42 results about "High frame rate" patented technology

This invention provides a lossless compression and decoding method for depth images. Lossless compression includes: dividing sequential data into multiple independent image blocks according to predetermined partitioning rules; selecting a corresponding thread model based on the characteristics of the depth image; calling a thread based on the thread model and using a reversible encoding algorithm to independently encode each image block; wherein, at the start of encoding for each image block, the encoder state is reset so that the encoding of each image block does not depend on the data of the preceding image block; generating metadata containing position index and compressed data length information and assembling it with the encoded image block for output. Lossless decoding includes: reading the metadata and decoding at least one image block based on the metadata. Through the above methods, parallel encoding and decoding are achieved; precise positioning via physical offsets and random access and on-demand region decoding are supported, improving the processing response of high frame rate depth streams under low computing power environments while ensuring lossless restoration.

The application discloses a kind of non-contact physiological signal monitoring systems based on image recognition, comprising: evoked collection module, for output light stimulation sequence and gaze guide sequence, and the high frame rate video image of the body area of target object is collected;Unit mapping module is used to construct multiple physiological response units, generate physiological response unit topology chart;Demodulation extraction module is used to perform encoding demodulation processing, generate unit evoked response feature set;Transmitting configuration module is used to construct physiological response transmission atlas;Separation correction module is used to separate and inhibit unit evoked response feature set;Inversion analysis module is used to carry out inversion, generate physiological inversion parameter set;Calibration compensation module is used to construct individual response calibration model and carry out compensation correction;Abnormal output module is used to generate abnormal monitoring result.The application multi-region image recognition, realize non-contact physiological signal monitoring.

The application discloses a kind of facial expression muscle dynamic coordination evaluation system based on deep learning, including data acquisition module, for collecting the multi-object high frame rate facial dynamic video of preoperative baseline period of subject and the multi-object high frame rate facial dynamic video of postoperative review period;Data preprocessing module is used to generate preoperative three-dimensional facial coordinate dataset and postoperative three-dimensional facial coordinate dataset;Vector field extraction module, extract preoperative full-face expression muscle three-dimensional displacement vector field and postoperative full-face expression muscle three-dimensional displacement vector field;Feature extraction module, extract the physiological asymmetry reference feature set of bilateral corresponding muscle area before operation, and the motion feature set of bilateral corresponding muscle area after operation;Model inference module, output eliminates the quantitative value of diagnosis and treatment related muscle movement symmetry deviation and dynamic coordination degree change quantitative value after innate physiological asymmetry;Result generation module generates the expression muscle dynamic coordination change evaluation result of corresponding subject before and after oral diagnosis and treatment.

Provided are a solid-state imaging device, a driving method for the solid-state imaging device, and an electronic apparatus, which have high GS function, AF function, and can fully achieve SEHDR performance and substantially achieve high dynamic range and high frame rate. A conversion signal reading unit (211) performs, in accordance with driving control of a reading unit (70), in an imaging mode, reading processing of a pixel signal corresponding to a combined charge of accumulated charges of a first photodiode PD11 and a second photodiode PD12, in accordance with a first conversion gain HCG, and then in accordance with a second conversion gain LCG, while maintaining a first transfer transistor TG11-Tr and a second transfer transistor TG12-Tr in an on state during the same transfer period.

This invention discloses an infrared high-speed target detection and tracking method and system based on alternating full-frame inspection and multi-window burst readout. The method includes: first, controlling the detector to operate in full-frame readout mode, establishing an inter-frame correlation analysis mechanism, and verifying the validity of suspected targets based on kinematic constraints; after confirming a valid target, performing dynamic window planning including state prediction, bias construction, and topology reconstruction based on the detector's readout circuit delay and the target's motion state; then switching to multi-window readout mode to output burst tracking sequences at a high frame rate, and real-time correcting the prediction model to achieve window following; finally, when a preset number of frames is completed or the target is lost, forcibly switching back to full-frame readout mode to eliminate accumulated errors and rescanning. This invention effectively solves the problem that large-area array detectors cannot simultaneously handle wide-area surveillance and accurate high-speed target tracking through the alternating mechanism of low-frequency inspection and high-frequency burst readout, significantly improving positioning and velocity measurement accuracy.

The application provides a multi-path data processing method, a real display device, equipment, a medium and a product. Pre-set row data output by at least two image signal processors is acquired, and the pre-set row data output by the at least two image signal processors is processed in parallel through network reasoning. The demand for image segmentation processing of the whole image is cancelled to improve the accuracy of image processing, and the requirement for the overall computing power and data cache space of the system is reduced. In addition, the pre-set row data output by at least part of the image signal is divided into multiple pieces, and network reasoning is performed on each piece of data according to a pre-set strategy. The pre-set strategy is that the mobile terminal starts from a first end, moves to a second opposite end, returns to the first end after reaching the second end, and continues until the piece data reasoning is completed. In this way, the memory overhead of the neural network processing unit is reduced, the ability of a single neural network processing unit to process multiple paths of image data in parallel is improved, high frame rate and high resolution are supported, and network reasoning delay is reduced.

This invention relates to a time-delay integration direct time-of-flight (DToF) LiDAR sensor and its processing method. Based on an area-array scanning operation, it completes quantization and readout within each short transit time, thus increasing the frame rate. The sensor's readout circuit employs a successive TDC (Time-of-Flight Difference) approach to approximate the histogram peak value, significantly reducing the amount of data generated by quantization. This reduces both the area overhead of the internal storage circuitry and the amount of data read out. Simultaneously, the system uses a coarse-quantization followed by a fine-quantization method, improving the quantization accuracy of the histogram TDC at a lower clock frequency. Benefiting from coarse quantization generating only one sign bit per stage, the TDI-DToF-LiDAR sensor exhibits a smaller data volume, higher frame rate, and smaller storage circuitry area overhead compared to traditional DToF-LiDAR sensors. Furthermore, the coarse-quantization + fine-quantization approach allows for high quantization accuracy without requiring a high-frequency clock input.

This invention discloses a method, system, and medium for detecting and warning of dynamic frames in silent video, relating to the field of data processing technology. The method includes: acquiring silent video of a power grid monitoring area in a low frame rate mode; extracting continuous frames and identifying dynamically changing frames using a lightweight model; upon detecting a dynamically changing frame, switching to a high frame rate mode to acquire high frame rate video; calling a large model for dynamic change identification and detailed feature extraction to determine power grid risk targets and events; combining the located targets for abnormal risk analysis; triggering an alarm mechanism; and generating warning information. This invention solves the technical problems of existing power grid monitoring technologies being unable to efficiently detect dynamic changes in low frame rate mode and exhibiting low monitoring accuracy and reliability in complex environments. It achieves the technical effect of improving the reliability of power grid monitoring by intelligently switching between high and low frame rate modes, enabling accurate identification and rapid response to risk events.

This invention discloses a message processing method and related apparatus, relating to the field of message processing. When the current message cache is not full, the main thread receives live messages and submits them to a HandlerThread for asynchronous parsing to obtain different message types. Corresponding caching strategies are then applied, retrieving image cache data from the ImageSpan cache and text cache data from the plain text cache, assembling them into a complete live message, and inserting it into a RecycleView queue for message aggregation caching. The complete live message is then inserted into a scrolling control in the RecycleView list for display. Based on a preset scrolling strategy, the complete live message is directly scrolled and displayed using a SpannableString cache. This invention utilizes HandlerThread for asynchronous message processing, a three-layer caching mechanism to optimize performance, and strategy parsing to decouple rendering logic, ensuring the main thread maintains a high frame rate.

This invention relates to a high frame rate target tracking method based on a deep spiking neural network. The method involves segmenting target events on an embedded platform within a given time period into an n-bin voxel grid, generating positive and negative event frames. These frames are then input into a YOLO neural network for spatial feature extraction, outputting the position and category information of all targets in each frame, forming spatial features. The positive and negative event frames are then input into a spiking convolutional neural network in chronological order for temporal feature extraction, outputting temporal information related to the tracked target. A target tracking network is constructed, and after spatiotemporal feature fusion of the spatial and temporal information, regression and classification calculations are performed to output the position and category of the tracked target at different times. This invention achieves fast and efficient target tracking on embedded, low-power platforms, effectively solving the problems of high energy consumption and low efficiency in high frame rate target recognition networks under complex environments.

This invention provides an image background noise suppression method based on anisotropic filtering with local contrast, belonging to the field of image processing technology. The method includes: Step S1, calculating the average gray value of pixels in a central neighborhood within a first specific range centered on the current pixel, obtaining the average gray value of the central neighborhood; Step S2, calculating the average gray value of background pixels in eight surrounding background neighborhoods within a second specific range centered on the current pixel, obtaining the average gray value of each of the eight background neighborhoods; Step S3, calculating the differences between the average gray value of the eight background neighborhoods and the average gray value of the central neighborhood in four directions, based on the average gray value of the central neighborhood and the average gray value of the eight background neighborhoods; Step S4, determining the gray value of the current pixel based on the differences in the four directions. This invention provides stronger local signal-to-noise ratio gain, clearer target boundaries, lower average background brightness, and faster calculation speed, enabling it to respond to high frame rate processing requirements.

This application provides a screen refresh rate adjustment method and an electronic device. In this method, the electronic device displays a target interface, detects user actions on the target interface, and obtains operation information, including the type of operation. The electronic device performs content recognition on the target interface to obtain a content recognition result, which indicates that the target interface includes at least one of video, animated images, or dynamic components. The electronic device adjusts the screen refresh rate based on the operation information and the content recognition result, and refreshes the target interface according to the adjusted screen refresh rate. This solution allows the electronic device to dynamically adjust its screen refresh rate based on user actions and the content displayed on the interface, without continuously maintaining a high frame rate, thus reducing power consumption while ensuring display quality.

An image generation method using a scanning electron microscope includes generating a first image by imaging a wafer sample at a low frame rate, generating a sample image by imaging a reference sample at a high frame rate, and generating a second image having a high signal-to-noise-ratio using phase information extracted from the first image and amplitude information extracted from the sample image, wherein the number of patterns included in the first image and the number of patterns included in the sample image are the same.

This invention relates to a method for achieving high frame rate and ultra-large dynamic range in a single scenario. The method involves determining the radiation characteristics of the target based on the detection task and obtaining the spectral range; determining the detector type based on the obtained spectral range and detection requirements; and determining the integration time increments based on the target's radiation characteristics, the dynamic range requirements of the detection scenario, and the capabilities of the detector devices, and calculating the integration time for each increment. T int_N The invention involves calculating the absolute radiometric calibration coefficients for each time level; performing multi-time integration imaging of the detected scene using a remote sensor; and real-time blind pixel replacement and radiometric correction processing on the integrated time imaging results for each time level using a video processing device. A saturation threshold is set, and the relationship between the image DN value and the saturation threshold is determined pixel-by-pixel. The DN value assignment operation is then performed iteratively to generate a fused image without saturation and rich in detail. This invention achieves ultra-large dynamic range target detection without increasing image data capacity.

A method for generating an image, a global shutter image sensor and an electronic device, the method is applied to a global shutter GS image sensor, the GS image sensor includes a first row of pixel circuits and a last row of pixel circuits, the method includes: controlling the first row of pixel circuits and the last row of pixel circuits to expose at the same time, generating a first electrical signal and a second electrical signal, wherein the first electrical signal is an electrical signal generated by the first row of pixel circuits, and the second electrical signal is an electrical signal generated by the last row of pixel circuits; after the first electrical signal is generated, and before the second electrical signal is generated, controlling the first row of pixel circuits to expose again, generating a third electrical signal; generating a first GS image according to the first electrical signal and the second electrical signal; controlling the last row of pixel circuits to expose again, generating a fourth electrical signal; generating a first RS image according to the third electrical signal and the fourth electrical signal. The method can further improve the output frame rate of the GS image sensor, and meet the needs of users for high frame rate applications.

This invention relates to a stereo calibration device adapted to high-speed binocular vision scenarios, comprising an L-shaped calibration plate body, which is composed of two perpendicularly intersecting calibration plates, and a preset calibration pattern composed of multiple self-illuminating calibration points is provided on the calibration surfaces of the two calibration plates; a control module, electrically connected to the self-illuminating calibration points, is used to control the emission timing of the self-illuminating calibration points, so that their flashing frequency is synchronized with the shooting frame rate of the camera in the high-speed binocular vision system. This invention also provides a stereo calibration method adapted to high-speed binocular vision scenarios based on the above device. This invention adapts to the short exposure and high frame rate requirements of high-speed cameras, eliminates the need for supplementary lighting and moving the calibration plate, and filters high-quality calibration images through intra-modal and inter-modal dual consistency indices, effectively improving the accuracy of calibration parameters while completing calibration quickly.

The application relates to the field of computer graphics and real-time rendering technology, and particularly discloses a three-dimensional animation scene construction method based on a real-time rendering engine. The method comprises the following steps: voxelizing three-dimensional scene geometry data; deploying a simplified computational fluid dynamics algorithm in a GPU computing shader to construct and initialize a fluid state field; iteratively solving a fluid dynamics equation every frame to update velocity, pressure and density fields; mapping the fluid state field to a volume rendering pipeline to drive the dynamic visual performance of environmental effects such as smoke and airflow; and synchronously detecting the interaction area of the fluid and flexible body or vegetation, and applying force feedback in line with the laws of fluid mechanics to drive deformation response. The application realizes movie-level fluid realism and physical accuracy while ensuring a high frame rate, and improves the immersion and interactivity of a three-dimensional animation scene.

This invention discloses an automatic finish line camera image interpretation system based on UWB ranging, belonging to the field of sports timing technology. The system includes: a starting sensor for capturing the race start trigger signal; a finish line camera timing unit for acquiring a sequence of finish line images; a UWB timing unit containing symmetrically arranged first and second base stations and electronic tags worn by athletes, employing multi-chip parallel ranging and linear fitting algorithms to calculate the finish line crossing time; and a data fusion and interpretation server that maps the UWB finish line crossing timestamp and identity information to image frames, generating a virtual auxiliary interpretation line. This invention achieves system-wide clock synchronization through a temperature-compensated crystal oscillator and time compensation mechanism, mapping the UWB-determined identity information and finish line crossing timestamp to high frame rate image frames to generate a virtual auxiliary interpretation line. This assists referees in quickly confirming results, effectively solving the recognition problems caused by overlapping occlusion, ambient light interference, and multi-lap races, achieving millisecond-level automatic image interpretation.

A video super-resolution device includes a down-sampler, video super-resolvers, and a selecting / averaging unit. The down-sampler divides an input low-resolution video into a plurality of frame rates. The video super-resolvers are video super-resolvers trained at different frame rates, and perform super-resolution of the low-resolution video. The selecting / averaging unit selects a video super-resolver according to the magnitude of an optical flow obtained as a result of video super-resolution. Specifically, in a case where the optical flow has a value smaller than 0.5 pixels, the selecting / averaging unit makes selection such as not adopting a result of video super-resolution at a high frame rate. Finally, the selecting / averaging unit obtains a mean value of the selected video super-resolution result, and outputs a final high-resolution video.

The present disclosure relates to an animation playing method and device, electronic equipment and medium. The animation playing method comprises: determining a first frame rate of a first opening screen animation to be played in an application; performing frame interpolation processing on the first opening screen animation according to the first frame rate to obtain a second opening screen animation, the second opening screen animation having a second frame rate, the second frame rate being greater than the first frame rate; and playing the second opening screen animation. The second opening screen animation is obtained by performing frame interpolation processing on the first opening screen animation, and the second opening screen animation can be played at a higher frame rate, thereby improving the smoothness of animation playing and effectively improving the user experience.

The application discloses a method for smoothly displaying a picture in screen projection, a related device and a system. In the method, in a mirror screen projection process, the screen projection device and the projected device can perform image prediction according to the latest received image frame, and display the predicted image frame when frame loss occurs. According to the application, the projected device side can continuously display the picture, maintain the continuity of the picture in vision, ensure the high frame rate of the projected device side, and avoid the problem of lag. In addition, the image prediction according to the latest received image frame can guarantee the smoothness and stability of the picture, avoid the visual jump, and provide a good screen projection experience for the user.

This invention discloses a bidirectional light alignment control method for MEMS optomechanical systems, comprising the following steps: using MEMS closed-loop phase-locked control to track the MEMS resonant frequency in real time via MEMS driving frequency; and using laser emission delay correction to compensate for bidirectional stripe misalignment caused by the deviation between the driving frequency and the resonant frequency. This invention can solve the bidirectional alignment problem of structured light when driving MEMS at the resonant frequency by coordinating MEMS closed-loop phase-locked control and laser emission delay correction, combined with precise driving of the DDS module and high-precision processing of the piezoresistive signal. This ensures high frame rate while avoiding optical power waste, significantly improving light energy utilization and 3D imaging quality.

The application relates to a high-frame-rate elastic imaging method and system. The method comprises the following steps: based on a specified frame frequency F, a target imaging area is regionally divided to form F sub-ROI areas which are sequentially arranged and meet the effective propagation of shear waves; during elastic imaging, at least one imaging frame rate expansion process is performed, wherein each time the imaging frame rate expansion process is performed, the following steps are included: two sequentially generated imaging area shear wave velocity sets are acquired, and velocity splicing processing is performed on the two acquired imaging area shear wave velocity sets to form at least F target area shear wave velocity information, based on each target area shear wave velocity information, a shear wave velocity image corresponding to the target imaging area is generated, and the generated shear wave velocity image is synchronously displayed with a B-ultrasound image. The application can effectively improve the imaging frame rate without increasing interpolation data.

The application discloses a target capturing method based on vision and 3D radar, and belongs to the field of data processing, aiming to solve the problem of high precision requirement of 3D laser radar in identifying target objects. In the method, the target object is detected in real time through visual recognition, and when the target object is detected, the central angle thereof is output and transmitted to the 3D laser radar algorithm part through ROS2 topic communication. After the 3D laser radar receives the output angle value a0, data processing is performed to obtain a fifth distance; and then the obtained M fifth distances are sequentially subjected to average filtering, low-pass filtering, Kalman filtering and difference filtering, so that the final distance is obtained. The application can realize real-time capturing and positioning of the target object at a frequency of more than 10 Hz at a low cost, and complete the target object tracking task with long distance, high precision, high reliability and high frame rate. The final distance error obtained by the method is less than 5 cm.

This invention provides an Electron ad-dominating optimization method based on polymorphic rendering and state awareness, applied in the field of data processing technology. This application targets Electron ad-dominating applications, using BrowserWindowAPI and Node.js powerMonitor to monitor window focus, power status, and user interaction to determine three playback states: active, low-frequency, and background. The rendering mode is dynamically switched based on the state: GPU acceleration + 60fps high frame rate during active playback, CPU software rendering + 15fps low frame rate during low-frequency playback, and off-screen rendering with cached frame data during background playback. Partial redrawing is achieved by combining OffscreenCanvas and WebWorker, reducing unnecessary computation. In the background state, resources such as DOM nodes and GPU memory are actively reclaimed, and focus is quickly restored, all without user awareness, significantly reducing resource consumption and energy usage, and improving operational stability.

This application relates to a video generation method, apparatus, computer device, and storage medium. The method includes: acquiring a first video stream captured by a main camera at a first frame rate, and a first event stream output by an event camera in a first acquisition mode; obtaining an event density based on the first event stream; if the event density meets a preset condition, controlling the event camera to switch to a second acquisition mode and acquiring a second event stream output by the event camera; and obtaining a second video stream based on the first video stream and the second event stream. This method enables low-power, long-duration, ultra-high frame rate slow-motion video generation.

This application discloses a DMD high frame rate display system and method that matches photoelectric integration time. The system includes a detector, a host computer, a main control board, a DMD, a light source driver, an illumination system, and a projection screen. The DMD calculates the frame period of the displayed image based on phase modulation information, the detector's synchronization signal, and photoelectric integration time data, and calculates the maximum frame rate of the displayed image based on binary pulse width modulation. When the detector's photoelectric integration time changes, the changed photoelectric integration time is acquired, and a new time base is calculated. The parameters of the registers in the control timing are modified according to the new time base to adjust the image display frame rate. This application can use the detector's synchronization signal to deduce the display time base from the detector's photoelectric integration time, achieving synchronization between the detector's photoelectric integration time and the DMD display frame rate. This avoids false signals, flickering, and image confusion, and achieves optimal imaging results.

The present application relates to the field of intelligent aquaculture and computer vision technology, in particular to a kind of abalone feed intake monitoring system based on image recognition, comprising: image acquisition module, for collecting high frame rate video stream data containing biological activity and feed distribution;Physical reference reconstruction module, based on initial feed characteristics, combined with fluid dynamics and dissolution parameters, deduce the theoretical evolution reference map of the current time;Behavior probability generation module, identify biological mouth key points and motion vectors, generate dynamic predation probability field;Double rail difference module, difference operation is carried out to real image and theoretical evolution reference map, and non-physical disappearance residual area is generated;Coupling verification module, residual area is mapped into probability field, and effective feeding behavior is determined based on spatial overlap degree;Data output module, according to the determination result, the accurate intake data is output cumulatively;The present application solves the counting problem caused by shielding, and significantly improves the monitoring robustness.

This application relates to the field of jitter processing for guide chair data, and discloses a method, apparatus, computing device, and storage medium for jitter processing of guide chair data. The method includes: acquiring point clouds collected by a lidar and video frames collected by a high frame rate camera; using two adjacent point cloud frames as target point clouds, acquiring target video frames that match the target point clouds; calculating target rotation and target translation matrices based on the target video frames; calculating target correction point clouds corresponding to the target point clouds based on the target rotation and target translation matrices; and fusing the target correction point clouds with the target video frames to obtain fused data. This application performs jitter processing on the collected lidar data using video frames collected by a high frame rate camera, and generates fused data after correction, reducing hardware costs and improving the accuracy of guide chair navigation and obstacle avoidance.

The present application relates to a kind of paper bag film multi-sensor online electric control and self-adapting calibration method, the present application is based on the sub-pixel level surface dynamic deformation acquisition method of high frame rate structured light vision system, combine multiple feature low-redundancy fingerprint vector Construction deformation-process variable mapping, realize typical coupling mode discrimination and independent compensation adjustment of leading influence variable.Through the combination of offline / on-line clustering and incremental learning, support the automatic adaptation and closed-loop dynamic optimization of new coupling mode.The present application can realize high-precision deformation monitoring under complex working conditions in paper-based material film coating process, accurate closed-loop regulation of main control variable and multivariable decoupling control, effectively improve the process stability and adaptive intelligent level.The present application introduces the leading mode recognition architecture based on the combination of offline clustering and online similarity matching, which greatly reduces the computational complexity of control algorithm;While ensuring control accuracy, reduce redundant adjustment action, effectively inhibit the risk of multi-loop interaction oscillation.