29 results about "Correlation filter" patented technology

An unmanned aerial vehicle target detection tracking method based on intelligent image matching comprises the steps that an unmanned aerial vehicle collects images to make a data set, a model file is obtained through twin neural network training, the model file is loaded to detect a video stream, a current frame target position and detection frame information are obtained, whether the frame is a first frame is judged, and if yes, the target is detected; using a target position information result to initialize a kernel filtering algorithm, and constructing a measurement vector; if not, calculating correlation by using a kernel correlation filter, taking the highest response position as a prediction bit, and constructing a measurement vector; calculating a Kalman gain, and inputting the Kalman gain into a Kalman filter to obtain a predicted state estimated value = (); comparing y-axis deviations of the detector and the tracker, and setting a threshold value for judgment; according to the method, target tracking under resource limitation is realized through light-weight twin network detection in combination with kernel filtering and improved Kalman filtering, the problems of high target loss rate and low precision are solved, and the method is suitable for the fields of military reconnaissance, intelligent autonomous attack and the like.

The application discloses a structure interlayer displacement automatic measurement method and system based on visual data and belongs to the technical field of structure displacement measurement. The measurement method comprises the following steps: acquiring video data in the vibration process of a building based on a monitoring camera; processing the video data by using an improved kernel correlation filtering algorithm to extract building interlayer displacement information; verifying the extracted displacement information and outputting building interlayer displacement results. On the basis of a classical kernel correlation filtering target tracking method, the application utilizes image pyramid, robust feature extraction and discrete Fourier transform up-sampling technology to expand the tracking range, enhance the edge features and improve the tracking precision; the application fully utilizes the existing monitoring camera equipment to measure the interlayer displacement, does not need to additionally arrange sensors and reduces the cost budget.

This invention relates to an object detection device. The object detection device (1) includes: a drive signal generation unit (5) that generates a drive signal including frequency modulation; a first correlation filter (62) that performs correlation detection between the received signal and a first reference signal corresponding to the drive signal; a first determination unit (8) that determines whether the received wave is a reflected wave of a detection wave transmitted from a transmitting unit (40A) based on the correlation signal of the first correlation filter; a second correlation filter (63) that uses a signal corresponding to a portion of the drive signal as a second reference signal and performs correlation detection between the received signal and the second reference signal; a third correlation filter (64) that uses a signal including a portion of the drive signal with a frequency higher than the second reference signal as a third reference signal and performs correlation detection between the received signal and the third reference signal, wherein the third reference signal is a signal corresponding to a portion of the drive signal; and a second determination unit (9) that determines whether an object is within the detection range based on the correlation signals of the second and third correlation filters.

The object detection device of the present application is provided with: a transceiver (4) that transmits an ultrasonic wave encoded by frequency modulation and receives the ultrasonic wave and outputs a reception signal; a first quadrature detector (61) that generates a complex reception signal by quadrature detection of the reception signal and outputs it; a second quadrature detector (71) that generates a complex reference signal by quadrature detection of a reference signal and outputs it; a correlation filter (62) that performs correlation detection of the complex reception signal and the complex reference signal and outputs a correlation signal; and a code determination unit (8) that determines a code contained in the reception signal based on the correlation signal.

A signal processing method for processing audio signals and in particular for adjusting the tone of music signals is disclosed. The method provides adaptive tonality control. Here, a shape dependent filter magnitude response is calculated, a level dependent filter magnitude response is calculated, a result response is calculated based on the shape and level dependent filter magnitude responses, and an output signal is calculated by convolving an input signal with the result response. In particular, the calculation of the shape dependent filter magnitude response comprises at least the steps of calculating a magnitude spectrum of a signal, filtering the resulting magnitude spectrum using a first smoothing filter, applying a predetermined weighting curve to the spectrum, detrending the weighted spectrum by fitting a curve to said weighted spectrum and subtracting said fitted curve from said weighted spectrum, and inverting the resulting spectrum to form the shape dependent filter magnitude response.

The application relates to a target tracking method for a high-speed rotating platform, which comprises the following steps: acquiring an initial frame image for target tracking; acquiring a next frame image, performing image rotation according to a rotation angle existing in the initial frame image, and obtaining a rotated image eliminating the rotation angle; taking a target point coordinate of a previous frame image as a center, adopting a correlation filtering model to perform target detection in the rotated image, and obtaining a target point coordinate in the rotated image; converting the target point coordinate in the rotated image to a target point coordinate in a pre-rotated image, and taking the target point coordinate in the pre-rotated image as a center to obtain a search region; updating the correlation filtering model, and performing target tracking on a next frame image; and further performing target tracking on all subsequent frame images. The application realizes stable tracking in the high-speed rotating process of the image.

The application discloses a visual target tracking system and method based on a nano unmanned aerial vehicle, aiming at the problem that a traditional kernel correlation filter tracking algorithm cannot adapt to target size change, introduces a scale filter, so that the tracking algorithm can adaptively adjust the size and position of a target frame according to the change of the target scale; a method based on time context information is adopted to realize template updating, so that the tracking template has time correlation and can more accurately reflect the change of the target in appearance and motion state; a long-short-term tracking strategy is adopted to combine the information of short-term tracking and long-term tracking, so that the target tracker can still maintain high precision and robustness in a long-time tracking task, and the real-time requirement is met; the application is experimented on a Crazyflie2.1 nano quadrotor platform, and realizes stable and rapid target tracking control on the nano quadrotor unmanned aerial vehicle.

A sensor circuit (620) for measuring physical quantities, comprising: a signal acquisition circuit (640) including sensors (621, 622) for providing input signals (v1, v2) related to the physical quantity; and a processing circuit (650) for receiving the input signals (v1, v2) and providing an output signal (θ) representing the physical quantity. 校正 The processing circuitry includes a closed loop comprising: a first sub-circuit (660) arranged to receive the input signals (v1, v2) and feedback signals (fbi, fbq), and configured to provide a first signal (θ). 估计 ); a frequency correlation filter (632) for receiving and filtering the first signal (θ) 估计 ), and is used to provide the output signal (θ) 校正 ); a second sub-circuit (670), which is used to receive the filtered signal (θ) 校正 ) and use a nonlinear function to filter the signal (θ) 校正 ) are converted into the feedback signals (fbi, fbq).

The application discloses a target tracking method and system combining correlation filtering and potential matching, and relates to the technical field of computer vision. The method generates a correlation filter and an image template by using a target image, and generates a boundary potential field, a depth field and a mask template of a point cloud template by using a target point cloud. Then, the method performs correlation operation on the image template and an environment image by using the correlation filter to obtain a potential region of the target. Then, the boundary potential field and the depth field in the potential region are rasterized, and local statistical information is calculated. Finally, potential matching is performed on the point cloud template and the potential region by using the local statistical information to obtain an accurate position of the target. The application fuses a visual sensor and a laser radar, and has high tracking accuracy and stability in scenes such as target deformation, illumination change and motion blur. Meanwhile, the application adopts a 'from coarse to fine' tracking mode, realizes the complementary advantages of vision and point cloud, and has high execution efficiency and reliability.

The application discloses an infrared image target tracking method based on a kernel correlation filter, which comprises the following steps: S1, extracting features of a target region to obtain a plurality of feature points and a descriptor of the target; S2, training and updating filter parameters according to the target features; S3, calculating a response map of a search region by using the parameters in step S2; S4, judging whether the tracking result is valid according to an APCE index of the response map in step S3, if yes, outputting the result and returning to step S1, if not, executing step S5; S5, matching the feature points of a current frame image with the target feature points, if the matching is successful, calculating the target position and size and outputting the same, and then returning to step S1, otherwise, reading a new frame image and executing step S5 again. The application can avoid parameter degradation when the target is seriously occluded or beyond the field of view, and re-track the target after the target appears.

Simulated audio signals in a virtual environment are decorrelated by obtaining (102) a first set of decorrelation filters (eg. frequency-selective phase shifting all-pass filters as in Fig. 3), evalua

The present application relates to the field of medical image processing, and particularly relates to an end-to-end ophthalmic surgery positioning navigation method and device. First, a coding and decoding network is used to extract eyeball features, and a multi-class segmentation layer constructed according to the eyeball features is used for semantic segmentation to determine the center position of the eyeball, so as to construct a polar coordinate sampler, so that the real-time collected intraoperative image is directly projected under the polar coordinate, and the eyeball feature map, the semantic segmentation result map and the preoperative image are converted to the polar coordinate system. Under the polar coordinate system, the current image and the target image are registered to complete eyeball rotation tracking. In the registration process, a twin network is combined with a correlation filter, the tracking target position, number and receptive field size are defined by defining a Gaussian template, the calculation efficiency is improved, the weighted sum of the inter-frame registration result and the long-time registration result is taken as the final rotation amount, the cumulative error is eliminated, and the stability and registration accuracy of the eyeball rotation continuous registration are improved.

The application discloses a deep spatio-temporal correlation video target tracking method and system, which realizes target tracking for each video sequence and guarantees to obtain accurate video target tracking results. Firstly, a spatio-temporal feature extractor is designed to extract spatio-temporal features of a template sequence and a search sequence. Secondly, a feature matching module is introduced, which is composed of a classification branch and a regression branch. The extracted template spatio-temporal features and the search spatio-temporal features are subjected to similarity matching through correlation filtering to obtain multi-channel correlation filtering features respectively. Then, a target tracking module is deployed, which includes a classification head and a regression head, and the classification score map and the regression score map are calculated according to the input multi-channel correlation filtering features respectively, which are used to predict the target position and estimate the target scale. Finally, the spatio-temporal correlation visual tracking model is optimized by minimizing the defined joint loss. In the target tracking test, a confidence region estimation strategy is proposed, so that the target tracking is kept robust and accurate in the video sequence.

The application provides a SAR deception jamming method of mixed domain efficient track deviation compensation, comprising the following steps: after a SAR radar enters a detection range of a jammer, relevant parameters of the SAR radar are acquired, and a jammer deception template is constructed according to false electromagnetic features generated in a SAR image by jamming; azimuth focusing and range migration compensation calculation are performed on the jammer deception template in an azimuth frequency domain, and then inverse Fourier transform in the azimuth direction is performed to obtain an initialization template; after the SAR radar enters a jamming area, real-time modulation is performed, track deviation compensation is performed on the initialization template after coupling term decomposition in the azimuth time domain, and a real-time SAR system correlation filter is constructed in the azimuth time domain pulse by pulse; the real-time SAR system correlation filter and the jammer correlation filter are multiplied in a range frequency domain to obtain a non-linear track jamming frequency domain response function, and convolution operation is performed on the non-linear track SAR deception jamming signal obtained after the SAR signal intercepted to obtain a non-linear track SAR deception jamming signal.

This invention discloses a robust tracking method for multiple moving targets using through-wall radar based on multi-domain data association and multi-state trajectory management. First, radar measurement results are obtained using a constant false alarm rate (CFAR) detection method. Second, the Euclidean distance between the measurements and the trajectories, the kernel correlation filter response, and the trajectory confidence cost are calculated to construct a minimum-cost maximum flow problem for data association. Subsequently, a 5-state trajectory management mechanism is established. Based on the target state and association results, new trajectories are generated, regular and unmatched trajectories are updated, and cross-trajectories are generated, updated, and separated. Specifically, in the cross-trajectory separation process, Bézier curves are used to fit the cross-trajectories, and the target state is estimated based on the Bézier curves to predict the target's position in the next frame. The re-matching of cross-trajectories and measurements is achieved based on the distance between the predicted position and the measurements.

The invention relates to a driver safe driving prompting method based on AR glasses, and the method comprises the steps: obtaining a forward video, collected by the AR glasses, of a first visual angle of a driver in a vehicle driving process; tracking a fixed marker in the forward video based on a target tracking algorithm of correlation filtering to form a dynamic anchor frame; according to the dynamic anchor frame, intercepting a front view subgraph in a video frame of the forward video; obtaining a vehicle lateral deviation state, a vehicle angle deviation state and a sight attention deviation state according to the front view subgraph, and determining a vehicle risk level; and outputting prompt information according to the vehicle risk level. While effective transmission of the prompt information and the navigation information is ensured, the interference to the driving task is reduced, so that the overall driving safety is improved.

The application discloses a target tracking method, system and device based on a multi-level loss recovery mechanism, comprising the following steps: collecting a video frame sequence and initial target region information; using a kernel correlation filter to perform target tracking on a current video frame, extracting HOG features of the target and generating a response map; using a Kalman filter to correct the motion state of the tracking result of the kernel correlation filter and outputting a corrected target position; evaluating the tracking confidence of the current target, and when the reliability evaluation result is lower than a preset threshold, starting a backup template mechanism for secondary verification; when a plurality of backup templates in succession fail, triggering a target loss redetection mechanism to redetect the target, obtaining a candidate region of the target, and performing multi-scale search and correction on the redetected candidate region to recover normal tracking of the target. The application can realize high-robustness and real-time embedded target tracking.

The present application belongs to the technical field of data processing, and discloses a kind of airborne terahertz video synthetic aperture radar data real-time processing method.The present application uses the image registration method based on kernel correlation filtering to estimate and correct the relative shift between image sequences;Secondly, the motion target shadow is detected using a deep neural network;Then, the detection results are used as the input of the improved SORT algorithm to track the motion target;Finally, the latitude and longitude coordinates of the motion target are obtained based on the pixel position and latitude and longitude coordinates of the auxiliary reference point.The present application fills the technical gap of real-time data processing methods that match the imaging rate of airborne terahertz video synthetic aperture radar in practical applications.

The application provides a video target tracking method and system based on feature fusion, and proposes a method combining a target detection algorithm and a correlation filter, a target detection frame of a target is recognized through the target detection algorithm, a target tracking edge frame is obtained through a local and global multi-feature fusion correlation filter, and the target detection frame and the target tracking edge frame are matched through a Hungarian matching algorithm, so that the problem of low accuracy and success rate of a target detection and tracking algorithm of a single correlation filter tracking algorithm or a neural network target detection algorithm is solved, target tracking stability in a complex scene such as a power scene is realized, and the real-time performance, accuracy and success rate of target detection and tracking are improved.

The application discloses a space-based distributed multi-sensor multi-target tracking and positioning method. The method first acquires a gray image of each sensor, extracts the coordinates of the target in the image, matches the serial number of the extracted target, and takes the first frame coordinates of the target and the image sequence as the tracking input. The tracking process is based on the discriminant correlation filter, adds the energy concentration threshold of small target features, constructs the double-constraint trajectory association model of the target distance and speed similarity between the continuous frames, and simultaneously uses the Kalman filtering algorithm to predict the target position. An adaptive covariance cross-fusion positioning algorithm is designed between multiple sensors, the geometric relationship between multiple detection satellites and the target is established, the sensor observation parameters are used as the basis for adjusting the fusion weight, and the three-dimensional position of the target is calculated. The method can significantly improve the target tracking accuracy and suppress the positioning deviation problem caused by the instability of part of the sensors, can adapt to the dynamic changes of the distributed detection scene, and ensures the target positioning accuracy.

The application provides a correlation filter tracking method based on saliency perception and space-time regularization, introduces a space regularization and a time constraint term into a target function, establishes a space regularization and time constraint model, converts the model into a frequency domain, decomposes the converted model into multiple sub-problems, solves the sub-problems respectively, obtains an optimized model, acquires a saliency map of a target region, fuses the saliency map of the target into initial space regularization weight coefficients to obtain new weight coefficients based on saliency perception, learns a correlation filter according to the optimized model, uses the correlation filter to locate the target and simultaneously completes scale estimation of the target, and finally updates model parameters to complete filter tracking, which is helpful to enhance time continuity and consistency of the model, effectively reduce boundary effects, effectively improve tracking performance and tracking efficiency, and enable the tracker to adapt to appearance changes and suppress background interference.

The invention discloses an unmanned aerial vehicle positioning method and system for urban rail transit vehicle inspection, and relates to the technical field of rail transit vehicle inspection. The method comprises the following steps: step 1, arranging bar code labels with unique coding characteristics at key parts of a bogie, a vehicle door and a pantograph of an urban rail transit vehicle; step 2, the unmanned aerial vehicle carries out self-inspection and preset inspection path parameter loading, and internal and external parameters of the depth camera and the scanning angle and resolution of the laser radar are calibrated; and step 3, based on initial pose information provided by the inertial measurement unit and the visual odometer, flying to a target correction point along a preset path. The position deviation is calculated in real time through a kernel correlation filtering algorithm, and accurate adjustment of the attitude of the unmanned aerial vehicle is realized by using multi-channel PID servo control, so that the accumulative error of the SLAM technology is effectively inhibited, and the centimeter-level positioning accuracy in long-distance inspection is guaranteed.

The application provides a fan operation state recognition method and system based on multi-modal optical fiber sensing monitoring. The method includes the following steps: acquiring continuous sensing signals collected by a distributed optical fiber sensor, performing time-space correlation filtering on the continuous sensing signals, generating a multi-physical field micro-change feature vector set, inputting the multi-physical field micro-change feature vector set into a pre-trained coupled response modeling network, performing correlation modeling on the strain field micro-change feature vector and the temperature field micro-change feature vector through the time-space cross attention mechanism in the network, and generating a dynamic response fingerprint spectrum; calling a fault evolution path set in a historical fault database, performing similarity matching on the dynamic response fingerprint spectrum, and generating an operation state deviation index; querying a pre-set health state evaluation strategy library, determining a fan overall health state grade and a corresponding potential fault risk probability, and generating an operation state evaluation result containing a fault area positioning coordinate. The application can improve the accuracy of fan operation state recognition.

The invention discloses an in-vehicle echo cancellation method based on a variable proportion decorrelation least mean square algorithm. The method comprises the following steps: sampling a vehicle signal; updating coefficients of the decorrelation filter; calculating partial decorrelation signals; self-adaptive filter coefficient updating; echo cancellation; calculating signal power; updating a proportional control factor; and repeating the steps until the dialogue is finished. The method has the advantages of being high in convergence speed and good in steady-state error.

In various examples, correlation filters may be learned for object tracking based at least on occlusion maps. When learning a correlation filter from an image area, an occlusion map may be applied to the image area that masks, excludes, and / or blurs occlusions of the target object. The correlation filter may be learned from a modified image, thereby reducing or eliminating learning from occlusions while still allowing for learning the target object from exposed portions. The occlusion maps may be generated using a machine learning model, such as a Gaussian Mixture Model (GMM) that is trained (e.g., using the image areas used to learn the correlation filter) using the target object as a background so that occlusions are detected as foreground.

An audio signal processing method and apparatus for adaptively adjusting a decorrelator. The method comprises obtaining a control parameter and calculating mean and variation of the control parameter. Ratio of the variation and mean of the control parameter is calculated, and a decorrelation parameter is calculated based on the said ratio. The decorrelation parameter is then provided to a decorrelator.

In various examples, image areas may be extracted from a batch of one or more images and may be scaled, in batch, to one or more template sizes. Where the image areas include search regions used for localization of objects, the scaled search regions may be loaded into Graphics Processing Unit (GPU) memory and processed in parallel for localization. Similarly, where image areas are used for filter updates, the scaled image areas may be loaded into GPU memory and processed in parallel for filter updates. The image areas may be batched from any number of images and / or from any number of single-and / or multi-object trackers. Further aspects of the disclosure provide approaches for associating locations using correlation response values, for learning correlation filters in object tracking based at least on focused windowing, and for learning correlation filters in object tracking based at least on occlusion maps.

The audio device includes a receiver 201 configured to receive a set of input audio signals. An audio beamformer 101 performs beamforming by combining the outputs of filters. A feedback circuit 103 includes a set of matching filters having a complex conjugate frequency response, and a beamform adapter 105 adapts the filters in accordance with a comparison between the input audio signal and the feedback audio signal. A fitting coefficient processor 207 determines decorrelation coefficients for a set of spatial decorrelation filters that produce a decorrelated output signal from the input audio signal, and the decorrelation coefficients are adapted based on the input signal. Sets of spatial filters 205, 301, and 401 are configured to apply spatial filtering to the input audio signal or the feedback signal, and the sets of spatial filters have coefficients determined from the decorrelation coefficients.

The invention discloses a kernel correlation filtering target tracking acceleration processing method and system of a programmable logic device. The method comprises the steps that the programmable logic device collects an input visible light image or infrared image; carrying out graying preprocessing on the collected visible light image, or carrying out image enhancement processing on the infrared image; calculating the gradient value of each pixel based on the preprocessed image, and obtaining the gradient intensity and the gradient direction; dividing the image into a plurality of cells with fixed sizes, and calculating a gradient direction histogram; combining a plurality of adjacent cells into blocks, carrying out normalization processing, and generating normalized feature sub-vectors; splicing all the normalized feature sub-vectors to form a final HOG feature vector; and the embedded processor executes kernel correlation filtering single target tracking based on the received HOG feature vector and the current tracking frame ROI information, and outputs target position and state information in real time. The delay can be obviously reduced, and the real-time performance is improved.