104 results about "Visual computing" patented technology

The invention relates to a visual computing and prewarning method of deviation angle and distance of an automobile from a lane line. The image processing and computer vision technologies are utilized, and the deviation angle and distance of the automobile from the lane line are computed in real time according to the road surface image acquired by a vehicle-mounted camera, thereby estimating the line crossing time for safety prewarning. The method comprises the following steps: detecting the lane lines of the road surface image to obtain a linear equation of partial lane lines; establishing a three-dimensional coordinate system by using the camera as the initial point, and recording the mounting height and depression angle of the camera; calibrating the focal length according to the lane detection result under the condition of a given deflection angle; computing the deflection angle and vertical distance of the automobile relative to the lane line according to a pinhole camera model; and estimating the deviation time from the lane according to the instantaneous running speed of the automobile, thereby obtaining the safety prewarning or intelligent control information of the running automobile.

The invention discloses a visual computing-based optical field imaging device and method. The visual computing-based optical field imaging device comprises an optical imaging system, an integrated near-infrared light source, a data transmission module, a high-capacity cache unit, a human-machine interaction device, a logic control unit and a processing unit. The invention provides an optical field imaging device based on a computing and imaging interactive mutual-beneficial mode; and with optical field imaging as the leading and deep perception and high-dynamic range imaging as the assistance, visual perception intelligent and visual cognition precision are realized, novel visual information with characteristics of large depth-of-field, wide dynamic range, actual three dimension, high resolution and capability of collecting and editing is obtained, and accuracy and robustness of a visual computing task are improved through data innovation of a visual perception source; and the optical field imaging device is widely applied to the fields of internet of things, security monitoring and control, biological recognition, intelligent transportation system, aerial remote sensing, digital medical treatment, animation media and the like.

The 3D active real-time positioning system based on binocular vision and laser ranging consists of three parts of vision processing subsystem, control subsystem and mechanical mechanism. The target image the video camera under the control of image acquisition card obtains is processed in the vision processing computer to obtain the 2D position information in image coordinate sent to the public memory area. Then, the control computer obtains the position information from the public memory area, performs coordinate conversion and real-time programming, outputs via D / A converter board to the driver to drive the motor and controls the laser via the angle encoder feedback loop to irradiate the target accurately. Finally, the depth information the range finder measures is fed back to the vision computer to obtain the 3D position information of the target. The system has the advantages of high distance precision, high activity, high real-time performance, low cost, etc.

The invention discloses a wearable three-dimensional gesture interaction system and a suing method thereof. The system consists of a micro projector, a common camera, a portable computer, and a group of simple and easy marks. The function of human-computer interaction by using gestures and a three-dimensional graphical interface is realized through simple and easy computer vision computation. In the wearable three-dimensional gesture interaction system, a three-dimensional graphical interface is more vivid and the interaction mode is more flexible through a projection screen marked three-dimensional graphical interface mode and a virtual mouse gesture. The wearable three-dimensional gesture interaction system has simple equipment and low cost and can meet the requirements on mobile equipment.

The present invention relates to machine vision computing environments, and more specifically relates to a system and method for selectively accelerating the execution of image processing applications using a hybrid computing system. To this extent, a hybrid system is generally defined as one that is multi-platform, and potentially distributed via a network or other connection. The invention provides a machine vision system and method for executing image processing applications on a hybrid image processing system referred to herein as an image co-processor that comprises (among other things) a plurality of special purpose engines (SPEs) that work to process multiple images in an accelerated fashion.

The CV-CAD (computer vision-computer-aided design) model is an enhanced CAD (computer-aided design) model that integrates local and global computer vision data in order to represent an object not only geometrically but also in terms of computer vision. The CV-CAD model provides a scalable solution for intelligent and automatic object recognition, tracking and augmentation based on generic models of objects.

The invention discloses a vision computing and inertial measurement unit-based virtual reality interactive device in the technical field of virtual reality interaction. The interactive device comprises a three-dimensional display device, an industrial camera module, three-dimensional glasses, a user interactive peripheral and a computer system, wherein the industrial camera module obtains two or more image signals in a view field range and inputs the image signals to the computer system; and the computer system distinguishes and tracks mark points or active luminous components with high-reflective coatings on the three-dimensional glasses worn by a user and the user interactive peripheral, and outputs a focal point of a sight line of the user and the three-dimensional display device, and spatial position and direction of the user interactive peripheral. The interactive device can be used on desktop-level or large-screen projection and has expansibility and maintainability; and high precision and low time delay can be achieved, continuous operation can be performed, and sight line shielding does not exist.

The invention discloses a high-precision mechanical arm hand-eye camera calibration method and system, and the method comprises the steps: collecting images of a calibration plate at different positions, and measuring the internal parameters of a camera and the external parameters between a wrist camera and the calibration plate through a visual method; measuring the external parameters between a mechanical arm target and the calibration plate through a high-precision self-collimating theodolite, a mechanical arm target design and key points selected from the calibration plate; and finally completing the calibration of the mechanical arm wrist camera through a visual computing method. According to the invention, the system employs two high-precision measurement instruments, takes a calibration plate coordinate system as the central coordinate system, achieves the high-precision coordinate system parameter conversion, and finally obtains the external parameters between the tail end of the mechanical arm and the camera. The method can achieve the high-precision calibration of the external parameters between the mechanical arm and the camera.

The invention discloses a multi-network integrated unmanned aerial vehicle, an Internet-of-things monitoring system and a method, wherein the unmanned aerial vehicle comprises a multi-communication module and a visual computing module, and the multi-communication module is used for receiving monitoring data and position information from a monitoring area and transmitting the monitoring data and the position information to a cellular network or other unmanned aerial vehicles; and the visual calculation module is used for shooting a monitoring area sent by the abnormal monitoring data when the multi-communication module receives the abnormal monitoring data, and transmitting a shot image back to the multi-communication module. The Internet-of-things monitoring system comprises an air networksubsystem and a ground network subsystem, and the air network subsystem comprises at least one unmanned aerial vehicle; the ground network subsystem comprises a plurality of ground terminal nodes andat least one ground aggregation node. The system can upload the monitoring area collected in the monitoring area without cellular network coverage to the far-end server so as to realize remote monitoring of the target area.

The invention discloses an image monitoring and transmitting method for an unmanned plane. The method comprises the steps as follows: S1, a monitoring computer starts a flight control program, and a satellite navigation unit starts a GPS navigation program; S2, a high-definition action camera collects a video image according to a track of the flight control program, and a vision computer processes the image; S3, a video image wireless transmitting module and a video image receiving module combine to finish the wireless transmitting and receiving of an image signal; S4, a central site image processing module processes the received image signal and displays the image signal on a display terminal. The method of the invention supports visual navigation, image recognition and obstacle avoidance, and fuses a satellite communication network mode and a traditional communication mode so as to achieve the high-speed exchange of large-capacity image data. The method of the invention has relative high safety.

The CV-CAD (computer vision-computer-aided design) model is an enhanced CAD (computer-aided design) model that integrates local and global computer vision data in order to represent an object not only geometrically but also in terms of computer vision. The CV-CAD model provides a scalable solution for intelligent and automatic object recognition, tracking and augmentation based on generic models of objects.

The invention provides a spatial three-dimensional vision-computing verification method. In a verification experiment, a reference coordinate system is formed by two theodolites a and b, a third theodolite c and a reference coordinate only have a translational relation. A camera is fixed on the theodolite c, and the relation between a camera coordinate system and a theodolite c coordinate system is obtained. The coordinate of three target points on a spatial target object under an own object coordinate system is given, then coordinates of three characteristic points under the reference coordinate system are obtained by utilizing the theodolites a and b, and further the relation between the object coordinate system and the reference coordinate system is solved. Finally, the relation between the target object coordinate system and the camera coordinate system is obtained as a reference value of a three-dimensional position of the target object. The reference value is used for verifying the three-dimensional position of the target object solved by other algorithms. The coordinates of spatial points under the camera coordinate system are obtained by utilizing the reference values of the position of the target object, and are projected to an image plane so as to verify the parameter calibrated by a vision system in the camera.

The invention discloses an autonomous cruising system and an autonomous cruising method of a water surface robot. A user can control the water surface robot to establish a water area map through a monitoring station through wireless network communication, learn the water surface environment, and prepare for autonomous navigation of a subsequent water surface robot; an integrated navigation technology is applied to autonomous navigation of the water surface robot to realize autonomous planning of a driving path; the combination of laser scanning point cloud processing and computer vision calculation is used for autonomous searching of a floating target on the water surface, so that autonomous planning of the driving path in the operation process is realized. The disclosed autonomous cruising method for the floating target on the water surface, which utilizes the synergistic effect of a water surface target detection laser radar and a water surface target detection camera, is particularly suitable for the situation that the number and the positions of the target points are uncertain, and the inverted reflection in water and water wave interference caused by illumination can be effectively eliminated. According to the method, the autonomous operation capability of the water surface robot is enhanced, the workload of manned operation is effectively reduced, and the water surface operation efficiency is improved.

The invention discloses a dynamic human body three-dimensional reconstruction method, device and equipment and a medium, and relates to the technical fields of computer vision, computer graphics, three-dimensional reconstruction, virtual reality, augmented reality and the like. According to the specific implementation scheme, a human body three-dimensional model is reconstructed according to an RGB image and a depth image, collected based on a single view angle, of a target human body; posture estimation is performed on the target human body according to the RGB image; according to the estimated corresponding relation between the two-dimensional posture and the human body three-dimensional model and the estimated corresponding relation between the three-dimensional posture and the human body three-dimensional model, bone movement and non-rigid movement of the human body three-dimensional model are estimated, and therefore the robustness of dynamic human body three-dimensional reconstruction based on a single view angle is improved. The voxel fusion is performed on the human body three-dimensional model after movement according to the matching result of the semantic information of the target node before and after movement, so that the fusion of the wrong human body surface is avoided.

The invention discloses a visual computing-based intelligent detection method and apparatus for the appearance and performance of an electricity meter. The method comprises the steps of 1, shooting image information of the to-be-detected electricity meter by frames, obtaining a high-resolution image by adopting a line-by-line scanning method, and generating to-be-processed original image data; 2,according to target region locating, segmenting the collected image, segmenting a target region by adopting an edge-based image segmentation method, extracting single character images, and generatinga to-be-identified character set; 3, comparing the segmented image information with stored information, judging whether the appearance structure of the electricity meter is intact or not and display parameters are normal or not, and generating an identification result; and 4, according to the identification result, performing statistics on analysis data, generating a detection report and uploadingthe detection report to an upper computer. The operation is simple and convenient; the accuracy is high; and through automatic detection, the electricity meter detection efficiency is improved, and the problems of low efficiency and high error rate of conventional manual detection are solved.

The present invention relates to machine vision computing environments, and more specifically relates to a system and method for selectively accelerating the execution of image processing applications using a multi-core processor system. To this extent, a multi-core processor system is generally defined as one that is multi-platform, and potentially distributed via a network or other connection. The invention provides a machine vision system and method for executing image processing applications referred to herein as an image co-processor that comprises (among other things) a plurality of multi-core processors (MCPs) that work to process multiple images in an accelerated fashion.

The invention provides an immersive visual campus system based on a 5G network. The system comprises a 5G classroom teaching system and a 5G practical training learning system, and the 5G classroom teaching system comprises a classroom three-screen full-space system, a teacher and student control three-terminal system, a student multi-terminal experience system, and a campus cloud which achieves the visual calculation functions of cloud storage, cloud rendering and cloud positioning through 5G communication. The 5G practical training learning system comprises an experience terminal and a campus cloud, and the campus cloud comprises a cloud storage center module, a cloud visual computing module, a cloud virtual object state computing module and a cloud rendering module. The beneficial effects of the invention are that the system integrates VR, AR and MR, constructs an immersible knowledge kingdom which can be seen by human eyes in a virtuality and reality combination mode on a real campus, and provides a brand-new three-dimensional immersive work, learning and living space for teachers and students in schools.

The invention provides a method for processing data. The method comprises the steps of providing a data adaptive interface, injecting the data through the data adaptive interface, providing a visual computing component according to the injected data, and carrying out computing processing on the injected data by dragging the visual computing component. The visual computing component is provided according to the injected data, data computing, monitoring and analyzing are all carried out in a visualization mode, and the method is simple, efficient and easy to use. The invention further provides a device for processing the data.

The invention relates to the field of structural design of visual information processing chips in integrated circuits, and discloses dual-bus visual processing chip architecture. The architecture comprises a first bus, a second bus, a visual computing and deciding module connected to the first bus, a first memory connected to the first bus, a characteristic combination and mode generating module connected to the second bus, an image characteristic extraction module connected to the second bus, a second memory connected to the second bus, and a bridge circuit for connecting the first bus and the second bus.

Evaluation based on sensitivity of an image that was performed by using the sensibility and the manual work of a person is automatically performed. There are provided image-for-evaluation input means 1 for inputting an image for evaluation, induction-field-in-vision calculation means 2 for calculating the induction field in vision of the image input to the image-for-evaluation input means 1, and image-evaluation-based-on-sensitivity means 3 for performing evaluation based on sensitivity of the image based on a potential value obtained by the induction-field-in-vision calculation means. The image-evaluation-based-on-sensitivity means 3 comprises induction-field-energy calculation means 31, calculates an induction-field energy based on the result obtained by the induction-field-in-vision calculation means 2, and performs the evaluation based on sensitivity of the image according to the magnitude of the energy.

The invention provides a pipeline cleaning robot control system and method. The control system comprises a robot used for cleaning a pipeline and further comprises a collection module, a visual calculation unit and a control unit, the collection module is used for collecting the inner wall of the pipeline in real time to obtain a first image, and the visual calculation unit is used for judging whether there is a to-be-cleaned object on the inner wall of the pipeline according to the first image; if YES, the initial position of the to-be-cleaned object is calculated, a walking route is planned,and the control unit is used for controlling the robot to advance to the initial position according to the walking route; the collection module is further used for collecting the to-be-cleaned objectto obtain a second image, and the visual calculation unit is further used for calculating the accurate position of the to-be-cleaned object according to the second image and planning a cleaning route; the control unit is further used for controlling the robot to execute the cleaning according to the cleaning route. According to the pipeline cleaning robot control system and method, intelligent control over the robot is improved, the cleaning cost for the interior of the pipeline is reduced, and a good technical effect is achieved.

The invention discloses a high precision trajectory tracking and control method. The method comprises of respectively sending the position and the rudder angle of a ship to a line of sight aiming navigation algorithm and a GPC controller, visually calculating a desired heading angle, and the GPC controller tracking the heading angle to obtain a commanded steering angle. A visual navigation algorithm can imitate the visual characteristics of humans according to the setting, track the desired point according to the distance, and then obtain the desired heading angle according to its own position. The invention designs a GPC controller by utilizing the advantages of multi-step prediction, rolling optimization, and feedback correction of GPC to track the heading angle of the visual output, calculate the commanded steering angle of the ship, and realize high precision tracking of the ship's trajectory.

The invention relates to the technical field of computer vision, in particular to an embedded visual computing system for face recognition, counting and temperature measurement. After edge computing equipment is started, the system is started to run; a FaceSDK / BodySDK algorithm module is initialized; a camera stream acquisition module acquires a video stream from predefined camera information; a video stream is pushed to a video stream algorithm processing module in a queue form by a single frame and is analyzed according to a loading algorithm; and the analyzed structured result is processed according to the business requirement and then presented to the user in a visual result form. According to the invention, the defects in the prior art are overcome, and the service scenes of employee sign-in, guest greeting, frequent hall approaching and the like are solved through face comparison by setting applications of face recognition, guest greeting, frequent hall approaching and the like based on face detection and face comparison algorithm derivation; and an AIOT framework based on the intelligent edge calculation is adopted, so the purpose of deploying the central server in an internet end or a far-end machine room can be achieved.

The invention relates to a human eye-like visual tracking device and a control method thereof. The device includes a head and neck robot body which includes rotation joints of three degrees of freedomand a CCD camera and is used for adjusting the orientation of the CCD camera to directly face a tracked target through the rotation joints, a vision calculating module which is used for calculating position information of the tracked target according to image information collected by the CCD camera and returning the position information to a robot controller, and the robot controller which is used for calculating joint angles that need to be rotated for tracking the target in real time and sending the joint angles to a joint controller for tracking. The human eye-like visual tracking device and the control method thereof adopt a geometrically decomposed joint angle solving method, which is different from a commonly used virtual connecting rod method and avoids the multi-solution problem of inversion solution of a robot of redundant degrees of freedom, are simple in calculation and high in solving efficiency, are introduced with a comfort angle optimization model, and most efficientlyutilize the dynamic characteristics of the robot on the basis of guaranteeing a feasible solution.

The invention relates to an intelligent security and protection system which comprises wideband video security and protection equipment, a shell, a power interface, a duty management module, an important protection task module and a video behavior analysis module, wherein the duty management module is used for creating a duty task, executing the duty task and recording and inquiring the execution condition; the important protection task module is used for creating important protection tasks and performing route description and video grouping on the important protection task; the video behavior analysis module is used for analyzing abnormal behaviors in video images by virtue of an image analysis treatment technology, a computer vision technology, a computer network technology and an artificial intelligence technology and giving out an alarm after discovering the behavior disorder. The intelligent security and protection system can be used for presenting the current police strength state and operation locus in real time, presenting videos for automatic route description and grouping of important protection tasks; by virtue of intelligent analysis on the video images and alarming for the abnormal behaviors, beforehand and event real-time early warning service is provided for working personnel.

The invention discloses an intelligent tracking shooting system for a long-distance high-speed moving target, and belongs to the field of computer vision, computer control and television relaying. The system comprises an optical system, a stable platform system and a display control system. According to the invention, a panoramic camera is used as a visual sensor, an adaptive target detection tracking method is fused with a vehicle-mounted two-axis three-dimensional stable platform, and in the target tracking method, a global camera is combined to guide a close-up camera to track and shoot and close-up camera closed-loop detection dual mechanism, so that the tracking precision of the whole system is improved; the target is in the center of the video picture. When a plurality of paths of videos are output, the system evaluates and automatically switches the plurality of paths of video streams in real time through a real-time video stream intelligent switching method, and outputs stable video pictures in real time, so that stable following shooting and real-time editing of a high-speed moving target in a severe environment and a complex scene are realized; the good technical support can be provided for extension applications such as subsequent remote director, cloud reconstruction, personalized rebroadcasting and the like.

The invention relates to the technical field of unmanned aerial vehicles, and particularly relates to a management platform and method for taking off and landing of a rotor unmanned aerial vehicle. The flight control system comprises a flight control device of the unmanned aerial vehicle and a control system running on the device, the unmanned aerial vehicle flight control device is jointly composed of an aircraft and a take-off and landing platform, the aircraft comprises a central fuselage, two main force arms, a longitudinal rotor rod and an undercarriage, and the two main force arms are fixed to the two sides of the central fuselage correspondingly and obliquely arranged upwards; the two longitudinal rotor rods are in one-to-one correspondence with the main force arms; unmanned dynamictake-off and landing are completed through cooperation of the flight control module, the visual calculation module and the image acquisition and transmission module, the unmanned flexibility is greatly improved, and search and rescue tasks can be executed more efficiently.

The invention relates to a real time three dimensional imaging system based on laser light and visual sense. The real time three dimensional imaging system consists of three major parts that are a visual sense processing subsystem, a control subsystem and a machinery mechanism. Object images are obtained by a camera and are then processed by a visual sense processing computer, three dimensional position information in image coordinates can be obtained and sent to a common storage area, and a control computer obtains the three dimensional position information from the common storage area and performs coordinate conversion and real-time programming. Information of measured depth is fed back to the visual sense computer to obtain the three dimensional position information of an object. The system has the advantages of high distance precision, high activity, real-time performance, high speed, low cost, simple technology and the like.

The invention relates to a visual system with real-time parameter monitoring and on-line calibration functions. Collected images are divided into a monitoring calibration region and a visual rebuilding region; the images in the monitoring calibration region are subjected to real-time feature point extraction, and in addition, the extracted feature points are compared with the feature points calibrated in a last time; when the errors of the feature points are smaller than the set threshold, the images in a scene rebuilding region are subjected to feature point extraction, in addition, camera inner and outer parameters calibrated in the last time are used for carrying out visual calculation, and scene information such as visual target position, scale and the like is output; and when the errors of the feature points are greater than or equal to the set threshold, the feature points extracted in real time in the monitoring calibration region are used for calibrating the visual system, the calibrated feature points of the monitoring calibration region and the camera inner and outer parameters are updated according to the calibration results, and the functions of visual calculation, parameter monitoring and on-line calibration are completed. The visual system provided by the invention has the parameter real-time monitoring function and the on-line calibration function, the accuracy of the visual calculation results is ensured, and the system resources are saved.