93results about How to "Improve perception accuracy" patented technology

The invention provides a communication and sensing integration method and system, electronic equipment and a storage medium. The method comprises the following steps: under the condition that a target object needs to be sensed, detecting the target object based on a passive detection technology to obtain a detection distance of the target object; and determining a target signal transmission mode adapted to the detection distance for sensing. And according to the target signal transmission mode, performing communication processing based on the target antenna in the communication time slot, and performing sensing processing based on the target antenna in the sensing time slot. Based on the scheme of the application, communication and sensing functions can be realized only by deploying a set of target antennas, so that spectrum and hardware resources are saved. In addition, in the sensing process, the signal transmission mode is adapted according to the detection distance of the target object determined by the passive detection technology, and the sensing precision for the target object can be improved. Meanwhile, the passive detection technology can assist in locking the target object in the sensing process, and the blind area possibly caused by switching the signal transmission mode of the target antenna is made up.

The invention discloses a visual inspection information acquisition device and method based on a horizontal included angle calibration image, and belongs to the technical field of machine visual inspection, the device comprises an industrial camera with a zoom lens, a holder device, a fixing device, a laser ranging sensor assembly, a signal acquisition synchronizer and an image automatic calibration module; wherein the industrial camera is used for capturing a surface area image of a measured structure; the laser ranging sensor assembly is used for determining a horizontal included angle formed between the camera lens axis and the surface normal of the measured structure; the signal acquisition synchronizer is used for ensuring data acquisition consistency between the camera and the laser distance measuring sensor assembly; and the image automatic calibration module is used for correcting distortion of the image captured by the camera due to the horizontal included angle to obtain a corrected image. By adopting the technical scheme of the invention, the accuracy of the image data can be ensured, so that the health detection and monitoring efficiency and precision of the engineering structure are improved.

The invention belongs to but not limited to the technical field of unmanned aerial vehicle communication, and discloses a cooperative detection method and system in an unmanned aerial vehicle communication and sensing integrated network, an unmanned aerial vehicle communication and sensing integrated system is constructed, and a base station and a detection unmanned aerial vehicle jointly execute a detection task; in a cooperative detection scene, establishing an unmanned aerial vehicle communication and sensing signal transmitting and receiving signal model; according to a maximum likelihood criterion, sensing data fusion is carried out on the detection sets of the base station and the unmanned aerial vehicle by adopting an EKF algorithm; mathematical modeling is carried out on the maximization problem of the reachable communication rate; the communication performance is enhanced by jointly optimizing the trajectory of the unmanned aerial vehicle and a beam forming algorithm, and the optimization problem of maximizing the communication rate is solved. According to the invention, the problems of high maneuverability, miniaturization and high possibility of being blocked by buildings of the black flight unmanned aerial vehicle can be solved, the sensing accuracy is effectively improved by using communication fusion sensing information, the cooperative sensing distance is increased, and the detection performance is enhanced.

The invention discloses a visual guide type tracheal intubation mirror system, relates to the technical field of guide type tracheal intubation mirrors, and aims to solve the problem of poor treatment effect of tracheal intubation. The constructed environment model not only comprises the diameter and the bending degree of the trachea, but also considers the surrounding tissue structure, so that a doctor can more comprehensively know the trachea and the surrounding environment of the trachea so as to make a more accurate decision, and the depth and angle information in the video image data is combined, so that the visual effect is improved. The real-time three-dimensional attitude of the intubation lens is estimated, the perception precision of the position and attitude of the intubation lens is improved, misoperation can be avoided, the structured light scanning method is combined with depth information and image coordinates for reconstruction, and compared with a traditional method, a three-dimensional model can be generated more quickly, the data quality of video data can be monitored in real time, and the real-time three-dimensional attitude of the intubation lens can be monitored more accurately. And the accuracy and stability of data are ensured. Therefore, data quality problems can be found and solved in time, and errors in subsequent diagnosis or treatment are avoided.

The invention provides a calf feeding robot with an autonomous navigation control system and a use method, and relates to the crossing field of robot technology and animal husbandry. A calf feeding robot with an autonomous navigation control system comprises a movable vehicle body chassis, a control screen, a main control bin, a first camera, a second camera, a laser radar base, a laser radar, a milk barrel, an electromagnetic valve, a nipple, a CAN communication interface module and a control panel. An autonomous navigation control system is arranged in the control panel and comprises an environment sensing subsystem, a navigation control subsystem, a milk feeding subsystem and a vehicle body chassis control subsystem, a highly-integrated control framework is achieved, and the integrated control system improves the operation efficiency and the response speed of the robot. Therefore, the robot can quickly adapt to the change of pasture environment and timely and accurately provide feeding service for calves.

The invention discloses a visual inspection information acquisition device and method based on a pitch angle calibration image, and belongs to the technical field of machine visual inspection, the device comprises an industrial camera with a zoom lens, a holder device, a fixing device, an inclination angle sensor, a signal acquisition synchronizer and an image automatic calibration module; wherein the industrial camera is used for capturing an image of a surface area of a measured structure; the inclination angle sensor is used for determining a pitch angle formed between the camera lens axis and the surface normal of the measured structure; the signal acquisition synchronizer is used for ensuring data acquisition consistency between the industrial camera and the inclination angle sensor; and the image automatic calibration module is used for correcting distortion of the image captured by the camera due to the pitch angle to obtain a corrected image. By adopting the technical scheme of the invention, the accuracy of the image data can be ensured, so that the health detection and monitoring efficiency and precision of the engineering structure can be improved.

The invention provides a phytoplankton chromatography sequence identification method and a phytoplankton chromatography sequence model building method, and belongs to the technical field of image enhancement identification. The method comprises the following steps: firstly, acquiring microscopic chromatography sequence data of phytoplankton, performing view field extraction and serialization recombination, and constructing a three-dimensional data set; then, constructing a three-dimensional recognition model containing physical perception and a sequence aggregation mechanism, extracting single-frame semantic features by the model by adopting a parameter-shared twin network, and introducing a physical definition prior module to calculate a space-frequency domain quality score of a slice; secondly, designing a deep perception sequence aggregation module, and adaptively aggregating key features of a high signal-to-noise ratio by taking definition scores as gating signals and combining spatial context information between slices; and finally, training and optimizing the model based on the image-level weak supervision label to obtain an optimal model. According to the method, the problems of information truncation and out-of-focus noise interference caused by extremely shallow depth of field of high-power microscopic imaging are solved, and full-depth-of-field stereoscopic perception can be realized under the condition that frame-by-frame fine labeling is not needed.

The invention provides a communication and inductance integrated waveform generation, sensing and communication processing method and a communication system, and belongs to the technical field of radar detection. At least one communication sub-signal and at least one sensing sub-carrier are acquired; performing direct-current bias processing on each communication sub-signal, and coding each sensing sub-carrier based on the biased communication sub-signal to obtain a single-carrier flux-sensing integrated waveform; and performing frequency conversion processing on each single-carrier flux-inductance integrated waveform, and based on the single-carrier flux-inductance integrated waveforms subjected to frequency division multiplexing combination frequency conversion, overlapping the frequency spectrums of the adjacent single-carrier flux-inductance integrated waveforms to obtain a multi-subcarrier flux-inductance integrated waveform. Direct-current bias processing is carried out on communication sub-signals, unique phases of sensing sub-signals can be reserved, communication and sensing can share the same time-frequency resource and software and hardware equipment conveniently, high communication rate and high sensing precision are achieved, frequency domain resources can be effectively saved through frequency conversion processing and frequency spectrum overlapping of multiple sets of single-carrier-wave-flux-sensing integrated waveforms, and the communication efficiency is improved. And spectrum efficiency is improved.

The present application relates to the field of data processing technology, and discloses a method, system and storage medium for controlling a robotic arm based on multimodal drive. The method comprises: collecting data through a sensor module for preprocessing to obtain a multimodal data set; extracting features from the multimodal data set and performing feature fusion to obtain a fused feature vector; estimating the state data of the fused feature vector through an unscented Kalman filter, and performing state prediction through a recursive neural network to obtain the current state of the robotic arm and the predicted state of the robotic arm; performing robotic arm operation trajectory planning and analysis on the current state of the robotic arm and the predicted state of the robotic arm to obtain the target execution trajectory; generating a control strategy for the target execution trajectory through an event-triggered distributed coordinated control algorithm to obtain a real-time joint control instruction sequence, and controlling the robotic arm to perform operations through the real-time joint control instruction sequence. The present application improves the efficiency and accuracy of robotic arm control based on multimodal drive.

The invention discloses a fusion and decision-making method based on multi-node spectrum data. The method comprises the following steps: arranging a plurality of nodes in an application area, constructing a receiving signal spectrum sensing model of the nodes, constructing a receiving spectrum matrix, calculating the power of each frequency value in the receiving spectrum matrix, and constructing a receiving power matrix of the nodes by utilizing the compensated power; determining a mean value of each node in a frequency band based on the receiving power matrix of each node, and determining a threshold value of each node by using a variance based on the mean value of each node, thereby setting a weight for the receiving power matrix of each node; setting parameters required by spectrum data fusion of multiple nodes; the multi-node spectrum data is fused, the spectrum data in a certain frequency band and a certain time range is compared with a threshold value, then the idle proportion in each frequency band range is calculated, and the frequency band with the maximum idle proportion serves as the optimal frequency band to be output. The system has the characteristics of high data acquisition efficiency, high reliability and wide application range.

The invention provides a safety protection method, device and equipment for a bridge inspection vehicle and a storage medium, and relates to the technical field of bridge detection.The safety protection method for the bridge inspection vehicle comprises the steps that environment point cloud data and environment parameter data around the bridge inspection vehicle are obtained in real time, and time and space synchronization processing is carried out; the environment point cloud data and the environment parameter data after time-space synchronization processing are preprocessed, the environment point cloud data are analyzed through a clustering algorithm, obstacle feature information on the advancing path of the bridge inspection vehicle is recognized, the environment parameter data are analyzed, and environment condition information of the bridge inspection vehicle on the advancing path is determined; and combining the detected obstacle feature information with the environmental condition analysis information, carrying out comprehensive risk assessment, and adjusting the inspection path and operation parameters of the bridge inspection vehicle in real time according to a comprehensive assessment result. The safety and inspection efficiency of the bridge inspection vehicle in a complex environment can be improved.

The invention discloses an unmanned aerial vehicle road intelligent cruise system based on internal signals, and relates to the field of unmanned aerial vehicle cruise. When an unmanned aerial vehicle performs road inspection, data information around an inspection road is obtained by using a laser radar sensor and a camera, and the obtained two different data are fused; more comprehensive and more accurate perception of the environment is realized so as to improve the environment perception precision of a subsequent built-in map chip, the current position of the unmanned aerial vehicle on the inspection road is acquired by using a scene matching mode, and during path planning, an unmanned aerial vehicle autonomous flight system masters regional map information, so that the path planning efficiency is improved. In this way, inspection can be intelligently completed without external signals, the problem that the unmanned aerial vehicle is difficult to complete road inspection work due to the fact that the unmanned aerial vehicle gets lost in a district with weak satellite signals and dark light can be avoided, the unmanned aerial vehicle inspection can achieve a better effect, and the practicability of the unmanned aerial vehicle is further improved.

The embodiment of the invention provides a signal sending method and device, a signal receiving method and device, equipment and a storage medium. The signal sending method comprises the following steps: sending a sensing signal on a flexible time slot or an uplink time slot or a downlink time slot; wherein the sensing signal adopts an extended cyclic prefix, or the time length of the sensing signal is greater than the length of an orthogonal frequency division multiplexing (OFDM) symbol.

The invention discloses a hybrid drive power-assisted exoskeleton device and method, and relates to the technical field of power-assisted robots, the hybrid drive power-assisted exoskeleton device comprises hip joint modules and a sensor assembly, the hip joint modules are rotatably connected to the two sides of a back body, the hip joint modules are connected with pneumatic artificial muscles and joint motors, and the pneumatic artificial muscles and the joint motors are connected with a control assembly; the sensor assembly comprises a pressure sensor, a tilt angle sensor and a gyroscope, the pressure sensor is used for acquiring the state of the pneumatic artificial muscle, and the tilt angle sensor and the gyroscope are used for detecting the tilt angle, so that the control assembly controls the pneumatic artificial muscle or the joint motor to work according to the tilt angle of the human body. Pneumatic driving is adopted when objects are taken out and put down, motor driving is adopted in the walking and object carrying moving process, the object carrying efficiency is improved, and the service life of the device is prolonged.

The invention discloses a multi-sensor fusion sensing method based on a high-precision map. The method comprises the following steps: sensing and acquiring information of static obstacles based on the high-precision map; performing original data fusion of dynamic obstacle information based on multi-sensor fusion, and summarizing data of all sensors to obtain a super sensor; obtaining the symmetry of the geometric structure of the multi-dimensional body according to the obtained super sensor, and extracting data information of a target obstacle based on edge detection and an obstacle IOU contour; and performing sensing fusion on a high-precision map based on stereo matching and the sensors, and performing shielding detection on inconsistent areas by adopting multi-source consistency check to complete final fusion of the multiple sensors. According to the invention, the possible road condition in front of the current position can be known in advance, and the function of the sensor module is dynamically adjusted, so that the performance requirement on the sensor is reduced, and vehicle-mounted computing resources are reasonably allocated.

The present application relates to the field of wisdom agriculture and image recognition technology, specifically to a crop disease and pest monitoring system based on image recognition, comprising: an environment perception module, which obtains the environmental context parameters of the monitoring area through multi-modal sensor nodes; a multi-source context analysis module, which constructs a disease and pest outbreak risk model in combination with historical disease and pest distribution spatio-temporal data and calculates the disease and pest risk level; an adaptive strategy generation module, which dynamically reconstructs the image acquisition mode and feature extraction strategy according to the risk level and the edge side resource constraint state, and generates adaptive perception instructions; an image acquisition execution module, which adjusts the physical acquisition parameters and algorithm calculation density of the visual sensor in response to the instructions, and executes image acquisition and recognition; the present application breaks the traditional rigid acquisition mode and realizes the optimal balance between monitoring efficiency and device survival period.

The invention provides a vehicle-road cooperative sensing method based on deep reinforcement learning, and the method comprises the steps: obtaining the positions of all CAVs through a road side unit (RSU) through a cooperative object selection method based on the RSU, distributing an optimal cooperative object for each CAV through employing a DQN network, and enabling the CAVs to achieve the cooperative object after obtaining an object needing to be cooperated, and selecting an optimal three-dimensional detection frame and a corresponding confidence score by using a DQN network based on invalid action shielding, and sending the optimal three-dimensional detection frame and the corresponding confidence score to a collaborative object to realize vehicle-road collaborative perception.

The invention discloses an obstacle sensing method, a cleaning robot, a storage medium and a program product, and the obstacle sensing method comprises the steps: obtaining sensing data of a sensing sensor of the cleaning robot; in response to the situation that the cleaning robot is detected to be in a diving state according to the sensing data, a distance measuring sensor of the cleaning robot is controlled to emit first emitting light of a first wavelength, and the diving state is used for representing that the cleaning robot is immersed in water; the distance measuring sensor is controlled to receive first reflected light, the first reflected light is formed by reflecting the first emitted light through a first obstacle, and the first obstacle is immersed in water; and determining a first distance between the first obstacle and the cleaning robot according to the first emitted light and the first reflected light. According to the method, when it is detected that the cleaning robot dives into the water, the first distance of the underwater first obstacle is sensed based on the light of the first wavelength emitted by the distance measuring sensor, and the underwater driving risk of the cleaning robot can be reduced.

The present invention relates to the field of process control technology, and specifically to an automatic driving method and system based on multi-sensor fusion deployment. The data to be fused and the driving behavior time series data of multiple sensors in the process of vehicle automatic driving are obtained, and the synchronous processing and real-time control of the data are realized by using a timer. In the early stage of automatic driving, the system quickly adjusts the preset window length according to the data fluctuation of the data to be fused to adapt to environmental changes; when the system is stable, the window length is comprehensively adjusted in combination with the similarity of environmental changes and driving behavior data to enhance the ability to respond to complex environments and emergencies. The method dynamically adjusts the window length and fuses and analyzes the data, improves the accuracy of data fusion and the perception accuracy of automatic driving, and provides a reliable basis for vehicle automatic control strategies.

The present invention relates to the field of industrial vehicle control technology, and specifically to an obstacle perception method of an industrial vehicle ADAS fused with a large model, comprising: when an obstacle is detected within the radar range of the industrial vehicle, obtaining the grayscale image of the obstacle and the height data and position data of the obstacle at each moment within a preset time period; obtaining a collision risk index based on the object contour distribution in the obstacle grayscale image at the initial moment within the preset time period, combined with the object attributes and height data of the obstacle at the initial moment; obtaining the degree of perceived influence of the obstacle on the industrial vehicle based on the difference between the position data of the obstacle at every two moments and the change between the height data at adjacent moments; dividing the obstacle warning level of the industrial vehicle according to the collision risk index and the degree of perceived influence, and obtaining the level perception result of the obstacle. The present invention effectively improves the reliability of the industrial vehicle ADAS system fused with a large model and the perception accuracy of the obstacle.

The application discloses a kind of pathological image classification methods based on dyeing perception intelligent feature modeling, belong to pathological image classification field, this method includes collecting the multi-staining full section image group data of different objects;Multi-staining full section image group data includes H&E dyeing image and several immunohistochemical dyeing images;Each multi-staining full section image group data is respectively preprocessed, obtains the several equal-size image blocks of each object;Based on the image block of each object, graph is constructed, and the multi-modal heterogeneous pathology graph of each object is obtained;With the multi-modal heterogeneous pathology graph of each object as training data, train pathological image classification model;The pathological image classification model of training completion is used to classify the pathological image to be measured.The application solves the problems of the existing method in processing digital pathology graph data, such as node category characteristics, edge information utilization, cross-staining interaction and non-perfect data processing.

The invention relates to a sensing method and device, a cleaning robot, a computer readable storage medium and a computer program product. The cleaning robot comprises a TOF sensor and a binocular camera, and the method comprises the steps that a first depth image of a target scene is collected through the TOF sensor, and three-primary-color images of different visual angles in the target scene are collected through the binocular camera; performing view angle alignment and pixel point alignment on the first depth image and three-primary color images of different view angles to obtain second depth images corresponding to different view angles; on the basis of a pre-trained image processing model, performing feature extraction and feature fusion on the second depth image and the three-primary color image of different visual angles to generate a target depth image; performing synchronous positioning and map construction based on the target depth image and the three-primary color image of any view angle to obtain a three-dimensional map of the target scene; the three-dimensional map is used for navigation perception of the cleaning robot. By adopting the method, the sensing accuracy of the cleaning robot to the target scene is improved.

The invention provides a multi-dimensional fish school sensing device, which can be applied to the technical field of fish school sensing, and comprises a fixed workbench, a plurality of pan-tilt cameras and an omnidirectional scanning sonar, the omnidirectional scanning sonar is located at the central position of the fixed workbench, and the omnidirectional scanning sonar is used for obtaining point cloud information in a 360-degree range around the device; the plurality of pan-tilt cameras are uniformly distributed on the fixed workbench, and the distances between the plurality of pan-tilt cameras and the central position are equal; the pan-tilt camera can rotate in a single degree of freedom in the horizontal direction; and the pan-tilt camera can actively rotate according to the distribution condition of the optical image feature points in the environment. Based on image data and sonar point cloud data acquired by the device, fusion of optical and sonar features and three-dimensional target detection of fish schools are completed. The invention further provides a multi-dimensional fish school sensing method.

The invention provides vehicle glass, a vehicle glass assembly and a vehicle, the vehicle glass comprises laminated glass, a first bonding layer and a first glass plate, the first glass plate is connected with the laminated glass through the first bonding layer or arranged in the laminated glass through the first bonding layer, and the first bonding layer is arranged between the first glass plate and the laminated glass. The first glass plate comprises a first signal transmission area and a first shielding area, and the first shielding area is arranged around the first signal transmission area. Therefore, the problems of local optical deformation and distortion of the glass are reduced, and the diopter performance of the vehicle glass in a sensor area is remarkably improved.

The invention provides a sensing signal transmission method and device, a sensing node and a storage medium. The method comprises the following steps: sending a sensing signal, and obtaining feedback information aiming at the sensing signal; determining environmental parameters of the to-be-measured area according to the feedback information; wherein the to-be-measured area is an area between a sending node and a receiving node of the sensing signal. According to the method, the sensing of the surrounding environment information is realized by sending the sensing signal.

The invention relates to the technical field of automatic driving, in particular to a multi-source time domain synchronization method and device in vehicle infrastructure collaborative awareness. The method comprises the following steps: the RSU periodically broadcasts configuration information of sensing equipment to vehicles in a coverage range; the vehicle receives the configuration information of the RSU after entering the coverage area, and calculates the time offset of the RSU sensing equipment according to the received configuration information; after the vehicle receives the sensing data of the RSU sensing equipment, the sensing data are cached, and the size of the cached data is predefined according to the fusion sensing capability of the vehicle; and the vehicle extracts self-sensing data and sensing data frames of the RSU sensing equipment according to a shortest time delay criterion, and fusion sensing is carried out. According to the method, a multi-source time domain synchronization scheme is provided, the amount of time needed to be shifted when different RSUs reach data is calculated, the amount of time is compensated, then similar data frames in the time domain are extracted for fusion, and the sensing precision is improved.

The application provides a multi-modal sensor time synchronization method for road monitoring, comprising: acquiring sensor data for target road detection, and determining a communication time delay; according to actual network conditions and environmental changes, optimizing a compensation value in real time to obtain an optimized data set; based on the complete optimized data set, determining a network topology structure; according to the optimized network topology structure, determining a time synchronization signal sequence of relative time delay information of each node in the network; and optimizing time synchronization of all sensors for target road detection. The application introduces a dynamic environment modeling and adaptive network topology optimization strategy, effectively overcomes the shortcomings of traditional time synchronization technology, and can realize high-precision multi-modal data time synchronization in a complex dynamic environment.

The invention provides a method for identifying fires in electric equipment based on a multi-parameter recursive graph, and relates to the technical field of electrical fire detection. The method comprises: using a smoothing filtering technology to pre-process monitoring video images, and creating a fire identification algorithm model based on artificial intelligence to detect and identify fire images in suspicious areas; the fire identification algorithm model identifies the hidden danger electrical monitoring data of the detection position, calculates the target feature vector corresponding to the trace features of the target video image, and builds a deep neural network sub-model and a power heterogeneous multi-parameter fusion model; then using a recursive algorithm module to arrange the power time series parameters into a recursive matrix, and constructing a graph convolutional neural network based on a multi-parameter recursive graph to extract features; using a fire warning judgment module to segment the suspicious unknown areas of the image sequence where the fire occurs, and timely predict and process them, and after identification and confirmation, automatically start a multi-parameter warning judgment program, and accelerate the issuance of warning signals by a background warning system.

The application provides a monocular three-dimensional target perception method, system and device based on instances, the method comprises the following steps: acquiring multi-view images collected by a vehicle-mounted camera; inputting the multi-view images into a monocular three-dimensional target perception model to obtain instance perception results; the monocular three-dimensional target perception model is used for feature extraction on the multi-view images, depth information enhancement on multi-scale feature maps, cross-attention operation on a pre-constructed query vector and a depth perception feature map, depth estimation on instance query features based on a pre-constructed reliability evaluation model, feature calibration on the instance query features based on a depth reliability weight factor, and decoding on the calibrated instance query features; the application can evaluate the depth reliability of each instance query feature and differentially calibrate the corresponding instance query features based on each depth reliability evaluation, thereby effectively improving the perception accuracy of targets in scenes such as occlusion and scale variation.

This invention discloses a multi-dimensional privacy information perception and collection method. It constructs a formal description model of multi-dimensional privacy information, characterizing privacy information from multiple perspectives, including its composition, attributes, permissions, and control. Based on active learning and transfer learning algorithms, it builds a metadata intelligent perception, identification, and labeling method adaptable to various application scenarios and privacy protection strategies. This patent's advantages in adapting to multiple scenarios and information types allow it to adapt to the impact of changing application scenarios on privacy properties, reduce the concealment of privacy information caused by multiple information types, and improve the accuracy of privacy information perception. It focuses on perceiving multiple types of privacy information, comprehensively perceiving different privacy types such as text, images, and sound. Compared to existing single-type privacy perception methods, it is more beneficial for subsequent privacy strategy formulation and privacy control implementation, improving user privacy security. It has strong privacy perception capabilities, can process different types of data in parallel, and offers accurate privacy positioning and fast data processing.