64 results about "High-motion" patented technology

The invention discloses a motion capture method based on sports equipment training, and relates to the technical field of intelligent motion capture, and the method comprises the steps: extracting a skeleton key point set of a detected object from a training ground image data stream through employing a human body key point detection algorithm, and extracting a two-dimensional contour region of sports equipment through employing a target segmentation algorithm, meanwhile, recording time index information of the corresponding image frames, performing synchronous comparison by utilizing timestamps of the training event data streams, establishing a time sequence mapping relation, and obtaining time sequence action data; taking the time sequence action data as a space anchor point, carrying out middle frame reconstruction on the high-speed motion stage based on a time sequence mapping relation, recovering an equipment motion track, and carrying out joint matching to generate a time sequence fusion sequence; establishing a multi-view geometric constraint through a time sequence fusion sequence and camera pose parameter information, and recovering a human body three-dimensional skeleton point cloud and an initial six-dimensional pose of the equipment; the space-time consistency and reconstruction precision of motion capture in a high-speed dynamic training scene are effectively improved.

The invention relates to the technical field of image communication, in particular to a video slow-action frame insertion playback system based on a generative adversarial network, which comprises a video signal acquisition module, a signal processing module, a frame rate conversion module and a playback code output module. The acquisition module outputs an original video frame sequence through an annular buffer; the processing module separates the denoised detail components and the edge gradient features in parallel; the frame rate conversion module uses a geometric structure as a boundary locking condition to guide detail components to execute nonlinear motion compensation and intermediate frame reconstruction; and the playback module executes time base remapping and distributed coding transmission. According to the invention, through a depth generation architecture of structural constraint textures, in combination with space-time consistency verification and a nearest neighbor pixel backfilling mechanism and a virtual time base rate decoupling technology, real-time slow-action video redisk with high signal-to-noise ratio and no artifacts for a high-speed moving target is realized.

The invention relates to the technical field of computer vision and image processing, in particular to a foreground detection method fusing wavelet transform and a prompt mechanism, which is characterized in that in a dynamic background, high-speed motion and small target scene, F-measure is averagely improved by more than or equal to 0.30, and the detection precision is obviously superior to that of FFNet and ASCNet. The wavelet transform attention module suppresses background noise, highlights target edges and textures, and improves input feature purity. The CFF, MFR and PB of the decoder collaboratively fuse multi-scale features, redundant information is reduced, and target details and contour integrity are reserved. According to integral end-to-end training, single-frame reasoning only needs 28 ms, and real-time performance and accuracy are both considered. Therefore, the problems that in an existing neural network foreground target detection algorithm, input data are often influenced by noise and background interference, redundant information is easily introduced into a decoder in the multi-level feature fusion and transmission process, and consequently the detection result is inaccurate are solved.

The invention discloses an event camera image reconstruction method of a multi-frame fusion network based on optical flow guidance, and belongs to the technical field of computer vision and image processing. The method comprises the following steps of: 1, acquiring event data, and converting the event data into continuous and smooth space-time voxels by adopting a full-time interval voxel coding scheme based on Gaussian distribution; step 2, constructing an FMF-Net network guided by an optical flow; step 3, designing a comprehensive loss function used for training the FMF-Net network; 4, performing supervised training on the FMF-Net network by using the public event data set; and step 5, inputting space-time voxels by using the trained FMF-Net network, and outputting a high-fidelity reconstructed image. According to the method, long-time motion clues and time domain dynamic features can be fully utilized, high-fidelity image reconstruction is realized under high-speed motion and sparse event input, the problems of loss of reconstruction details and poor consistency of an existing method in a dynamic scene are solved, and the structural similarity and visual quality of a reconstructed image are remarkably improved.

The invention provides a camera-based AI visual recognition and prediction control method and related equipment, and the method comprises the steps: obtaining continuous image frames of a moving object, and recognizing the skeleton key point coordinates and appearance feature information of the moving object; inputting the skeleton key point coordinates and the appearance feature information into a preset time sequence neural network model to obtain a motion prediction track; performing space mapping on the centroid coordinates of the moving object according to the motion prediction trajectory, and determining focal length control parameters of the camera; meanwhile, angle mapping is carried out on the position offset of the moving object, and pan-tilt control parameters are determined; and tracking and shooting the moving object according to the focal length control parameter and the holder control parameter. Through the mode, the camera can complete the pre-adjustment of the focusing mechanism and the composition direction before the target motion trend occurs, thereby effectively reducing the delay error in the focal length switching process, and improving the clear imaging rate and composition stability of a high-speed motion main body.

1. Name of the product in this design: Camera (High-speed camera - SH3). 2. Purpose of this design: This design is used for photographing high-speed moving targets. 3. The key design features of this product are its shape. 4. The image or photograph that best illustrates the design's key points: 3D view 1.

This invention relates to the field of 3D geometric modeling engine technology, specifically to a method for rapidly displaying Points of Interest (POI) information from a 3D real-world model. The invention first monitors the motion state of a virtual camera, reusing historical states during high-speed movement to avoid image jumps caused by aliasing. Second, it calculates the original occlusion distance of the POI and quantifies the physical depth difference between it and the scene. Then, by analyzing the local window depth distribution, it obtains discrete suppression weights to correct the original distance, adaptively distinguishing between solid and discrete structures, suppressing vegetation noise while maintaining high sensitivity to building edges. Finally, it constructs a state update cost based on the corrected distance and introduces hysteresis decision logic, setting a potential energy threshold for state switching, effectively reducing flickering in critical areas. This significantly improves the accuracy and stability of POI information display.

A virtual stereovision imaging array system and a moving target three-dimensional reconstruction method, the system comprises multiple cameras and multiple galvanometer mirrors, the galvanometer mirrors and the cameras are matched one by one, and a real-time controllable virtual camera array is formed; through synchronous control of the galvanometer mirrors, each virtual camera simultaneously tracks and locks the moving target; based on the real-time pose relationship between the virtual cameras and combined with the synchronously acquired image information, real-time three-dimensional reconstruction of the moving target is completed. The reconstruction method comprises joint calibration of the matched galvanometer mirrors and the cameras, determination of camera internal parameters and virtual camera external parameters; through real-time synchronous adjustment of the galvanometer mirrors, each virtual camera cooperatively tracks the moving target, acquires the moving target image, and updates the virtual camera external parameters in real time; in each frame of image, stereo matching is performed, and combined with the real-time updated virtual camera external parameters, the three-dimensional point cloud of the moving target is solved and obtained. The present application can complete high-efficiency and high-precision three-dimensional reconstruction of a high-speed moving target in a large field of view.

The application discloses a high-speed motion three-dimensional point tracking and trajectory reconstruction method and system, and relates to the technical field of three-dimensional point tracking. The method comprises the following steps: synchronously collecting an RGB image sequence and an event stream through a binocular multi-modal observation device; performing time alignment on the RGB image sequence and the event stream, extracting an RGB feature map based on the RGB image sequence, and extracting an event feature map based on the event stream; mapping the RGB feature map and the event feature map to a three-dimensional space; supplementing a three-dimensional feature point cloud through a linear interpolation algorithm; balancing the spatial distribution of the supplemented three-dimensional feature point cloud through an anchor point sampling algorithm to obtain a fused three-dimensional feature point cloud; and generating high-speed three-dimensional point trajectory data based on the fused three-dimensional feature point cloud through a time correlation transformer. The application effectively fuses the texture fidelity of the RGB image and the high time resolution of the event data, and can realize high-frame-rate and high-precision three-dimensional point trajectory reconstruction in a high-speed motion scene.

The application discloses a motion training evaluation system and method based on cooperation of an event camera and a visual camera, and is applied to the field of computer vision, and aims at the problems that an existing training evaluation system based on an optical camera cannot capture fast motion trajectories and limb shapes, cannot effectively evaluate action quality, and cannot provide effective training guidance; the application provides continuous human motion events in the form of event streams, can achieve millisecond-level motion response, is not affected by motion blur effects of high-speed moving objects, and can provide a higher dynamic range, and can provide more effective motion training evaluation in scenes with strong light, backlight, and sharp changes in brightness. Through cooperation of the visual camera and the event camera, not only the accuracy of fast actions can be evaluated, but also standard action teaching and training in the form of visual scene interaction can be performed.

This invention discloses a precision detection method and system for surface defects in metal strips, relating to the interdisciplinary fields of mechanical engineering and machine vision. The method includes: synchronously acquiring multi-band images using a multispectral light source and an industrial camera array; establishing a spatiotemporal synchronization model based on encoder speed signals and performing sub-pixel-level image registration to correct motion distortion; adaptively fusing multispectral images based on a material-spectral response database to enhance the signal-to-noise ratio of micro-defects; employing a lightweight deep learning model to achieve pixel-level defect segmentation and classification; and extracting defect geometric parameters and automatically classifying them using a process tolerance library. This invention aims to solve the problems of narrow material applicability, severe image distortion under high-speed motion, low micro-defect detection rate, and complex system deployment in existing technologies. It achieves high-precision, high-speed online detection of non-ferromagnetic metal strips, significantly improving the sensitivity to micron-level defects, system robustness, and ease of deployment.

The invention discloses a projection screen image quality optimization processing method and terminal based on an intelligent agent identification video source, and relates to the technical field of video processing, and the method comprises the steps: selecting a projected screen device through a voice projection screen instruction to complete projection screen connection; the voice control instruction is converted into a control signal for current projection screen connection, and the control signal is sent to the projection screen end through a reverse channel, so that the projection screen content of the projection screen end is directly controlled, or the projection screen equipment is directly controlled according to the voice control instruction; identifying the signal source type of the current input video source; analyzing the inherent frame rate and the resolution of the input video source, and when the screen projection device detects that the video source is a predetermined low-frame-rate content or a predetermined high-speed motion content, determining that the video source is a high frame compensation income scene; and a motion estimation and motion compensation frame compensation function is automatically started to insert a calculation frame between original frames for frame compensation. According to the invention, the video content can be automatically identified, optimization functions such as frame compensation can be intelligently started, and meanwhile, the screen projection experience of voice interaction can be realized in the whole process, so that the use convenience and immersion of a user are greatly improved.

The embodiment of the invention provides a time sequence controllable non-contact deformation measuring method and device, computer equipment and a medium, and relates to the technical field of optical measurement experiment mechanics, and the method comprises the following steps: determining time sequence control parameters, internal and external parameters and a coordinate system transformation matrix of an industrial camera; collecting a sequence diagram of the surface of the measured object based on the trigger signal, triggering the light source to provide illumination to generate a first data stream, and controlling the event camera to output a second data stream of pixel brightness change of the surface of the measured object; analyzing the second data stream, and dynamically controlling image acquisition of the industrial camera according to the instantaneous motion state parameters; and carrying out digital image correlation matching and three-dimensional reconstruction to obtain a high spatial resolution deformation field of the measured object at the sampling moment, and carrying out event-driven three-dimensional reconstruction to generate a three-dimensional deformation measurement result. According to the scheme, the contradiction between clear imaging and long-time continuous acquisition of the high-speed moving object is solved through sequential control parameters combining short exposure and long interval.

The invention belongs to the technical field of industrial camera anti-blurring, and discloses an industrial camera system for resisting high-speed motion blurring, which forms a complete closed loop from physical compensation to software repair through coupling of hardware active anti-shake and software deconvolution, and can effectively solve the problem of mixed blurring in an industrial scene. The hardware end is driven by piezoelectric ceramics and flexible damping in a combined manner, and is matched with real-time data of a six-axis IMU, so that low-frequency uniform-speed vibration and high-frequency instantaneous impact can be quickly counteracted; a software end deduces a fuzzy kernel based on an IMU track, residual fuzzy is restored through Fourier domain deconvolution, and fuzzy elimination is more thorough than that of a traditional single hardware or software scheme; when the device is actually used for metal part detection, tiny cracks which cannot be seen clearly due to fuzziness can be clearly presented now, so that the reliability of defect identification is improved, and the 0.1 mm level detection precision requirement in industrial production is met; and the imaging frame rate is stabilized to be more than 100fps while the precision is ensured, and the method can cope with scenes with high real-time requirements.

This invention discloses an anti-blurring dual-stream surface defect detection method and apparatus for industrial vision, relating to the fields of image processing and computer vision. The method acquires surface images of metal parts under high-speed motion environments; inputs a pre-constructed anti-blurring dual-stream feature enhancement network, which includes an adversarial detail recovery subnetwork and an inter-layer feature reshaping subnetwork set in parallel with a shared backbone; the backbone network includes a multi-scale guidance module, fusing residual dense connections, spatial-channel attention, and multi-scale convolution to enhance feature representation; outputs high-frequency detail images through the adversarial detail recovery subnetwork and pixel-level defect segmentation images through the inter-layer feature reshaping subnetwork; locates defect regions based on the segmented images and extracts image patches from the high-frequency images; finally, inputs the defect recognition subnetwork to output classification results; this invention improves the accuracy and robustness of defect detection in high-speed environments by collaboratively optimizing detail reconstruction and segmentation through a dual-stream network.

This invention discloses a motion blur compensation image enhancement method for high-speed cameras, relating to the field of computer vision technology. The method includes acquiring raw sequence data recorded by a high-speed imaging device, analyzing overlapping regions of adjacent frames in the sequence using a temporal convolutional network to determine the temporal continuity features corresponding to highly overlapping frames, obtaining a separated frame sequence, and using an optical flow estimation algorithm to track the object's movement trajectory based on the separated frame sequence. If the optical flow vector exceeds a preset threshold, it is judged as an interference factor in a dynamic scene, resulting in a motion-compensated frame sequence. This motion blur compensation image enhancement method for high-speed cameras not only improves the accuracy of image enhancement but also enables real-time and accurate extraction of key information frames in complex dynamic environments, optimizing the defect detection process. It has broad application prospects, particularly in high-speed motion analysis and industrial inspection, significantly improving detection efficiency and accuracy.

The application provides a stage light tracking control method and system based on target positioning, the method comprises collecting a video stream and obtaining a candidate box through target detection, and determining a main feature area of a human body through pose estimation; the input length of three-dimensional convolution is adjusted based on a time sequence light fluctuation index, global features are extracted, and attention weights are assigned to local texture features of the main feature area; a motion pattern signature is constructed using sparse optical flow of skeletal key points, a target re-identification vector is generated by fusing the global features and weighted local features, and a historical feature library is matched; the fusion weight of position prediction and feature measurement in a cost matrix is adjusted according to the amplitude of the motion pattern signature, and target matching is completed; a virtual aiming point is calculated based on the key point confidence of the matched target, a position compensation amount is generated by combining a light mapping model and sparse optical flow, and a control instruction is output to drive the light to continuously track stage characters, thereby improving the tracking stability and alignment accuracy under complex lighting and high-speed motion scenes.

An event camera image reconstruction method based on an optical flow-guided multi-frame fusion network belongs to the field of computer vision and image processing technology. Step 1: Acquire event data and convert it into continuous, smooth spatiotemporal voxels using a Gaussian distribution-based full-time-interval voxel encoding scheme. Step 2: Construct an optical flow-guided FMF-Net network. Step 3: Design a comprehensive loss function for training the FMF-Net network. Step 4: Supervised training of the FMF-Net network using a publicly available event dataset. Step 5: Utilize the trained FMF-Net network, input spatiotemporal voxels, and output a high-fidelity reconstructed image. This invention fully utilizes long-term motion cues and temporal dynamic features to achieve high-fidelity image reconstruction under high-speed motion and sparse event input, solving the problems of detail loss and poor consistency in reconstruction under dynamic scenes in existing methods, and significantly improving the structural similarity and visual quality of the reconstructed image.

The application provides a system and method for online evaluation of the opening quality of tobacco tow, which comprises a high-speed image acquisition system, a high-speed area array camera, a high-brightness area array light source and a light source controller, and high-quality images are acquired through the cooperative control of the camera and the light source; the control system is responsible for calculating the tow width and uniformity of the acquired tow images, so as to online evaluate the opening quality of the tow. The application can acquire clear images of the tobacco tow under high-speed motion through the selection and cooperative control of the area array camera and the area array light source, avoids the evaluation error caused by the blurred images under high-speed motion, and simultaneously adopts a simpler width evaluation and uniformity evaluation method, thereby ensuring the calculation efficiency. The measurement system and method introduced in the application have higher automation degree, measurement efficiency and measurement precision, thereby improving the opening quality of the tow and improving the control level of the production process.

The invention discloses a monocular vision three-dimensional measurement method and system for a high-speed moving target, and the method comprises the steps: scanning a linear reference object located in a target motion plane through a laser radar, obtaining point cloud data, and carrying out the fitting of the point cloud data, and determining a first straight line of the central axis of the reference object; acquiring a target motion sequence image by adopting a camera, and extracting a second straight line passing through the optical center of the camera and the target based on the motion sequence image; and based on the position relationship between the first straight line and the second straight line in the same coordinate system, obtaining the three-dimensional space coordinates of the target. According to the method, the high precision of vision measurement and the reliability of laser radar ranging are fully utilized, the instability of a pure vision method during feature missing or motion blur is overcome, and the measurement robustness and accuracy of the system under real and complex working conditions are improved by introducing a deviation degree judgment mechanism.

The invention relates to a motion blurred image reconstruction method in a low-illumination environment, and the method comprises the steps: S1, collecting an original image through an RGB camera, and obtaining a motion blurred image under the conditions of long exposure, high-speed motion or camera shaking; s2, constructing an event-frame fusion algorithm based on an event camera principle, and jointly utilizing high time domain motion information of an event stream and the pixel intensity of a frame image; s3, a joint strategy of edge alignment constraint and fuzzy alignment constraint is provided, the edge alignment constraint guarantees the authenticity of the geometric contour, and the fuzzy alignment constraint is kept consistent with the physical imaging process; s4, designing a lightweight double-branch feature extraction network (a local branch, a global branch and a fusion algorithm), and capturing fine-grained features and a global structure of the sample at the same time; and S5, completing reconstruction and outputting a clear image. According to the method, the quality of blurred image reconstruction is effectively improved under the scenes of low illumination, high-speed target motion and the like, a large amount of noise and artifacts are inhibited, and the method has good robustness and generalization.

The present application relates to the technical field of camera simulation, and particularly relates to an event camera physical characteristic simulation method and device, wherein the method comprises the following steps: a description file of a lens group of a target event camera is established to select a plurality of sampling interval points; an exit pupil condition of each sampling interval point is calculated, and the exit pupil condition on the entire straight line is generated according to the exit pupil condition of each sampling interval point; whether a light beam of a target scene is in the exit pupil condition on the entire straight line is judged based on a three-dimensional scene model obtained by modeling the target scene and the target event camera, if the light beam is in the exit pupil condition on the entire straight line, the light beam is converted into a photoelectric current, otherwise the light beam is ignored; and the photoelectric current is simulated by using a sensor of the target event camera to obtain a three-dimensional simulation result. Thus, a physical-based full-link event simulator is obtained, which can provide data for various event camera vision tasks, and solves the problem that a traditional camera is difficult to detect with high precision in high-speed motion, excessive illumination and low-illumination scenes.

The invention relates to the technical field of computer vision, and discloses a moving target three-dimensional reconstruction method and system, electronic equipment and a storage medium, and the method comprises the steps: obtaining a binocular image pair after binocular correction; monocular depth priori and binocular features are extracted based on the corrected binocular image pair; constructing a cost body based on binocular features, and generating initial parallax estimation through three-dimensional volume accumulation; fusing and refining the initial disparity estimation based on monocular depth prior, optimizing the disparity estimation through a multi-stage circulation unit, and outputting a refined disparity map; depth information is calculated based on the refined disparity map, and a three-dimensional point cloud is generated according to the depth information; and performing post-processing on the three-dimensional point cloud, and suppressing noise and outliers to obtain a three-dimensional reconstruction result. According to the invention, high-precision and complete three-dimensional reconstruction of a high-speed moving target can be realized under the condition of a large field of view.

The invention discloses a steel rail surface defect three-dimensional reconstruction device based on double line-scan digital cameras and a detection method. Comprising a light source and motion control device, a computer, a mounting frame, a first linear array camera, a second linear array camera, a first R-axis electric rotary sliding table, a second R-axis electric rotary sliding table, an X-axis electric linear sliding table, a first X-axis displacement base and a second X-axis displacement base. The problem of dark current generated by an area-array camera in a high-speed motion scene is effectively solved, the problem of image splicing when the area-array camera shoots a large-view-field scene is avoided, high-quality images in a large view field are obtained, a high-quality image set is provided for subsequent algorithm detection and binocular three-dimensional reconstruction of steel rail surface defects, and the method is suitable for large-view-field and large-view-field detection. The depth resolution and the detection rate of tiny defects are improved, the automation degree is high, and the system has universality and can adapt to different detection scenes.

The application provides a two-dimensional code level evaluation method, device, medium and product. The method comprises the following steps: obtaining an original digital image of a two-dimensional code to be evaluated; performing static feature analysis to generate a static structure reference score; simultaneously, obtaining a dynamic disturbance image set through a digital disturbance transformation matrix, and traversing the image set to test the decoding success degree to obtain a decoding success metric value; performing weighted processing on the decoding success metric value based on the static structure reference score to obtain a dynamic robustness score; and calculating a final evaluation value through a preset weighted fusion function based on the static structure reference score and the dynamic robustness score. The scheme solves the technical problem of inaccurate evaluation caused by only relying on static analysis by combining double evaluation of static quality and dynamic robustness, can more accurately predict the real decoding performance of the two-dimensional code in a complex dynamic environment such as industrial high-speed motion, and thus effectively avoids the risk of production interruption or data traceability failure.

This invention provides a time-controlled, non-contact deformation measurement method, apparatus, computer equipment, and medium, relating to the field of optical experimental mechanics. The method includes the following steps: determining the time-series control parameters, intrinsic and extrinsic parameters, and coordinate transformation matrix of an industrial camera; acquiring a sequence of images of the surface of the object under test based on a trigger signal and triggering a light source to provide illumination to generate a first data stream; controlling an event camera to output a second data stream showing the pixel brightness changes on the surface of the object under test; parsing the second data stream and dynamically controlling the image acquisition of the industrial camera according to instantaneous motion state parameters; performing digital image correlation matching and 3D reconstruction to obtain a high spatial resolution deformation field of the object under test at the sampling time; performing event-driven 3D reconstruction to generate a 3D deformation measurement result. This scheme resolves the contradiction between clear imaging of high-speed moving objects and long-term continuous acquisition by combining short exposure and long interval time-series control parameters.

The invention provides a video frame rate adjusting method and system based on an LED display device, a medium, a program and a terminal, different frame compensation strategies are selected according to a second motion intensity value, and when low motion is needed and only one frame needs to be inserted, frame repetition, zero motion estimation and zero additional delay are adopted to ensure that the rhythm is natural and the real-time performance is met; and switching to motion compensation frame insertion when the system is in medium-high motion or needs multi-frame insertion, reconstructing an intermediate state according to displacement, continuously restoring a motion track, and inhibiting jitter and trailing. Moreover, when the motion compensation frame insertion strategy is executed, afterglow compensation is carried out on the insertion frame, so that the phenomena of smear, ghosting, local brightness overexposure and the like caused by the inherent afterglow effect of LED pixel points in the LED display device when a high-speed motion picture is played are avoided.

The application provides a film surface defect detection method and system, relates to the technical field of defect detection, and fundamentally improves detection performance by introducing a defect confidence evaluation mechanism based on morphology and energy distribution and constructing an intelligent secondary verification link. The mechanism can accurately distinguish real physical defects with regular morphology and concentrated energy from pseudo defects caused by electromagnetic interference, film instantaneous wrinkles and the like, effectively solves the high false alarm rate problem caused by the dependence of the prior art on single signal strength while retaining high detection rate for low contrast defects. In addition, the judgment model based on physical characteristics has natural robustness to background noise generated in high-speed motion, and the adjustable confidence threshold gives the system high practical flexibility, enabling it to adapt to complex industrial environments and different quality control standards, significantly enhancing the accuracy, reliability and intelligence level of the entire detection system.

The invention discloses a mobile instance segmentation method and system, and belongs to the field of pattern recognition. In order to solve the problem of insufficient small target segmentation precision caused by image degradation, event appearance and motion feature entanglement, event sparsity and modal division solidification in extreme scenes of low illumination, high-speed motion and the like in the existing method, the invention provides an innovative multi-modal fusion framework. According to the framework, firstly, a double-decoupling encoder is adopted, and appearance and motion double features are extracted from an image and an event stream at the same time; cross-modal feature alignment and fusion are realized by using adversarial training and reversible conversion; promoting feature interaction in combination with cross attention; and finally, outputting an instance mask and a motion state through task decoding. The method is suitable for the fields of automatic driving, intelligent monitoring and the like, and has the advantages of high small target segmentation precision, high fusion robustness, excellent extreme scene performance and the like.