42 results about "Motion vision" patented technology

The invention discloses a method for extracting a video interested region based on a visual attention. The interested region extracted by the method combines a static image domain visual attention, amotion visual attention and a depth perception attention, and the method efficiently restrains the internal unicity and inaccuracy of each visual attention extraction and solves the problem of noisescaused by a complex background in the static image domain visual attention and the problem that the motion visual attention can not extract an interested region with small local motion and motion amplitude, thus the method improves the calculation accuracy, enhances the algorithm stability and extracts an interested region from a background with complex veins and the motion environment; furthermore, the interested region obtained by the method accords with the depth perception feature interesting on an object having a strong depth perception or a short distance in the stereovision and the semantic characteristic of human eye stereovision besides according with the vision feature of human eye interesting a static grain video frame and interesting a motion object.

The communication device which implements the target device location indicating function, the motion visual image producing function, the non-motion visual image producing function, and the auto backing up function. The target device location indicating function indicates the current location of the target device, the motion visual image producing function retrieves the motion visual image data from the camera, the non-motion visual image producing function retrieves the non-motion visual image data from the camera, and the auto backing up function transfers the data stored in the communication device to another computer for purposes of storing backup data therein.

A system for creating the perception of visual depth in scenery comprising lateral movements. A polarizing projection filter 21 controls the plane of polarization of projected light. The plane of polarization is changed by changing the orientation of the polarizing projection filter. An observer wears special viewing glasses 22 with polarizing filters when viewing scenery illuminated by the polarizing projection filter. The plane of polarization of the polarizing viewing filter for the right eye 23 is rotated 45 degrees clockwise from horizontal. The plane of polarization of the polarizing viewing filter for the left eye 24 is rotated 45 degrees counterclockwise from horizontal. When the polarizing projection filter is rotated clockwise from horizontal the scenery appears brighter to an observer's right eye than to the observer's left eye. Objects moving laterally 42 from right to left from the observer's perspective appear closer to the observer 43. Objects moving from left to right appear farther away from the observer. When the polarizing projection filter is rotated counterclockwise from horizontal the scenery appears brighter to an observer's left eye than to the observer's right eye. Objects moving from right to left from the observer's perspective appear farther away from the observer. Objects moving from left to right appear closer to the observer. In the first and second embodiment of the invention, the scenery comprises polarized imagery projected on a viewing screen. In the third embodiment, the scenery comprises live scenes illuminated by polarized light. The scenery can be viewed in full color. The dead eye effect can be avoided by periodically changing the plane of polarization of the polarizing projection filter.

A system (1) for detecting dynamic secondary objects (55) that have a potential to intersect the trajectory (51) of a moving primary object (50), comprising a vision sensor (2) with a light-sensitive area (20) that comprises event-based pixels (21), so that a relative change in the light intensity impinging onto an event-based pixel (21) of the vision sensor (2) by at least a predetermined percentage causes the vision sensor (2) to emit an event (21a) associated with this event-based pixel (21), wherein the system (1) further comprises a discriminator module (3) that gets both the stream of events (21a) from the vision sensor (2) and information (52) about the heading and / or speed of the motion of the primary object (50) as inputs, and is configured to identify, from said stream of events (21a), based at least in part on said information (52), events (21b) that are likely to be caused by the motion of a secondary object (55), rather than by the motion of the primary object (50).Vision sensors (2) for use in the system (1).A corresponding computer program.

The invention discloses a binocular vision positioning method and system in a dynamic environment applied to the technical field of robot positioning. The method comprises the following steps of extracting image feature points, distinguishing the feature points on the moving object points and the feature points on the non-moving object points in a mode of clustering the feature points. The featurepoints on the moving object can be effectively eliminated so that elimination of feature points on a static dynamic attribute object is avoided, more available feature points are reserved, and poseswhich are closest to IMU data and are estimated based on different feature point sets are screened through an inter-frame pose transformation matrix obtained through an IMU pre-integration algorithm,so that some external points such as feature points belonging to a moving object are effectively eliminated. According to the poses and depth estimation method integrating monocular motion vision andbinocular stereoscopic vision, the estimation precision can be effectively improved.

The invention discloses an old people indoor fall-down detection method based on computer vision. According to the method, first, a GMM model is utilized to extract a human body original contour, andhuman body coordinates are positioned by use of an ellipse fitting method; second, an ellipse is divided into different regions, centroids of all the regions are tracked through a closest distance method, and multi-centroid offset vectors are extracted to serve as motion features; third, a K-means clustering algorithm and a TF-IDF algorithm are used to establish a human body motion vision statement; and last, the human body motion vision statement is classified through a DAG-SVM to detect whether old people fall down. The method has the advantages that a fall-down action, a similar fall-down action and a lying action can be distinguished, a fall-down action parallel to a camera principal axis can also be detected, and the fall-down detection rate is high; and the method can be implementedjust through simple hardware equipment and is low in cost and easy to implement.

The invention discloses an active hand training system and method based on brain-computer interaction and deep learning. The active hand training system comprises an electroencephalogram cap, an FPGAacquisition device, an image stimulation module, an upper computer and a hand movement support; the electroencephalogram cap and FPGA acquisition equipment are used for acquiring a motor imagery electroencephalogram signal of a testee, carrying out preprocessing operation and wirelessly transmitting the pretreated signal to the upper computer; the image stimulation module generates corresponding motion visual prompts according to the training items so as to prompt the testee to perform imagination of corresponding action; meanwhile, after receiving the electroencephalogram signal, the upper computer decodes the electroencephalogram signal and generates a corresponding control signal; and the hand movement support receives the control signal to pull the hand of the testee to move, and corresponding training is completed. The device has the characteristics of high safety, labor saving and high interestingness, the initiative of a user can be exerted to the maximum extent, and the rehabilitation effect is improved.

The invention relates to a six-degree-of-freedom motion capture method for a wind tunnel model, and belongs to the field of hypersonic wind tunnel non-contact measurement. A high-speed binocular stereoscopic vision system is built, feature points arranged on the surface of a model are captured in real time, and spatial coordinates of the model in the test process are reconstructed based on deep learning and stereoscopic vision, so that six-degree-of-freedom motion data of the model are calculated; and based on high-speed binocular stereo vision and deep learning, the invention provides a six-degree-of-freedom spatial motion vision capture method suitable for a hypersonic wind tunnel test, and model motion parameters and a visualization result are obtained.

A sports vision training device is formed from a piece of material having a thickness sufficient to interfere with an individual's ability to look down at a sporting object. The piece of material has an adhesive coating or layer for positioning the material on an individual's cheek, preferably beneath an individual's eye, so that the piece of material interferes with the individual's ability to look down at the sporting object. The sports vision training device has a thickness sufficient to direct the individual's vision toward the field of play and up and toward at least one player on the field of play, and in some cases permitting intermittent and / or momentary glances at the sports object being controlled by the individual.

A contrast sensitivity detection method based on motion visual evoked potential comprises: designing concentric ring stimulating paradigms, namely a stimulating paradigm contrast scanning gradient, setting up a brain-computer interface platform, and carrying out stimulating paradigm displaying; subjecting electroencephalogram data to feature extraction, and judging a paradigm response value by calculating SNR (signal-to-noise ratio) of canonical correlation analysis spectrum so that accuracy of SSMVEP (steady-state motion visual evoked potential) contrast sensitivity detection is improved; and feeding back out the electroencephalogram objective detection results of contrast sensitivity through a screen. A group of concentric ring scanning SSMVEP paradigms is designed herein to contrast sensitivity inspection; the corresponding threshold judging standard is provided; objective contrast sensitivity of electroencephalogram is attained; an objective quantitative measuring method can be provided for for detecting contrast sensitivity in ophthalmic testing.

A visual acuity detection method based on motion visual evoked potential comprises: designing a concentric ring stimulating paradigm, designing a stimulating paradigm spatial frequency gradient, setting up a brain-computer interface platform, displaying the stimulating paradigms, performing visual acuity threshold judgment on electroencephalogram results, and feeding back visual acuity detection results. The concentric ring SSMVEP (steady-state motion visual evoked potential) paradigm for visual acuity examination is designed herein; corresponding threshold judging standards are provided; so that objective visual acuity of electroencephalogram is attained. The objective quantitative measuring method is provided for visual acuity detection in ophthalmic testing.

A sports vision training device is formed from a piece of material having a thickness sufficient to interfere with an individual's ability to look down at a sporting object. The piece of material has an adhesive coating or layer for positioning the material on an individual's cheek, preferably beneath an individual's eye, so that the piece of material interferes with the individual's ability to look down at the sporting object. The sports vision training device has a thickness sufficient to direct the individual's vision toward the field of play and up and toward at least one player on the field of play, and in some cases permitting intermittent and / or momentary glances at the sports object being controlled by the individual.

The embodiment of the invention discloses a visual training system. The visual training system comprises training glasses, a display device and a central processing module, wherein the training glasses are used for being worn at the eyes of a human body; the display device is mounted in front of lenses of the training glasses and used for displaying image information for visual training; and the central processing module is in communication connection with the display device, the central processing module is used for transmitting the image information for visual training to the display devicefor display, and the central processing module controls images and contents of the display screen, the training time and the number of training times. According to the visual training system and method provided by the embodiment of the invention, visual training and intelligence development are integrated, motion vision and vision diversification cross-channel interaction are realized, eyeball movement training is carried out according to the vision triple movement principle, coordination and sensitivity among all parts of the eyes are improved, vision is improved, visual fatigue is effectively relieved, the purposes of eliminating pseudomyopia and controlling occurrence and development of true myopia are achieved, and the visual training system has important significance in reducing myopia complications and preventing blindness.

The invention discloses a rocket engine spraying pipe motion vision measuring method and a portable infrared light emitting device used in same and relates to the rocket engine spraying pipe motion vision measuring field. The rocket engine spraying pipe motion vision measuring method and the portable infrared light emitting device used in the same solve the problem that a cooperative target cannot be correctly identified from a camera image plane due to the huge difference between a cold and hot trial running process and a simulated spraying pipe operating condition of the rocket engine when motion parameters of the rocket engine spraying pipe are measured. According to the rocket engine spraying pipe motion vision measuring method, a plurality of infrared luminous points are arranged on the outer surface of the rocket engine spraying pipe, an infrared optical filter is arranged before a lens of an image collecting device which collects images of the plurality of infrared luminous points which are arranged on the outer surface of the rocket engine spraying pipe, collected images are sent to a vision measuring system controlled computer, the rocket engine spraying pipe motion vision measuring purpose is achieved, and the infrared luminous points are achieved through a high luminance infrared light emitting device. The rocket engine spraying pipe motion vision measuring method is applicable to rocket engine spraying pipe motion vision measuring.

The invention discloses a training and mode switching method of a pure idea control rehabilitation robot. The method comprises the steps that a user generates a motor imagery electroencephalogram signal through a motor vision stimulation module; an electroencephalogram collection module collects synchronous electroencephalogram signals of a user, carries out amplification filtering preprocessing and transmits the signals to the upper computer in a wireless mode; after the upper computer receives the electroencephalogram signals, feature extraction is carried out through adoption of a filter bank CSP method, a fruit fly algorithm optimizes a tree-shaped SVM classifier to carry out feature classification, and corresponding rehabilitation training control signals are transmitted to a rehabilitation exercise device; and after receiving the training control signal, the rehabilitation exercise device performs rehabilitation exercise training on the limbs according to the corresponding control signal. The brain of the user directly participates in rehabilitation training through pure thought control, reconstruction of brain neurocortex of the user and restoration of the motor neural network are facilitated, meanwhile, the initiative and enthusiasm of the patient are stimulated, the convenience and comfort of rehabilitation treatment are improved, and the effect of rehabilitation treatment is improved.

A system for harvesting produce from a tree has a drone capable of hovering, a video camera gathering visual data of movement, a cutting implement, a remote control station with a display screen, wireless circuitry, and input mechanisms to control movement of the drone and operation of the cutting implement, and circuitry in the body of the drone enabling two-way communication with the remote control station, transmission of video data from the video camera, and response to commands from the remote control station. The video data from the camera on the drone is displayed on the display screen of the remote control station, and an operator viewing the display screen operates the input mechanisms, maneuvering the drone to position the cutting implement relative to produce in the tree, and triggers the cutting implement by command, severing a stem to separate the produce, causing the produce to fall from the tree.

The invention provides a human body motion measurement method and a system based on computer vision and electronic equipment, and the method comprises the steps: obtaining a human body motion video stream through a human body motion measurement method based on optics, and extracting a current reference image and a current frame image from the human body motion video stream; according to the current reference image and the current frame image, obtaining motion information of image pixel points in the current frame image through calculation, obtaining human skeleton key point coordinate information in the current frame image, and at least according to the motion information of the image pixel points in the current frame image and the human skeleton key point coordinate information, obtaining the motion measurement information of the human body joint motion through calculation, and combing semantic vision and motion vision, so that the human body motion measurement method is more convenient to operate, and the obtained motion measurement information is more robust.

The invention relates to the field of motion vision, and particularly relates to a multi-selectivity attention evaluation and training method and system, and aims to solve the problems that a trainingresult cannot be quantified and precise training cannot be realized for different users. The evaluation method comprises the steps of displaying a plurality of target objects with different appearances on a display screen; recording an ID number of a selected target object in the plurality of target objects; shielding the appearance distinguishing features of the plurality of target objects to enable the appearances of the plurality of target objects to be the same; controlling the plurality of target objects to move on the screen according to the movement parameters, simultaneously acquiringan eye movement track and an electroencephalogram map when the testee observes the selected target object, and acquiring a judgment value of the testee for the position of the selected target objectat the end of movement; and generating an evaluation result according to the actual motion trail, the eye movement trail and the electroencephalogram map of the selected target object and the judgmentvalue of the testee for the position of the selected target object at the movement end moment. According to the invention, targeted attention training and improvement of the user are realized.

The invention relates to the field of motion vision, in particular to a target predictive thinking evaluation and training method and system, and aims to solve the problem of how to accurately evaluate observation ability and predictive ability and improve training efficiency. The method comprises the following steps: controlling an observed target object to move and disappear on a display screenaccording to movement parameters, acquiring an eye movement track of a testee in the process of observing the target object, and acquiring a judgment value of the testee for the position of the targetobject at the end of observation; and generating an evaluation result according to the actual motion trail and the eye movement trail of the target object and the judgment value of the testee to theposition of the target object at the observation end moment. According to the method, the testee can be helped to accurately evaluate the observation ability and the pre-judgment ability of the testee, the method is not limited by sites, a large amount of physical strength does not need to be consumed, the observation ability and the pre-judgment ability can be trained more quickly in a targeted mode, and therefore the training efficiency is effectively improved.

The invention discloses a movement vision measurement method of a rocket engine nozzle and a high-brightness infrared light-emitting device used in the method, relates to the field of movement vision measurement of the rocket engine nozzle, and solves the problems that a cooperative target cannot be correctly identified on a camera image plane caused by a great difference between a cold and hot commissioning process of the nozzle and a simulated nozzle working condition in measurement on movement parameters of the rocket engine nozzle. A plurality of infrared luminous points are arranged on the outer surface of the rocket engine nozzle and connected to form a light-emitting circuit in a multi-channel parallel manner; image collection equipment collects the images of a plurality of infrared luminous points at the outer surface of the rocket engine nozzle in movement and transmits the collected image to a vision measurement system control computer, so as to achieve the target of movement vision measurement of the rocket engine nozzle; the infrared luminous points are achieved by adopting the high-brightness infrared light-emitting device. The movement vision measurement method is applicable to movement vision measurement of the rocket engine nozzle.

The invention relates to the field of automation equipment, and provides compartment position detection equipment which is used for detecting whether a compartment of a truck is parked in a designated loading position or not. The compartment position detection equipment comprises a truss, a cross beam, a moving arm, a visual sensing device and a compartment detection treatment device, wherein the truss is arranged above the loading position and at least comprises a front beam and a side beam; the cross beam is arranged on the truss and can move in the length direction of the truss; the moving arm is arranged on the cross beam and can move in the length direction of the cross beam; the visual sensing device is arranged on the moving arm, moves on the truss along with the moving arm and the cross beam and collects appearance and position data of the compartment; and the compartment detection treatment device calculates the appearance data and the position data of the compartment according to the data collected by the visual sensing device. The compartment position detection equipment has the advantages that integration and automation of goods delivery and loading are achieved, and the goods loading efficiency and accuracy are greatly improved.

The invention discloses a method for displaying the results of visual detection of a curved object to be tested. The visual detection device is fixed on a robot and moves at a uniform speed synchronously with the end of the robot. The visual detection device then acquires an image of the corresponding position of the curved object to be measured; overview There are multiple display frames evenly distributed on the map, and the display frames are associated with the detection results of the corresponding positions; the multiple display frames distributed on the overview map are obtained by the following steps before the test: 1) Obtain an overview of the curved object to be tested Figure, manually mark the first and last two endpoints, and then manually mark the schematic points along the curved object to be measured, all points are recorded as the first type of points; the first type of points can feedback the shape characteristics of the curved object to be measured; 2) in Evenly insert the second-type points between two adjacent first-type points; 3) draw the display frame according to the positions of all the first-type and second-type points. The display position of the method corresponds to the visual detection result, and can display the actual position of the detection result on the part in real time.

The invention discloses soft frame attaching equipment. The soft frame attaching equipment comprises a machine frame, a first positioning assembly, a first lifting device, a visual positioning device, a bearing disc, a position adjusting module and a second lifting device, wherein the machine frame is provided with a positioning platform, the first positioning assembly is arranged on the positioning platform and defines a set shape so as to position a soft frame, the first lifting device is arranged below the positioning platform and is connected to the first positioning assembly so as to drive the first positioning assembly to move in the direction perpendicular to the upper surface of the positioning platform, the visual positioning device is arranged below the positioning platform and corresponds to shooting holes in the four corners of the soft frame so as to acquire the position information of the four corners of the soft frame, the bearing disc is arranged above the positioning platform and is used for bearing an object to be attached, the position adjusting module is arranged below the positioning platform and is connected to the bearing disc so as to adjust the position of the object to be attached to correspond to the position of the soft frame, and the second lifting device is arranged below the positioning platform and is connected to the position adjusting module so as to drive the object to be attached to make contact with the soft frame. According to the attaching equipment, the positioning precision and the working efficiency of the soft frame can be improved.

The invention relates to a surgical assistance system and a control method thereof. The surgical assistance system comprises a visual marker and surgical assistance equipment. The visual marker is disposed in a surgical bed system. The surgical assistance equipment comprises a support arm, a support unlocking mechanism, a visual tracking device, and a control device. The visual marker is disposedin the surgical bed system. The visual marker performs synchronous movement with the surgical bed system. The visual tracking device is used for acquiring a first optical signal emitted by the visualmarker. The control device is used for determining whether the surgical bed system is in a safe state or not according to the first displacement signal. If the surgical bed system is in the non-safe state, the control device controls the support unlocking mechanism to unlock to separate the support arm from the surgical bed system. The surgical auxiliary system effectively prevents a situation that serious injury is caused to a surgical site of a patient due to that the surgical site of the patient is still fixed on a bracket fixedly connected with the support arm when the surgical bed suffersfrom collapsing, tilting or other accidents.

The invention discloses a system and method for calibrating motion of a motion module and a robot, and the system comprises a calibration plate, a visual detection device, a controller and a servo motor. The servo motor controls the motion module to move according to the target coordinates issued by the controller; the visual detection device at least collects images of the motion module and the calibration plate once in the motion process of the motion module; the controller obtains actual coordinates of the motion module according to the image, obtains error coordinates according to the target coordinates and the actual coordinates and obtains an error coordinate table, the controller corrects the target coordinates through the error coordinate table to obtain corrected coordinates so as to control the servo motor to control the motion module to move to the target position according to the corrected coordinates, and then the motion module moves to the target position according to the corrected coordinates. For a large-stroke module and an equipment machine with a high precision requirement on the large-stroke module, the problems that the precision performance of the equipment machine is difficult to test, the test quantitative index is lacked, the test cost is high, the efficiency is low and the like in the development or mass production stage can be effectively solved.

The invention discloses a glass bottle tracking and counting system. The system comprises a visual imaging module, a motion control module and an information processing module; the visual imaging module and the motion control module are connected with the information processing module; the information processing module is used for inputting an opening signal to the motion control module, inputting a stop signal to the motion control module, receiving a video image sent by the visual imaging module, and carrying out ID labeling on the updated Tracks according to a glass bottle tracking counting method to complete counting; the motion control module is used for controlling the conveyor belt to drive glass bottles on the conveyor belt to reciprocate within a preset boundary range according to the received opening signal, and stopping the motion according to the received stop signal; and the visual imaging module is used for collecting video images of the glass bottles moving back and forth and sending the video images to the information processing module. The speed and accuracy of tracking and counting the glass bottle opening are effectively improved.

The invention relates to a wearable continuous flexible outer limb robot for narrow space operation, which comprises a wearable back support plate, two continuous rope-driven mechanical arms, a driving motor module, a visual sensing module, a driving control module and a power supply module, each joint is formed by embedding a plurality of connecting frameworks end to end; the drive control module is connected with the drive motor module, the drive motor module drives the drive rope to control the mechanical arm to move flexibly, and the visual sensor can rotate by 360 degrees and bend by 90 degrees. The problems that large assisting equipment cannot enter a narrow space, and an operator is difficult to complete complex work are solved, additional assistance can be provided for the operator when the two hands work at the same time, the visual sensing module is used for path planning and flexible obstacle avoidance of the mechanical arm, and therefore the working difficulty and labor intensity of the operator are reduced, and the working efficiency is improved.