Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

192results about How to "Improve tracking performance" patented technology

Autoregressive visual target tracking algorithm based on token fusion

The invention discloses an autoregressive visual target tracking algorithm based on token fusion, and the algorithm specifically comprises the following steps: 1, obtaining a public visual target tracking data set, and converting the data set into a central coordinate format; 2, preprocessing an initial template of a given video sequence, an online template at the current moment and a search image to obtain visual embedding of the initial template, a dynamic template and the search image; the method comprises the following steps: preprocessing an angular coordinate sequence of a target frame in previous historical N frames of a current search image of an input video to obtain historical track word embedding of the current search image; 3, constructing an autoregressive visual target tracking model based on token fusion, and designing a loss function; step 4, obtaining a final autoregressive visual target tracking model based on token fusion; the method solves the problem of insufficient model feature extraction capability caused by insufficient utilization of spatio-temporal information and redundant background information of a search area in an existing method.
Owner:XI'AN POLYTECHNIC UNIVERSITY +1

AL-NN neural network integrated navigation method based on attention mechanism

The invention discloses an AL-NN neural network integrated navigation method based on an attention mechanism, and relates to the technical field of navigation positioning, and the integrated navigation method comprises the following steps: firstly, obtaining information including GNSS position information, data of an accelerometer and a gyroscope of an IMU, and position, speed and attitude information calculated by an INS; then, establishing a relation model of IMU output, INS solution and GNSS position increment, and constructing an AL-NN neural network model; when the GNSS signal is available, training the model; when the GNSS signal loses lock, the model is used for predicting the GNSS position increment during the lock losing period; and finally, adaptive Kalman filtering fusion is carried out through the GNSS position output by the model and the adjusted measurement noise covariance matrix. According to the AL-NN neural network integrated navigation method based on the attention mechanism, the calculation complexity can be reduced, and meanwhile, the navigation robustness and precision are effectively improved.
Owner:XIANGTAN UNIV

Servo hydraulic system optimization control method and system for press machine test platform

The invention provides a servo hydraulic system optimization control method and system for a press machine test platform, and relates to the technical field of press machines, and the method comprises the steps: building a state space expression and building a mathematical model of a servo hydraulic system based on dynamic characteristics by deducing mathematical expressions of all components of the system; building a predictive control algorithm based on the mathematical model, solving a quadratic programming problem by adopting a rolling optimization strategy through building a state space predictive model and determining a multi-objective optimization function to obtain an optimal control sequence, and taking a first element of the optimal control sequence as control input at the current moment; feedforward compensation and pressure feedback compensation are introduced, the feedforward compensation constructs a compensation model based on an inverse system of a press transfer function and pre-corrects a control signal, and the pressure feedback compensation designs a digital filter based on a pressure error, fuses a feedforward compensation signal with a pressure feedback compensation signal and superposes the fused signal with the control signal; and obtaining a final control signal.
Owner:TIANJIN HANYANG METAL EQUIP

Camera adaptive adjustment method and system, and storage medium

The invention relates to the technical field of camera adjustment, in particular to a camera adaptive adjustment method and system and a storage medium. The method comprises the following steps: acquiring a multi-source environment sensing data set and an initial environment image; performing environment feature recognition on the initial environment image according to the multi-source environment sensing data set to obtain a scene environment feature map; performing semantic scene segmentation on the scene environment feature map to obtain a semantic scene macroscopic feature map; performing time sequence attention target tracking on the scene environment feature map to obtain a target vehicle motion feature map; performing feature constraint fusion on the semantic scene macroscopic feature map and the target vehicle motion feature map to obtain a target vehicle-scene motion semantic map; and performing parameter adjustment on the camera according to the target vehicle-scene motion semantic map to obtain a camera parameter adjustment strategy set. According to the invention, the performance and the image quality of the camera in a changeable environment are improved.
Owner:SHENZHEN TUNGSON AGES TECH CO LTD

Path planning method and system for automatic driving trolley

The invention provides a path planning method and system for an automatic driving trolley, and relates to the technical field of path planning, and the method comprises the steps: obtaining the data of an acoustic sensor of an underwater environment, and obtaining the contour of the underwater terrain and the position information of an obstacle; carrying out dynamic environment modeling processing; the Euclidean distance is used as a heuristic function to calculate the dynamic model of the underwater environment, and a second optimized path conforming to kinematics and dynamics constraints of the automatic driving trolley is obtained; carrying out smoothing processing on the second optimization path through a dynamic motion primitive technology; constructing a three-dimensional map for path planning and obstacle avoidance; predicting an action track of the robot in a period of time in the future; and through real-time adjustment processing of the self-adaptive control system, path planning of the automatic driving trolley is completed. According to the method, the dynamic characteristics of the underwater environment and the motion constraint of the automatic driving trolley can be comprehensively considered, the three-dimensional map containing the obstacle and topographic information can be constructed in real time, and the future motion trail of the obstacle can be predicted.
Owner:CHENGDU UNIVERSITY OF TECHNOLOGY

Unmanned aerial vehicle target intelligent identification and perception system and method based on deep learning

The invention discloses an unmanned aerial vehicle target intelligent identification and perception system and method based on deep learning, and relates to the technical field of artificial intelligence, and the method comprises the steps: extracting the importance feature of each target region, calculating a region importance score, generating a target region monitoring plan, and obtaining continuous frame images; an improved YOLOv5 model is adopted, continuous frames of images are input, a bounding box of a suspected target wild animal is output and mapped to an actual geographic coordinate system, and actual position coordinates are obtained; key points of the t-th frame image are detected, coordinates of the key points are extracted, visibility scores and confidence coefficients are calculated, and feature descriptors are formed; performing target detection and key point detection on the (t + 1) th frame of image to obtain actual position coordinates and feature descriptors of candidate targets, performing track matching, updating the actual position coordinates and the feature descriptors of the suspected target wild animals if the actual position coordinates and the feature descriptors are matched, and defining new target animals if the actual position coordinates and the feature descriptors are not matched; counting according to the track matching result to obtain the total number of the target wild animals.
Owner:TIANJIN ZERO ONE INTELLIGENT TECH CO LTD

Dynamic scene multi-target tracking method fusing trajectory association technology

The invention provides a dynamic scene multi-target tracking method fusing a trajectory association technology, and belongs to the technical field of computer vision and artificial intelligence. The technical problem that in the prior art, the pedestrian tracking capability is insufficient when complex, diversified and crowded scenes are processed under the dynamic background is solved. According to the technical scheme, the method comprises the following steps: S1, video frame preprocessing and target detection; s2, tracking optimization of the fast moving target; s3, track linking and fusion; and S4, track interruption and fragmentization are repaired. The method has the beneficial effects that in order to improve the tracking capability of multi-target pedestrians under a complex dynamic background, accurate extraction and tracking of pedestrian features are realized by adopting an advanced deep learning algorithm and a computer vision technology, and more reliable technical support is provided for the fields of intelligent monitoring, automatic driving and the like.
Owner:NANTONG UNIV

Target detection tracking method and system based on automatic detection

The invention discloses a target detection tracking method and system based on automatic detection, and relates to the technical field of target detection tracking, and the method comprises the following steps: carrying out the target position and category labeling of an image through a labeling tool based on an image data set of a typical target, and obtaining a training data set with a label, establishing a target detection model according to a YOLO architecture, and introducing a dynamic weighted loss function; according to the training data set, a training batch is generated by using a hierarchical sampling strategy, an optimized training sample is obtained through an online difficult case mining mechanism, and iterative training is performed on the target detection model based on the generated training batch and the optimized sample; and performing target detection on the real-time video stream by using the trained target detection model to obtain the tracking position and category information of each frame. According to the method, the problem of class imbalance can be effectively solved, the detection capability of rare targets is improved, and meanwhile, the overall performance of the model is improved through difficult case mining.
Owner:HEFEI HEZHIXIN MICROELECTRONICS CO LTD

Target tracking method based on gated attention mechanism and space-time memory network

The invention relates to the technical field of target tracking in computer vision, and particularly discloses a target tracking method based on a gated attention mechanism and a space-time memory network, which mainly comprises a feature extraction module, a gated attention mechanism feature fusion module, a space-time memory feature reading module and a double-branch prediction module. The method comprises the following steps: inputting a memory frame image and a current query frame image into a backbone network to extract features; fusing the enhanced memory frame image features and the query frame image features by adopting a gating attention mechanism; respectively predicting a target position frame and a segmentation mask result by using a double-branch detection module; and dynamically screening the memory frame template according to the confidence score and the time distance constraint. According to the method, fusion of memory frame image features and query frame image features is enhanced through a gating attention mechanism, space-time memory information is fully utilized, interference caused by conditions that a target is shielded for a long time and non-target similarity is high is effectively reduced, and accuracy and robustness of a tracking model are improved.
Owner:ZHEJIANG UNIV

Mask-guided frequency domain-spatial domain hybrid network target tracking method

The invention provides a mask-guided frequency domain-spatial domain hybrid network target tracking method, which relates to the field of pattern recognition and computer vision, and comprises the following steps: carrying out visual feature extraction on an image pair comprising a search image and a template image to respectively obtain multi-layer and multi-resolution search branch features and template branch features; fusing the extracted multi-layer and multi-resolution search branch features and template branch features to obtain fused backbone network features; stacking the template mask and the template image, and encoding to generate a learnable template mask vector; jointly inputting the fused backbone network features and the generated template mask vector into a frequency domain-spatial domain attention interaction module to promote full interaction and enhancement of template branch features and search branch features, and generating final search branch features; and inputting the generated final search branch features into the tracking prediction head network to generate coordinates of the to-be-tracked target frame. The target tracking performance can be improved.
Owner:UNIV OF SCI & TECH OF CHINA

Unmanned vehicle road obstacle sensing system based on laser radar

The invention relates to the technical field of laser radars, in particular to an unmanned vehicle road obstacle sensing system based on a laser radar, and the system comprises a laser radar data collection module which adjusts the scanning frequency of the laser radar, collects point cloud data from static and dynamic obstacles, and obtains preliminary scanning data; and filtering the preliminary scanning data, and removing noise data. According to the invention, by calculating the dynamic characteristics of the point cloud data, the key point and the moving track of each obstacle are extracted, and the ability of sensing the obstacles in the dynamic environment is optimized. By adopting a time sequence analysis method, the moving state and track of the obstacle can be directly deduced according to the point cloud change, so that the ability of tracking the moving obstacle in a complex dynamic scene is improved. In the spatial reconstruction process, geometric subdivision is utilized to reconstruct point cloud data, and dynamic information is combined with a static point cloud structure, so that the analysis capability of spatial data is enhanced.
Owner:TIANJIN DEXIN AVIATION TECH CO LTD

Target tracking method and device for fast moving small target

The invention provides a target tracking method and device for a fast moving small target, and the method comprises the steps: monitoring the position change of a target in continuous frames of an original target tracking data set, and estimating the motion speed of the target through calculating the difference of the positions of the target of adjacent frames; dynamically adjusting the sampling probability of the video frame according to the target movement speed; according to the adjusted sampling probability of each frame, probability resampling is carried out on the original target tracking data set, and a fast moving target tracking training set is generated; determining hyper-parameters of a speed sensitive loss function based on image features of the fast moving target tracking training set; and performing iterative optimization training of the target tracking model by using the fast motion target tracking training set and the speed sensitive loss function to obtain an optimized target tracking model suitable for the fast motion small target. The problems of inaccurate tracking, loss and the like in a high-speed moving small target tracking process in the prior art can be solved.
Owner:ACADEMY OF MILITARY MEDICAL SCIENCES

Low-energy-consumption three-dimensional point cloud single-target tracking method based on spiking neural network

The invention provides a low-energy-consumption three-dimensional point cloud single-target tracking method based on a spiking neural network, and the method comprises the steps: obtaining the three-dimensional point cloud data of a previous frame and a current frame of a target, designing a voxel grid module, and dividing the point cloud into 3D voxels; a feature extraction module is designed, a three-stage pulse convolution block is used as a feature extractor, voxelization point cloud features are extracted, a Transform encoder layer is added, and global context information of the point cloud is captured through a self-attention mechanism; a part-to-part motion modeling module is designed, continuous point cloud features on a time sequence are modeled and coded part by part through a two-dimensional pulse convolution block, and feature information changes of a previous frame and a current frame in corresponding space parts are obtained; and designing a regression prediction module, and inferring the relative motion of the target through a pulse regression prediction module. According to the method, energy consumption is reduced, efficiency is improved, and meanwhile it is guaranteed that the network accurately tracks the target in diversified scenes and motion states.
Owner:ANHUI UNIV

Flexible joint mechanical arm system finite time adaptive impedance control method and system

The invention belongs to the technical field of force / position hybrid control of mechanical arm systems, and particularly discloses a finite-time self-adaptive impedance control method and system for a flexible joint mechanical arm system. In order to solve the problems that the control effect of a mechanical arm system is affected by nonlinear and unknown system model parameters at present and derivation is complex in backstepping design, the fuzzy self-adaptive finite time impedance controller is designed based on the impedance control technology so as to achieve hybrid control over the track and force of the tail end of the mechanical arm. In the design process of the controller, a real control law is constructed to realize finite time stability of the system; and meanwhile, command filtering and fuzzy self-adaptive technologies are introduced to solve the problems of calculation explosion in backstepping design and unmodeled dynamics of a mechanical arm system. By means of the control method, the influence caused by system uncertainty can be effectively overcome, the position tracking error of the mechanical arm system is reduced, and meanwhile it can be guaranteed that all signals in a closed-loop system are bounded within finite time.
Owner:QINGDAO UNIV

Non-cooperative spacecraft active tracking method based on deep reinforcement learning

The invention discloses a non-cooperative spacecraft active tracking method based on deep reinforcement learning, and relates to a non-cooperative spacecraft active tracking method. The invention belongs to the field of space flight and aviation. The objective of the invention is to solve the defects that an existing algorithm cannot fuse a spacecraft dynamics model and satellite orbit dynamics, cannot effectively improve the robustness of the algorithm on the basis of guaranteeing the tracking precision, cannot effectively extract time sequence related information about a target from a training sample, and the like. The method comprises the following steps: 1, obtaining the position and speed of a chasing spacecraft; 2, constructing an actor network, a commentator network and a loss function; 3, constructing a reward function, and obtaining a trained actor network and a reviewer network; and 4, obtaining the difference between the position between the target spacecraft and the chasing spacecraft and the expected position, and the speed difference between the target spacecraft and the chasing spacecraft, inputting the difference into the trained actor network, and outputting the action of the current time step by the trained actor network.
Owner:HARBIN INST OF TECH

RGBT target tracking method based on modal perception feature learning

An RGBT target tracking method based on modal perception feature learning is characterized in that a target tracking model is constructed for an RGBT target tracking task, and the target tracking model comprises a feature extraction module, a modal perception module, a Transform fusion module, a three-branch prediction head and a template updating module; a transfer learning strategy is adopted for training. According to the invention, a modal sensing module combines an aggregation and distribution mechanism of channel features and a similarity sensing mechanism of spatial features to fully mine multi-modal discrimination features; the Transform fusion module is used for fusing features of a template and a search area, and effectively capturing a global dependency relationship by adopting mixed attention so as to enhance instance representation; the three-branch prediction head is combined with a mutual constraint loss function to improve the target positioning precision; the template updating module is used for optimizing a target template, the influence of challenge factors such as scale change, background interference and partial shielding is relieved, and the stability of target tracking is further improved.
Owner:NANJING UNIV

Fuel cell throttle valve control and simulation method based on fuzzy LADRC

The invention discloses a fuel cell throttle valve control and simulation method based on fuzzy LADRC, and relates to the technical field of hydrogen fuel cells, and the method comprises the steps: firstly building an empirical model of the air pressure and backpressure throttle valve opening of a hydrogen fuel cell air system; secondly, taking the air target pressure of the air system of the hydrogen fuel cell as program input; secondly, designing LADRC to ensure good working performance of the system; and finally, designing fuzzy logic, combining the fuzzy logic with the LSEF in the LADRC, and improving the tracking performance and rapidity of the opening degree of the backpressure throttle valve by using a fuzzy LADRC controller. The demand of adaptively adjusting the LSEF parameters is realized through fuzzy control, the difficulty of parameter setting is reduced, and the adaptability and the control effect of the system are also improved; through an interpolation table generated through fuzzy control, the model structure is simplified, and the operation speed is increased; through a fuzzy LADRC controller and an empirical model of air pressure and backpressure throttle opening, controller parameters are adjusted through simulation analysis, the test time is shortened, and the research and development cost of a hydrogen fuel cell system is reduced.
Owner:SHENYANG AEROSPACE MITSUBISHI AUTOMOBILE ENGINE MFG CO LTD

Automatic driving path planning and tracking control method

The invention discloses an automatic driving path planning and tracking control method, which belongs to the technical field of automatic driving and comprises the following steps of: 1, generating a global path and a tracking path according to a current position and a planned target point in a vehicle driving process; 2, constructing a local cost map by adopting a multi-sensor fusion method, carrying out local path planning, and generating a vehicle real-time obstacle avoidance path; 3, evaluating and selecting a tracking path, an obstacle avoidance path and a lane keeping and changing path, and deciding an optimal local path; 4, performing transverse / longitudinal control on the vehicle by using a path tracking transverse / longitudinal control system; the method has the advantages that the tracking path conforming to the kinematics law can be planned for the vehicle, the vehicle can autonomously run according to the set speed and path according to the expected path output by the planning module and under the cooperation of the integrated navigation system, and the problems that in the prior art, single vehicle path planning and tracking control cannot be achieved are solved.
Owner:ROCKET FORCE UNIV OF ENG

Sensorless model prediction speed control method for induction motor

The invention discloses a sensorless model prediction speed control method for an induction motor, and the method comprises the steps: carrying out the modeling of a mathematical model of the induction motor based on a two-phase static coordinate system, and obtaining a mathematical model of the induction motor; constructing a full-order observer, and obtaining an estimated value of the rotating speed of the rotor of the induction motor and an estimated value of the flux linkage through the full-order observer; parameter identification of stator resistance is introduced into a motor rotating speed estimation and model prediction module; establishing a prediction model, predicting the stator flux linkage and the rotating speed at the k + 1 moment, and obtaining a stator flux linkage prediction value and a rotating speed prediction value; and finally, constructing a reduced order load torque observer, carrying out feedforward compensation on a speed outer ring of the control system, and improving the dynamic response of the system. According to the invention, the problem in the prior art that the system performance is reduced and even the system crashes due to sensor faults is solved.
Owner:XIAN UNIV OF TECH

Multi-target tracking method and device for complex scene

The invention provides a multi-target tracking method and device for a complex scene, and the method comprises the steps: carrying out the target detection of a to-be-processed video frame through a YOLOX detector, and dividing a detection frame into a high-confidence detection frame and a low-confidence detection frame according to a set confidence threshold value; performing state division on the tracking trajectory of the previous video frame; dividing the target trajectory into a confirmed state trajectory and an unconfirmed state trajectory based on the state of the tracking trajectory; predicting a target tracking frame of the target trajectory in the current video frame through a Kalman filter; constructing a target matrix according to the high-confidence detection frame, the low-confidence detection frame, the confirmed state track, the unconfirmed state track and the target tracking frame; matching the target track according to the target matrix; and repeating the above steps on a to-be-processed video frame sequence to obtain a track set comprising a plurality of target tracks. According to the method, the problems of identity switching, wrong matching and target loss of multi-target tracking in a complex scene are reduced.
Owner:QINGYUAN HUAYUN ZHIKONG TECH CO LTD +1

Online estimation method and system suitable for cluster type intelligent electric meter error analysis

The invention discloses an on-line estimation method and system suitable for cluster type intelligent electric meter error analysis, and the method comprises the steps: building an intelligent electric meter error on-line calculation model through an energy conservation principle based on the characteristic that most intelligent electric meters are distributed according to a tree topology structure in reality, and solving the model to obtain an error estimation value. On this basis, non-uniformity of error information contained in measurement data in time is considered, a weight coefficient represented by attenuation factors is introduced to construct a loss function of a calculation model, different attenuation factors are adopted according to the condition number of a coefficient matrix to improve the tracking capability of an algorithm, an error estimation value is obtained by deriving the loss function, and the tracking precision of the algorithm is improved. And the error estimation value of the electric energy meter is updated in real time through newly collected data, so that the solving precision of the error estimation value is improved.
Owner:STATE GRID SICHUAN ELECTRIC POWER CO MARKETING SERVICE CENT

Scale-adaptive active visual tracking method

The invention discloses a scale-adaptive active visual tracking method, and aims to enhance the tracking capability of an active tracker under the condition of target scale change and improve the tracking performance. According to the technical scheme, a scale-adaptive active visual tracking system composed of a data acquisition module, a data preprocessing module, an active visual tracking teacher module and an active visual tracking student module is constructed; and training the active visual tracking teacher module by adopting a reinforcement learning mode, training the active visual tracking student module by adopting a supervised learning mode, and obtaining the trained scale adaptive active visual tracking system. And the trained scale adaptive active visual tracking system tracks a target. According to the method, the conventional-scale training, the small-scale training and the large-scale training are respectively carried out in the training, so that the multi-scale features of the visual observation image can be extracted, the multi-scale features are adaptively fused, and the tracking performance of a tracker under the condition that the target scale changes and interferents exist in the environment is enhanced.
Owner:NAT UNIV OF DEFENSE TECH

ViT-YOLOX-Dynamic Head-based cattle face image dynamic detection and tracking method

The invention relates to a cattle face image dynamic detection and tracking method based on ViT-YOLOX-Dynamic Head, and the method comprises the steps: obtaining a cattle herd face video, and carrying out the data preprocessing; constructing a cattle face image dynamic detection and tracking model based on a YOLOX model; performing training to obtain a trained cattle face image dynamic detection and tracking model; and obtaining a to-be-detected cattle face image, performing preprocessing, inputting the preprocessed to-be-detected cattle face image into the trained cattle face image dynamic detection and tracking model, and obtaining a cattle face image detection result. According to the method, the capability of capturing context information of the model is enhanced, the parameter efficiency of the model is fully utilized, global information in the image can be better captured, and the method has higher recognition capability for cow face targets with complex backgrounds and different scales; a dynamic head method is adopted to perform multi-scale feature fusion and adaptively adjust feature levels by associating spatio-temporal features and facial features, so that the cow face detection and tracking capability of the model in different scales and complex backgrounds is improved.
Owner:ANHUI UNIV

Unmanned aerial vehicle ground multi-target tracking method and system

The invention discloses an unmanned aerial vehicle over-the-ground multi-target tracking method, and the method comprises the steps: carrying out the initialization processing of a video image shot in the inspection process of an unmanned aerial vehicle, and reading a first frame image in the video image; extracting optical flow points in the first frame of image, and meanwhile, endowing each optical flow point with different PIDs (Proportion Integration Differentiation); tracking the optical flow points, setting the same PID for the same optical flow points, deleting the optical flow points which are not tracked in the current frame, and adding new optical flow points according to judgment; using a target detection algorithm to detect all supported targets in the current video frame, including the pixel positions, sizes and categories of the targets; a sort algorithm architecture is adopted, and the motion of a target in a picture is divided into decoupling of aircraft motion and target motion and decoupling of a static target and moving target tracking strategy; and counting the targets of the same TID, the categories of all detection frames and static and dynamic types, and updating the category and static and dynamic identifiers of the target by using the target with the highest occurrence frequency.
Owner:XIAN INNO AVIATION TECH CO LTD

Robust visual inertial odometer method and device in dynamic scene

The invention discloses a robust visual inertial odometer method and device in a dynamic scene, and belongs to the technical field of visual inertial odometers. The method comprises the following steps: step 1, acquiring and preprocessing a grayscale image, a depth image and IMU (Inertial Measurement Unit) information to acquire depth information; 2, performing feature rough matching based on depth information and inter-frame IMU integral, and constructing a corresponding descriptor and a matching function by using gray information and depth information of feature points; step 3, carrying out shake state detection by using IMU information, after a shake state is detected, carrying out re-extraction on feature points, and carrying out uniform distribution on the re-extracted feature points according to a gridding method; and step 4, constructing an inverse depth residual by using the re-extracted feature points and the depth information obtained by triangularization, and designing an optimization function. According to the invention, stable positioning in a complex dynamic scene can be realized by using the depth camera under an indoor condition, and the robustness of positioning is greatly improved.
Owner:AEROSPACE INFORMATION RES INST CAS

Learning effect evaluation system and method for artificial intelligence image generation training course

The invention discloses a learning effect evaluation system and method for artificial intelligence image generation training courses, and relates to the technical field of computers. In order to solve the technical problem that in the prior art, a learning management system cannot feed back learning conditions of students in time, the technical scheme provided by the invention is that the system comprises a plurality of modules used for collecting and analyzing data in the learning process of the students, and comprises a student single course homework quality evaluation module, the quality evaluation module is used for performing quality evaluation on a single homework of a student in an artificial intelligence image generation course to obtain a quality score; the overall learning ability evaluation module is used for carrying out comprehensive learning state analysis on the quality scores and the homework process logs of the multiple homeworks of the students to obtain overall learning ability scores of the students; and the visual feedback module is used for generating learning diaries of the students according to the change of the homework quality and the learning ability of the students. The method is suitable for being applied to learning effect evaluation work of artificial intelligence image generation courses.
Owner:HARBIN INST OF TECH

Unmanned underwater vehicle trajectory tracking control system based on preset time and performance

The invention provides an unmanned underwater vehicle trajectory tracking control system based on preset time and performance, and aims to improve trajectory tracking precision and stability in underwater operation tasks. The method comprises the following steps: establishing an unmanned underwater vehicle mathematical model; designing a preset time preset performance function, and introducing an adjustable function changing along with time to enable the boundary function to have a smooth characteristic in a convergence process; a time-varying gain auxiliary system is adopted to process an input saturation problem; designing a disturbance observer to realize rapid attenuation of disturbance, and enhancing the robustness of the system by introducing a residual adjustment function based on a smooth nonlinear model; and finally, integrating the interference observer and an auxiliary system to deduce a trajectory tracking scheme. The result shows that the longitudinal speed fluctuation is reduced by 69.2%, and the swaying speed fluctuation is reduced by 32.7%; the north position tracking performance is improved by 30.8%, the east position tracking performance is improved by 16.7%, the speed fluctuation is effectively reduced, and the tracking performance, the system stability and the anti-interference capability are improved.
Owner:HARBIN UNIV OF SCI & TECH

KCF relocation method based on multi-core DSP fusion target detection and motion information

The invention discloses a KCF relocation method based on multi-core DSP fusion target detection and motion information. The method is used for tracking and detecting a tracking target in an image. The method comprises the following steps: S1, carrying out global detection on an initial image of an input image, marking all objects, and selecting a tracking target for directional tracking from the objects; s2, confirming the total frame number m of the image to be read after the tracking target is determined, t = 1, 2... m, and extracting the tracking target characteristics of the first frame image; s3, on the basis of the tracking target features of the first frame of image, tracking target feature acquisition and tracking target trajectory data capture are performed on the second frame of image to the mth frame of image frame by frame; and S4, arranging the trajectory data of the tracking target and outputting a trajectory diagram of the tracking target. According to the invention, motion detection of the tracking target in the image is realized.
Owner:SHANGHAI RADIO EQUIP RES INST

Precise position tracking control algorithm based on real-time iterative compensation

The invention discloses a precise position tracking control algorithm based on real-time iterative compensation, and the algorithm comprises the following steps: taking a single axis of an X-Y motion platform as a research object, and obtaining a transfer function of the X-Y motion platform through parameter identification; the position tracker adopts a discrete PID (proportion integration differentiation) feedback control algorithm to obtain a feedback control output signal; an FIR (finite impulse filter) parameterization fixed structure controller is adopted to carry out feedforward control; a disturbance observer module is designed, disturbance observer output is compensated to control input, and a low-pass filter is introduced; and performing real-time iterative compensation control based on a Kalman filter. The Kalman filter is introduced into a real-time iterative compensation strategy provided by the invention, so that the noise influence of the system can be inhibited to a certain extent, and the anti-interference capability and the tracking precision of the system are further improved; in combination with a gating recursive unit prediction result and a model state space equation recursive prediction result, a better noise processing capability is provided by estimating a system state and a noise covariance.
Owner:ZHONGKE SHANHAIWEI (HANGZHOU) SEMICONDUCTOR TECHNOLOGY CO LTD

Target tracking method for hyperspectral video

The invention discloses a target tracking method for a hyperspectral video. The method comprises the following steps: initializing weight parameters in a hyperspectral target tracking algorithm; reading an initial frame of a tracking video as an initial template of target tracking and an online updating template image, and reading position information of a target in the initial frame; reading the next frame of the tracking video and cutting the next frame as a search image for target tracking; an initial template, an online updated template image and a search image serve as input of a hyperspectral target tracking algorithm, a three-channel image is obtained through dimensionality reduction of a dimensionality reduction module, then the three-channel image is input into a feature extraction network after low-rank adaptive learning, obtained feature information is input into a target position decoder, and a target position is obtained. Outputting the predicted position of the target in the current frame and the confidence of the target, and judging whether to update the online update template image or not according to the confidence; continuously repeating each frame of the tracked video and recording the predicted position of the target in each frame of the video; the method has the advantages of high target tracking accuracy, strong extraction feature representation, low training cost and the like.
Owner:XIDIAN UNIV