25results about How to "Eliminate jitter" patented technology

The invention discloses an abnormal behavior identification method based on a visual large model and a cognitive Agent, and relates to the technical field of computer vision, and the method comprises the steps: collecting an RGB image and an infrared thermal imaging IR image in a monitoring scene, and extracting a human body key point sequence; performing time sequence difference processing on the key point sequence, and constructing a skeleton velocity vector containing a motion trend; inputting the skeleton velocity vector into a Mamba model, adjusting model parameters by using a selective state space mechanism, and carrying out long-time-sequence modeling on the action sequence; predicting the speed variation of the next frame by adopting a residual prediction architecture and reconstructing a human body posture; and calculating an abnormal score according to the prediction error, triggering a cognitive Agent when the score exceeds a preset threshold value, calling a visual large model (VLM) to perform semantic analysis on the current scene environment, and generating a natural language warning containing an abnormal type and an environmental cause. According to the method, the forgetting problem of a traditional model in a long sequence and the jitter problem of a prediction action are solved.

This application discloses a low-power temperature-compensated RC oscillator, comprising a low-voltage linear buffer, an RC core circuit, a comparator, a Schmitt trigger, a non-overlapping clock generator, and a level shifter. The low-voltage linear buffer uses intrinsic transistors to reduce operating voltage and power consumption; the RC core circuit employs temperature compensation technology, including a series low-temperature drift resistor and a temperature-dependent current source, causing the equivalent resistance to increase with temperature; the comparator compares a triangular wave signal with a reference voltage and outputs a reference clock; the Schmitt trigger filters out clock glitches; the non-overlapping clock generator generates a two-phase clock with dead time to control charging and discharging; and the level shifter converts the low-voltage domain clock to the high-voltage domain. This oscillator eliminates the influence of comparator offset voltage through two-phase switching control and employs low-voltage and temperature compensation technology to achieve a high-precision, low-temperature drift reference frequency in a small area with low power consumption, making it suitable for compact sensing and measurement systems.

This application provides a mobile communication message transmission and reception system and method. The system includes a BBU unit, a HUB unit, and an RRU unit. The HUB unit includes a message receiving module, a first FIFO memory, a timer module, a message reading module, a message decompression module, a message merging and compression module, a second FIFO memory, and a message sending module. The timer module sends a reminder message to the message reading module when a set time is reached. The message reading module reads the message data sent by each RRU unit stored in the first FIFO memory. Upon receiving the reminder message from the timer module, it sends the currently read message data from each RRU unit to the message decompression module. In this embodiment, because a timer is used, messages can be sent to the BBU unit at a constant interval in different RRU unit relocation scenarios, eliminating message jitter and saving the capacity of the second FIFO memory.

This invention belongs to the interdisciplinary field of power management and radiation measurement, aiming to solve the problem of inaccurate counting caused by shared power supply in radiation measurement. It provides a seamless power switching device and method for radiation measurement equipment. The switching device includes a controlled power-off unit, a transient competitive power supply network, and a static path maintenance unit. The controlled power-off unit is configured to connect the external power supply to the transient competitive power supply network in non-measurement states, and to disconnect the electrical connection between the external power supply and the transient competitive power supply network upon receiving a radiation measurement trigger signal. The transient competitive power supply network is configured to automatically transfer the power supply path of the static path maintenance unit from the external power supply to the internal energy storage power supply when the output voltage of the controlled power-off unit drops below the voltage of the internal energy storage power supply. The static path maintenance unit is configured to maintain connection with the radiation measurement equipment. This invention can solve the problem of inaccurate counting in radiation measurement equipment under shared power supply scenarios.

The application relates to a video frame processing method and device, computer equipment, a storage medium and a computer program product. The method can be applied to cloud conference, cloud storage, artificial intelligence and intelligent transportation scenes, and the method comprises the following steps: determining the posture of a key part of a target object in each video frame carrying noise; determining the deflection degree value of the key part in each video frame according to the posture of the key part; generating a control signal based on the deflection degree value of the key part in each video frame; performing back gap filtering on the control signal according to a back gap filtering coefficient through a back gap filter to obtain a deflection update value of the key part in each video frame; and sequentially performing image processing on each video frame based on the deflection update value in each video frame. The method can effectively eliminate the jitter caused by noise when the video frame is subjected to image processing, and is beneficial to improving the stability of the video frame after image processing.

The invention relates to the technical field of optical engineering, and discloses a handheld laser directional energy system which comprises a Shack-Hartmann wavefront detection system, a visual observation system, an atmospheric turbulence correction system, a laser emission system, a laser ranging system and an anti-shake system. By integrating wavefront detection and adaptive correction, coaxial design, target background difference enhancement through light splitting, MEMS reflector dynamic adjustment, jitter compensation and other technologies, the problems that the imaging quality is interfered by atmospheric turbulence, correction is inaccurate due to system space separation, the contrast ratio of a dark and weak target is poor, aiming is difficult, and jitter miss is caused are solved; the effects of high-resolution imaging, energy concentration, easy target discovery, automatic locking and stable aiming are achieved.

The embodiment of the invention provides an electronic image stabilization method, which is used for processing visible light video data of an airborne photoelectric hanging bin, and comprises the following steps of: 1, acquiring video sequence basic information; step 2, feature point detection and motion estimation; step 3, performing abnormal processing on the motion model; step 4, motion trail smoothing processing; 5, image stabilization transformation and post-processing are carried out; and step 6, outputting a result. Aiming at visible light video data of an airborne photoelectric pod, various forms of jitter motion can be accurately captured and compensated through multi-scale feature point tracking and intelligent motion estimation. Compared with a traditional method, the method shows higher precision and stability when a complex motion mode is processed. Especially in scenes with slight motion blur, illumination variation or partial shielding, the method can still keep reliable image stabilization performance. The smooth processing of the motion trail enables the finally output video to eliminate uncomfortable jittering and maintain the intentional motion of the camera, thereby conforming to the visual perception characteristics of human beings.

The utility model discloses a kind of hardware anti-shaking electronic switch circuit, including signal input filter voltage stabilizing circuit, signal anti-shaking inverting circuit and signal buffer circuit;Wherein, the input end of signal input filter voltage stabilizing circuit is connected after the input signal needing anti-shaking outside, the output end of signal input filter voltage stabilizing circuit is connected with the signal anti-shaking inverting circuit, signal buffer circuit;The output end of signal anti-shaking inverting circuit is connected with the input end of signal buffer circuit, and the output end of signal buffer circuit is connected to the external circuit of receiving anti-shaking signal.The utility model can effectively eliminate mechanical contact shaking, ensure the stability and accuracy of signal, applicable to most circuits needing signal anti-shaking.Meanwhile, circuit structure is easy to realize, and cost is low.

A defect detection method and system for photovoltaic cells, an initial point pair set corresponding to each line laser profiler is obtained, a first correction slope corresponding to the line laser profiler is calculated, and then the initial point pair set corresponding to the line laser profiler is profile corrected to obtain a real-time point pair set corresponding to the line laser profiler; a first type of defect detection result is obtained according to the real-time point pair set corresponding to each line laser profiler; and / or a second type of defect detection result is obtained according to the real-time point pair set corresponding to each line laser profiler and the standard profile data of the three-dimensional vision sensor corresponding thereto. The error caused by mechanism jitter, installation non-parallelism and the like is suppressed or removed through profile correction and one-dimensional signal filtering and the like, so as to improve the repeatability and stability of measurement, so that the measurement data is close to the true value, and then the accuracy of the defect detection result is ensured, and on this basis, integrated measurement of various defects is realized.

The invention discloses an automatic nipple opening positioning system and guiding system for duodenal nipple intubation guidance, solves the problem of how to realize high-precision and high-robustness positioning of a nipple opening under multiple constraints, and belongs to the technical field of medical image processing and computer-aided surgical navigation. According to the invention, a generative model is utilized to actively deduce anatomical features of a shielded area, and continuous tracking in an instrument shielding scene is realized; establishing an absolute mapping relation between a pixel position and a target coordinate through injection explicit space geometry prior, and obtaining a nipple opening pixel-level coordinate; visual uncertainty estimation and instrument-tissue kinematics coupling are fused, a physically constrained adaptive filtering mechanism is constructed, and smooth coordinates are obtained; the positioning system is used for constructing a guiding system, closed-loop interaction of instrument control through a handle and inlet coordinate positioning feedback is achieved in an operation environment, and dual-scene application of preoperative training and intraoperative navigation is supported.

The invention relates to the technical field of puncture in the interventional department, and provides an intelligent depth-adjustable puncture device for the interventional department, which comprises a universal flexible arm, and one end of the universal flexible arm is provided with a fixed seat connected with external equipment and an angle adjusting assembly. The angle adjusting assembly is installed at the other end of the universal flexible arm through a connecting frame, the puncture assembly is arranged on the angle adjusting assembly, through the arrangement of the structures such as the multi-section flexible arm and the transmission line, the multi-section flexible arm is composed of a plurality of hinged joint units, and the transmission line is connected with the multi-section flexible arm. The transmission lines are symmetrically arranged on the two sides of the arm body and connected to the tail end joint, the retracting and releasing amount is accurately controlled through the first power end and the second power end, the arm body is accurately bent in a three-dimensional space, it is ensured that an end effector stably reaches a preset target spot area, the problem of path deviation caused by the fact that the bending direction of a traditional flexible pipe is uncontrollable is solved, and the bending efficiency is improved. The problems of easy deviation and insufficient stability in the prior art are solved.

This invention discloses a method for generating smooth trajectories for robotic arms that combines visual keypoint cues with a hierarchical diffusion model, applicable to complex robotic arm operations based on visual and verbal commands. The method constructs a hierarchical robotic arm operation framework. The high-level layer utilizes a sub-target generator guided by a visual-language model to predict the six-DOF pose sequence of the end effector as operation sub-targets from visual observations and verbal commands. The low-level layer proposes a dual diffusion model trained based on a random sampling strategy to generate smooth, low-curvature joint position trajectories between sub-targets. A keypoint extraction module that requires no training is introduced to provide visual cues. This invention improves the success rate of operations, reduces trajectory jitter and redundant motion, and enhances the smoothness and execution stability of robotic arm operations through the synergy of hierarchical planning and diffusion trajectory optimization.

This application discloses a self-adjusting transmission device for a circular vibrating screen, relating to the field of circular vibrating screen technology. The application includes a frame, a motor mounted on the frame, and a second coupling mounted on the vibrator. A first pulley is mounted on the output shaft of the motor. A bearing housing is mounted on the frame, and a main shaft rotatably passes through the bearing housing. Two spaced-apart roller bearings are disposed between the bearing housing and the main shaft. The second pulley and the first coupling are mounted on the main shaft. This application abandons the direct V-belt connection transmission method of the prior art. In use, the V-belt transmission drives the main shaft and the first coupling to rotate together, causing multiple rubber belts to move and deform. Through the pulling action of the multiple rubber belts, the second coupling rotates, enabling the vibrator to operate. Because the second coupling and the first coupling are transmitted through multiple rubber belts, their connection is close to flexible.

The application discloses a time segmentation-based dimming control method and device and a medium, and relates to the technical field of dimming, and comprises the following steps: acquiring a current PWM value and a target PWM value of a lighting device, and calculating a difference value between the target PWM value and the current PWM value; determining a current adjustment stage from a plurality of preset adjustment stages according to a numerical interval of an absolute value of the difference value, and calculating an adjustment step length based on a step length strategy corresponding to the current adjustment stage; and updating the current PWM value based on the difference value and the adjustment step length. The mechanism of dynamic adjustment based on the difference value interval dynamic segmentation avoids repeated oscillation caused by improper adjustment strategy near the target value when a single fixed step length or a single proportional step length is used, thereby achieving smooth transition of PWM dimming and effectively eliminating the jitter near the target value.

The surgical assistive robot and control method based on a continuum configuration described in this invention belong to the field of surgical assistive equipment technology. It includes a support mechanism, a motion positioning mechanism, a mounting mechanism, and a mechanical execution mechanism. The support mechanism consists of a horizontal support frame, a vertical support frame, and an inverted T-shaped frame. The motion positioning mechanism includes a gimbal and a sliding table mechanism. The mounting mechanism utilizes a hydraulic telescopic device I and a rotary motor I to flexibly adjust the height and rotation angle of the upper and lower fixed platforms. The mechanical execution mechanism, through a rotary motor II, drive motor I, drive motor II, drive motor III, and an adjusting motor, achieves precise adjustment of the extension and end-effector posture of the movable upper and lower arms. The control method includes positioning initialization, mounting mechanism adjustment, mechanical execution mechanism deployment, execution device fine-tuning, surgical operation execution, and dynamic feedback and correction, achieving high precision and flexibility in operation and effectively overcoming the difficulties in master-slave hand mapping and the limitations on surgical flexibility.

The application discloses a double-cavity air pressure buffer with controllable reset process, which comprises a buffer system and a limiting system, and the limiting system is connected with the buffer system through a clamping pin base and a thrust bearing which are matched with each other; the buffer system comprises an outer cylinder, an inner cylinder, a piston rod, a separation valve, a reset spring, an upper end cover and a lower end cover; the end, which is in contact with the inner cylinder, of the outer cylinder is respectively provided with an inwardly-contracted protrusion and an outwardly-expanded protrusion; the outer surface of the outer cylinder is provided with an outwardly-expanded circular table structure; the outer cylinder is uniformly provided with six air holes; and the piston rod is provided with four evenly-distributed damping holes. The application can limit the reset movement of the buffer after the buffer is compressed by impact, reduce the shaking and shaking after the buffer process is finished, and realize the active control of the reset process of the buffer; the double-cavity design with adjustable high-pressure chamber pressure enables the buffer to still have good buffering capacity under different loads, and widens the working range of the buffer.

This invention relates to the field of remote monitoring and image processing technology for autonomous driving, and discloses a remote video image processing method and system based on an autonomous driving vehicle platform. The method includes: acquiring images from multiple wide-angle cameras and six-axis IMU data, and synchronizing multi-source sensor data; performing attitude prediction and vibration mode separation based on the synchronized IMU data, outputting pixel displacement prediction values, predicted homography matrix, and relative rotational perturbations between cameras; obtaining an independent stabilized image through adaptive image stabilization transformation; performing dynamic compensation and spatiotemporal consistency fusion of the surround-view stitching seams based on the relative rotational perturbations to generate a surround-view panoramic image; and outputting the final image based on adaptive distortion suppression using a human spatial perception model. This invention improves the smoothness and spatial perception reliability of remotely controlled images under vibration conditions on unpaved roads by jointly optimizing IMU high-frequency prediction and sparse optical flow, and dynamically adjusting seams and fusion weights in the stitching transition zone.

The present application belongs to the technical field of power management and radiation measurement, and aims to solve the problem of inaccurate counting caused by shared power supply during radiation measurement. A power seamless switching device and method for radiation measurement equipment are provided. The switching device includes a controlled power-off unit, a transient competitive power supply network, and a static path retention unit. The controlled power-off unit is configured to turn on the external power supply to the transient competitive power supply network in the non-measurement state, and cut off the electrical connection between the external power supply and the transient competitive power supply network at the moment of receiving the radiation measurement trigger signal. The transient competitive power supply network is configured to automatically transfer the power supply path of the static path retention unit from the external power supply to the internal energy storage power supply when the output voltage of the controlled power-off unit drops below the voltage of the internal energy storage power supply. The static path retention unit is configured to maintain the connection with the radiation measurement equipment. The present application can solve the problem of inaccurate counting of radiation measurement equipment in the shared power supply scenario.

The application discloses a kind of multi-screen display synchronous display optimization methods.This method includes: obtaining the extended display identification data of each display and timing parameter;Based on the actual time stamp of the vertical synchronization signal of each display clock monitor, construct timing drift characteristic model;Select timing reference source or generate virtual reference clock;Delay compensation instruction is inserted in GPU rendering command stream, and the scanning start offset address of the frame buffer of secondary display is adjusted dynamically, and the physical clock phase difference between each display is compensated.In addition, for the mixed refresh rate scene, a proactive double buffering queue is established, and a complete frame is selectively output based on the tear risk index.The application can achieve microsecond-level accurate synchronization of multi-screen display at the software level without external Genlock hardware, effectively eliminating the tearing phenomenon caused by heterogeneous screen clock drift and the periodic stutter caused by mixed refresh rate.

The invention relates to the technical field of hydraulic pressure, in particular to a control mechanism for a rotary hydraulic device, which comprises an oil port A, an oil port B and a speed change valve, the speed change valve comprises a speed change valve core and a speed change channel, and a speed change valve cavity is arranged at the end part of the speed change valve core and is used for controlling the speed change valve core to move between a first speed position and a second speed position; the control valve is communicated with the oil port A and the oil port B and comprises a control valve core, and a load valve cavity is formed in the control valve core; when high-pressure oil enters the oil port, the control valve element is at the initial position, the variable-speed valve element is arranged at the first speed position, and the control valve element is switched to the control position, the high-pressure oil is introduced into the variable-speed valve cavity through the variable-speed channel from the control valve, the variable-speed valve element moves to the second speed position, the high-pressure oil enters the load valve cavity and acts on the control valve element, and when the pressure reaches a set value, the variable-speed valve element moves to the load valve cavity. And the control valve element is pushed back to the initial position, so that the variable-speed valve element moves back to the first speed position. The technical problems that in the prior art, the speed needs to be switched manually, and the speed cannot be adjusted automatically are solved.

The application discloses a kind of high-efficiency filtering interframe jitter's in-car human body key point detection method and system, belong to computer vision technical field.The method includes: by vehicle-mounted camera is photographed in-car, obtains in-car multiple frames of pictures;Using pre-trained convolution network model, the current frame of picture is handled according to preset seat number Human regression, determine the preset seat number personal body bounding box on current frame of picture;The human body in human body bounding box is detected, determine human body key point;The human body key point in each human body bounding box of current frame of picture is carried out with the human body key point in the corresponding human body bounding box of historical frame of picture Smooth filtering processing.The application determines a certain number of human body bounding boxes according to the number of in-car seats, reduces the difficulty of model processing problem, reduces data processing amount, simultaneously using a certain number, independent human body bounding box is associated between multiple frames The human body key point belonging to the same human body is efficiently eliminated interframe jitter.

The application belongs to the field of aerial remote sensing stable platform control method, and relates to an aerial platform control method based on P-I inverse model and self-disturbance sliding mode, which comprises the following steps: a P-I hysteresis nonlinear model of a piezoelectric actuator is established, and an inverse model is constructed as a feedforward controller, which is used for compensating the known hysteresis nonlinearity; a feedback controller containing an extended state observer (ESO) and a super-twisting sliding mode (STSM) is constructed, the ESO is used for estimating the lumped unknown disturbance in real time, and the feedback control quantity is calculated based on the disturbance estimation value through the STSM; finally, the feedforward and feedback control quantities are superposed to drive the stable platform. This control mode can simultaneously solve the hysteresis nonlinearity of a piezoelectric ceramic (PZT) actuator, the chattering problem of a traditional sliding mode and the suppression problem of external strong disturbance, and high-precision tracking and strong robustness of the aerial remote sensing stable platform are realized.

This invention relates to the field of graphics card chip technology and discloses a method and system for dynamic scheduling of computing power in a graphics card cluster. The method includes acquiring graphics card utilization and network traffic sampling sequences, obtaining computationally and communication-intensive stage label sequences through timestamp alignment and stage analysis; calculating the peak computing power and bandwidth requirements of the two types of stages and prioritizing tasks; matching target graphics card groups, generating reserved bandwidth channels by pre-allocating network paths in conjunction with stage switching windows; and synchronizing execution parameters and updating the running status sequence in a closed loop. This method can achieve precise dynamic scheduling of graphics card cluster computing power, meeting the dual requirements of high-performance computing efficiency and resource utilization.

The invention relates to the technical field of image and video anti-shake, and discloses a hybrid anti-shake method and system combining optical anti-shake and electronic anti-shake. The hybrid anti-shake method is applied to a hybrid anti-shake unit, and specifically comprises the following steps: S11, acquiring original data of angular velocities and linear accelerations of a terminal device in three axial directions acquired by a multi-axis IMU, and acquiring actual displacement vectors of a position sensor and an image sensor in the terminal device during optical anti-shake execution; and establishing a uniform timestamp to carry out precision synchronous alignment on the data of the multi-axis IMU, the optical anti-shake position sensor and the image sensor, and generating a motion track estimation feature. According to the method, the actual displacement feedback of the optical anti-shake position sensor is used as an independent sensing source, hardware-level high-precision synchronous fusion is carried out on the independent sensing source, the multi-axis IMU and the image data to construct the motion model, intentional motion and non-intentional shake are accurately distinguished from the source, and the problems of rough motion sensing and weak cooperative basis are solved.

The present application relates to the technical field of thermal power generation, and discloses a combined cycle generator unit one-key start-stop energy-saving regulation and control device, which comprises a pipe body installed on a drain main pipe at the bottom of an auxiliary steam header in an ABS; a control group installed in the inner cavity of the pipe body along the drain direction, the control group comprising a plurality of floats, the moving direction of the floats being set as a first direction, the heights of the floats gradually increasing along the drain direction, a first permanent magnet being fixed at the bottom of the float through a second plate body, and a second permanent magnet with opposite magnetic poles being arranged at the bottom of the first permanent magnet; by designing the float as a bellows and filling the bellows with phase-change working medium, the volume of the float is self-adaptively changed with temperature, the temperature compensation of the buoyancy is realized, the problems of delayed opening of the valve in cold-state starting and early opening of the valve in hot-state starting are solved, the stability of the water level of the valve in different temperature conditions is ensured, the warm-up efficiency is improved, and the waste of high-temperature steam is reduced.