel information within its neighborhood, including gradient, color, local contrast, and texture features, to enrich the description of the feature point; c. Fuse the feature point information from different sensors or image sources, calculate the fusion weight for each feature point, and perform weighted averaging on the feature points based on these weights to achieve effective fusion of multi-source information; d. Based on the fused feature points and their neighborhood information, use the matching algorithm to determine the correspondence of feature points between consecutive frames; e. On the basis of feature point matching, combine the block matching algorithm and use the pixel information within the feature point neighborhood for motion estimation; f. Use the Kalman filtering algorithm to perform real-time tracking of the moving target, and update the target state according to the observed value and the predicted value to improve the stability and accuracy of tracking; g. According to the motion vector, perform fine tracking of the moving target, including target recognition, trajectory plotting, and speed measurement; h. Objectively evaluate the tracking result through the accuracy evaluation formula to verify the accuracy of tracking; i. Post-process the motion estimation result to improve the accuracy and integrity of motion detection.

2. The method for motion detection based on feature point analysis according to claim 1, wherein In step b, the extraction of feature point neighborhood information also includes calculating the local contrast of pixels within the neighborhood to further enhance the stability and distinctiveness of the feature points.

3. The method for motion detection based on feature point analysis according to claim 2, wherein In step c, the calculation formula for the fusion weight of feature points is as follows: where ω i is the fusion weight of the i-th sensor; d i is the distance between the target detected by the i-th sensor and the sensor; μ is the average value of the target distances detected by all sensors; σ is the standard deviation of the target distances detected by all sensors; SNR i is the signal-to-noise ratio of the i-th sensor; n is the total number of sensors; The calculation formula for the target position after multi-source information fusion is as follows: Ρ is the estimated target position after fusion; ω i is the fusion weight of the i-th sensor; Ρ i is the target position detected by the i-th sensor.

4. The method for motion detection based on feature point analysis according to claim 3, wherein In step d, the feature point matching algorithm adopts the RANSAC algorithm, and estimates the inliers that best conform to the model from the matching point pairs through iteration to improve the accuracy and robustness of matching and reduce the possibility of false matching.

5. The method for motion detection based on feature point analysis according to claim 4, wherein In step e, the selection and parameter setting of the block matching algorithm are adaptively adjusted according to the application scenario and image characteristics to achieve more fine-grained tracking of the moving target and more accurate motion estimation.

6. The method for motion detection based on feature point analysis according to claim 5, characterized in that, In step f, the update formula of the Kalman filtering algorithm is as follows: mathbfX k|k = X k|k-1 + K k (Z k - Η k X k|k-1 ) mathbfΡ k|k =(Ι-K k Η k )Ρ k|k-1 ; where X k|k is the state estimate at time k; X k|k-1 Predict the state at time k from the state at time k-1; K k is the Kalman gain matrix; Z k is the observed value at time k; H k is the matrix that maps the state vector to the observation vector; R k is the covariance matrix of the observation noise; Ρ k|k is the covariance matrix of the estimation error at time k; Ρ k|k-1 is the error covariance matrix for predicting the k-th moment at the (k-1)-th moment; Ι is the identity matrix.

7. The method for motion detection based on feature point analysis according to claim 6, characterized in that, In step h, the accuracy evaluation formula for moving target tracking is as follows: Among them, Accuracy k is the tracking accuracy at time k; X k|k is the state estimate at time k; X true is the true state vector for the target.

8. The method for motion detection based on feature point analysis according to claim 1, wherein It also includes performing trajectory analysis, speed change analysis, acceleration analysis, and interaction analysis between targets on the moving target based on the trajectory and speed measurement data obtained in step g to extract useful behavior information.

9. An apparatus for motion detection based on feature point analysis, characterized in that, It includes: A feature point extraction module for extracting feature points in the image using the SIFT algorithm; A feature description module for extracting the pixel information within the neighborhood of each feature point, including gradient, color, local contrast, and texture features, to enrich the description of the feature point; An information fusion module for fusing the feature point information from different sensors or image sources, calculating the fusion weight for each feature point, and performing weighted averaging on the feature points based on these weights to achieve effective fusion of multi-source information; A feature point matching module for determining the correspondence of feature points between consecutive frames based on the fused feature points and their neighborhood information using the matching algorithm; A motion estimation module, which is used to perform motion estimation by combining a block matching algorithm based on feature point matching and using pixel information within the neighborhood of feature points; A real-time tracking module, which is used to perform real-time tracking of a moving target by using the Kalman filtering algorithm, and update the target state according to the observed value and the predicted value to improve the stability and accuracy of tracking; A fine tracking module, which is used to perform fine tracking of a moving target according to the motion vector, including target recognition, trajectory drawing, and speed measurement; An accuracy evaluation module, which is used to objectively evaluate the tracking result through an accuracy evaluation formula to verify the accuracy of tracking; A post-processing module, which is used to post-process the motion estimation result to improve the accuracy and integrity of motion detection; A behavior analysis module, which is used to perform trajectory analysis, speed change analysis, acceleration analysis, and interaction analysis between targets on the moving target based on the trajectory and speed measurement data obtained in the fine tracking module to extract useful behavior information.

Images