363 results about "Visual angle" patented technology

The invention relates to the technical field of chip optical detection, and discloses a high-precision defect detection method and system for a semiconductor chip. Initial detection data are obtained through laser scattering and photo-thermal microscopic detection, key visual features are extracted and converted into a multi-dimensional visual feature matrix, and the multi-dimensional visual feature matrix is used for detecting the defects of the semiconductor chip; according to the method, a fuzzy sensing area is positioned and potential defects are identified by using a map clustering algorithm, adaptive analysis of a supplementary visual angle is carried out and supplementary detection data is generated, and finally defect identification is carried out through the supplementary data and popularity information to obtain a chip defect result. The method effectively improves the precision and reliability of defect detection, especially has obvious advantages in recognition of complex and fuzzy regions, and solves the problem of insufficient precision of single-path optical imaging detection in the prior art.

The invention discloses a visual angle tracking grating stereo display system comprising a 3D camera grating display device used for obtaining viewer 3D information; the eyeball position tracking module is used for acquiring real-time spatial position coordinates of the eyes of the audience relative to the 3D camera; the pose calibration module is used for acquiring a spatial relationship between the 3D camera and the display; and the visual angle rendering and grating interlaced graph generation module is used for calculating the spatial position of the human eyes on the display screen and the mapping of each grating strip in a screen pixel coordinate system after acquiring the positions of the eyes and the relative poses of the display screen and the 3D camera, generating a grating stereoscopic display image suitable for the eyes, and sending the grating stereoscopic display image to a display to be output. According to the method and the device, the viewpoint self-adaptive three-dimensional display of the eyes of one or more audiences is realized, the calculation complexity is reduced, the three-dimensional vision consistency can be ensured, the picture which should be seen by the viewing angle can be accurately rendered, and the problem of blurring or 3D picture distortion is avoided.

The invention discloses a scoliosis rapid screening method and device, and the method comprises the steps: S1, obtaining frame-by-frame images of a human spine through a visual sensor, and forming a human standing video sequence; s2, acquiring three-dimensional coordinate information of main skeleton key points of a human body; s3, analyzing the three-dimensional coordinate information of the main skeleton key points of the human body; s4, collecting human body front and back double-view-angle stoop motion videos through a color vision sensor; s5, carrying out frame-by-frame processing segmentation on the motion video sequence of human body stooping; s6, splicing the two-dimensional back contours extracted frame by frame to generate a complete back three-dimensional curved surface model; and S7, judging whether the scoliosis risk exists in the testee or not by combining the dynamic change indexes of the back contour features in the motion state, combining the back contour features of each frame extracted frame by frame and a plurality of indexes calculated according to the skeleton key points and inputting the indexes into the joint learning model.

The invention relates to the technical field of titanium alloy bar detection, and discloses a titanium alloy bar quality detection system and method. The system comprises a multi-modal data acquisition module, a detection reference dynamic optimization module, a data standardization processing module, a multi-modal feature fusion module, an intelligent defect identification and evaluation module, a quality grade evaluation module and a quality detection report generation module. When the quality of the titanium alloy bar is detected, through a multi-mode nondestructive detection mechanism integrating ultrasonic detection data, optical surface detection data and eddy current electromagnetic detection data, all-dimensional defect information of the interior, the surface and the near surface of the bar is synchronously obtained, and complementation and cross validation of the multi-mode detection data are achieved; the problem of limitation of a single sensor detection visual angle is solved, the missed detection and misjudgment risks are reduced, and the comprehensiveness and reliability of a defect detection result are ensured.

The present application provides an iris recognition method and apparatus, an electronic device, and a storage medium. The method comprises: obtaining an iris image group acquired for an object to be recognized, wherein the iris image group comprises at least two iris images acquired at different acquisition angles for a same eye region of said object; performing stereo matching on the at least two iris images to obtain a disparity map between the at least two iris images, wherein the disparity map comprises disparity elements, and each disparity element represents a displacement amount in a specified direction between two pixel points corresponding to a same eye region element in the at least two iris images; in one iris image among the at least two iris images, determining a pupil edge on the basis of the displacement amounts; determining an iris region in the iris image on the basis of the pupil edge, and performing feature extraction on the iris region to obtain an iris feature; and performing identity recognition on said object on the basis of the iris feature.

The embodiment of the invention discloses an aluminum plastic package detection method and system based on machine vision, and the method comprises the steps: carrying out the noise reduction processing of a single-visual-angle detection image of an aluminum plastic package, obtaining a noise reduction detection image, extracting an abnormal feature region, and obtaining an initial detection result; establishing a semantic association rule of a pre-trained semantic rule mapping model and the quality detection demand label of the aluminum plastic package; a synchronous quality control instruction in the production link of the aluminum plastic package is used as a semantic adaptation trigger condition, and an adjustment strategy of a semantic weight distribution unit is calibrated and activated; inputting the initial detection result into a semantic rule mapping model to perform semantic dimension priority ranking on abnormal features in the initial detection result to obtain an abnormal feature ranking result; and in combination with the defect judgment model and a defect grading rule in a current production scene corresponding to the production link, performing matching judgment on the abnormal feature sorting result, and outputting a defect judgment conclusion adaptive to the current production scene.

The invention discloses a maintenance platform and new energy automobile maintenance equipment and method, and belongs to the technical field of new energy automobile maintenance. Aiming at the problems that visual angle blind areas exist in existing detection, manual errors are large, bolts are prone to missing, and a cover plate is difficult to disassemble, the equipment is provided with a clamping structure driven by a two-way lead screw through a maintenance platform body, and two sets of liftable moving plates are matched with L-shaped clamping plates to achieve multidirectional stable clamping of a battery pack; according to the double-high-definition-camera mobile detection mechanism, a rack-gear-worm transmission assembly triggered by the end of a mobile table is used for driving a battery pack to rotate intermittently in a one-way mode, and a secondary driving mechanism is formed by a hydraulic cylinder and a trapezoidal plate in a matched mode. Maintenance equipment is integrated with a magnetic type bolt collecting device, and an adsorption seat moves to drive a vacuum suction cup to automatically lift a cover plate. Through mechanical transmission and hydraulic linkage design, full-process automatic improvement of battery pack detection, bolt management and cover plate disassembly and assembly is structurally realized, and the safety and efficiency of new energy automobile battery maintenance operation are effectively improved.

The invention discloses a monocular visual odometer method and system based on aerial view representation and differentiable weighted Procrustes solver, and the method comprises the steps: constructing a pose estimation model which can carry out the extraction and matching of key points in an aerial view, and carries out the relative pose estimation through an interpretable weighted Procrustes solver; the method comprises the following steps: constructing a loss function based on pose information, optimizing parameters of a pose estimation model by using the loss function, and performing pose information estimation between two frames of monocular images by using the parameter-optimized pose estimation model to realize the monocular visual odometer, so that the monocular visual odometer is superior in the aspect of reducing scale drift; and stable performance is shown under different weather and illumination conditions.

The invention discloses a construction safety management method based on excavator visual angle image fusion, and belongs to the technical field of construction early warning, and the method specifically comprises the steps: obtaining the position and working condition of excavation equipment, and setting a dynamic alarm fence; triggering an equipment vision device to scan a personnel target in real time, identify a personnel bounding box and extract a personnel identity identification code; synchronously calling original coordinates of the corresponding ultra-wideband tag, capturing a posture visual frame, and generating a posture feature vector through skeleton joint point topology analysis; inputting the attitude vector into an attitude-error compensation model, and outputting a positioning offset predicted value to correct an original coordinate; according to the method, real-time compensation is carried out through dynamic accurate matching of the construction dangerous area and personnel positioning errors, misjudgment is avoided to the greatest extent, the safety risk active pre-judgment and alarm accuracy is improved, and the safety risk active pre-judgment and alarm accuracy is improved. And intelligent safety management of a construction site is realized.

The endoscope comprises an insertion tube, a tip part, a first bending control wire and a second bending control wire, and the insertion tube comprises a main body section, a first bending adjustment section and a second bending adjustment section which are sequentially arranged from the near end to the far end; the far end of the first bending control wire is connected to the first bending adjusting section; the far end of the second bending control wire is connected to the second bending adjusting section; the tip part is connected to the far end of the second bending adjusting section, and a camera is arranged on the tip part. According to the endoscope, the first bending adjusting section and the second bending adjusting section which can be bent independently are arranged on the insertion tube, the visual angle of the camera on the tip part can be adjusted through double-section bending adjustment, the visual field range which can be observed by the endoscope is wider, the operation of surgical instruments and the excision condition of diseased tissue can be better observed through the endoscope, and the operation efficiency is improved. And the surgical operation difficulty of surgical operators can be reduced to a great extent.

The invention discloses a three-dimensional style migration method and device and a storage medium, and relates to the technical field of computer vision, and the three-dimensional style migration method comprises the following steps: adjusting a preset stylized three-dimensional Gaussian sputtering model based on an obtained target intensity vector of an adjustable style adjuster, the style intensity vector of the adjustable style regulator is associated with the attribute of each Gaussian ball in the stylized three-dimensional Gaussian sputtering model, and the style intensity vector of the adjustable style regulator is associated with the attribute of each Gaussian ball in the stylized three-dimensional Gaussian sputtering model; and carrying out style migration on the target scene based on the adjusted stylized three-dimensional Gaussian sputtering model. According to the method, through the stylized three-dimensional Gaussian sputtering model with the adjustable strength, the style images with the ideal migration degree and multiple visual angles can be obtained, and the flexibility of migration degree control of three-dimensional style migration is improved.

The invention discloses a dynamic semantic vision SLAM (Simultaneous Localization and Mapping) method based on point and line feature adaptive weighting. The method comprises the following steps: firstly, obtaining potential dynamic region prior by combining target detection and image segmentation; then adaptive weighting is carried out on point and line features in two stages: in the first stage, initial weighting is carried out based on motion consistency check and dynamic levels, and feature matching and pose preliminary estimation are realized; in the second stage, weighting is further carried out on the key frame level through multi-view consistency verification, and weighted optimization and map maintenance are combined; and finally, loopback verification and global optimization are completed based on weighted features, and cumulative drift is effectively eliminated. According to the method, interference of dynamic and low-credibility features is suppressed through differential feature weighting, the adaptability to an intermittent moving object is improved through a double-stage weighting mechanism, and the stability and observability of a low-texture scene are enhanced through point-line combined modeling.

The invention provides a nickel sheet welding visual angle control fool-proof method, and relates to the technical field of metal processing, and the method overcomes the interference of nickel sheet surface optical characteristic difference, spatial pose change and complex optical environment through gray scale adaptive processing, pose correction, included angle control and fool-proof determination, and realizes accurate positioning and poor welding interception. By means of self-adaptive gray threshold segmentation, a three-dimensional pose compensation model and Gaussian filtering edge enhancement, interference caused by nickel sheet surface roughness, workpiece inclination and ambient light changes is effectively overcome, the coarse positioning error is reduced to + / -0.2 mm from + / -0.5 mm, the fine positioning error is controlled to + / -0.05 mm, the grabbing success rate is increased to 98% or above, the welding reject ratio is remarkably reduced, and the welding quality is improved. According to the method, module scrapping is avoided, stable operation of a production line under the high-precision welding requirement is ensured, different nickel sheet materials and welding scenes are automatically adapted through the technological parameter table, the system universality is enhanced, technical support is provided for the procedures of tab cutting, battery cell stacking and the like, and the adaptability is high.

The invention discloses a method and a device for measuring the circle center of an outfitting pipe fitting with a shielded sight line, the device for measuring the circle center of the outfitting pipe fitting is designed, the device is arranged on a pipe flange of the outfitting pipe fitting, and the circle center of the pipe is guided and transferred into a measurement visual angle of a total station by using the device, so that the measurement of the circle center of the outfitting pipe fitting is realized. The three-dimensional coordinate of the target center of the measuring target on the device is measured by the total station, and then the three-dimensional coordinate of the circle center of the pipe is calculated, so that the actual position deviation of the circle center of the pipe is judged, the problems that the circle center of the pipe is shielded and cannot be measured and the total station needs to be carried back and forth are effectively solved, the measuring difficulty is reduced, and the measuring efficiency is improved.

The invention relates to the technical field of displays, in particular to a double-view-angle mini LED vehicle-mounted display device and a control method thereof. The double-view-angle mini LED vehicle-mounted display device comprises a main body frame which comprises an upper rubber shell, an inner rubber frame, a middle rubber shell and a lower frame, the middle rubber shell is connected with the lower frame in a clamped mode, the upper rubber shell is connected with the middle rubber shell in a clamped mode, and the inner rubber frame is connected with the bottom of the lower frame; the first visual angle backlight assembly comprises a first light source assembly, and the first light source assembly is arranged on one inner side wall of the lower frame; the second visual angle backlight assembly comprises a second light source assembly and a visual angle restraining piece, and the visual angle restraining piece is located on one side of the light emitting direction of the second light source assembly; the visual angle restraining part is located on the side, opposite to the light emitting direction, of the first visual angle backlight assembly. According to the invention, two different visual angles can be conveniently and rapidly switched.

The invention relates to the technical field of ophthalmologic operations, and provides an intelligent head position recognition and correction system in ophthalmologic operations, comprising: an intelligent pillow which is designed according to human head engineering, is provided with a self-adaptive adjustment module, and is used for performing automatic fine adjustment according to head contours and body types of different patients to ensure that the heads of the patients are stably supported; the infrared eyeball tracker is installed on a mechanical arm beside an operating bed, the infrared eyeball tracker is composed of a high-sensitivity infrared emitter, a wide-view-angle receiver and a matched optical filter, and the infrared eyeball tracker can stably work in a complex light environment and accurately capture infrared signals reflected by eyeballs. Through the high-precision nine-axis inertial measurement unit integrated in the intelligent pillow, various fine actions such as translation, rotation and inclination of the head can be monitored accurately in all directions, and the measurement precision reaches the sub-millimeter level and the sub-degree level. Therefore, the system can timely and accurately obtain the head position change information, and a reliable basis is provided for subsequent judgment and processing.

The invention relates to a robot operation planning method and system based on self-adaptive visual prompt. The method comprises the following steps: firstly, acquiring a first visual angle image and a task text instruction of a robot, and extracting a task related object list and a preliminary positioning area by using a pre-trained visual language model; visual key point detection is carried out by combining the data, and an operation candidate key point set and corresponding pixel coordinate marks are generated; according to a task instruction, candidate key point marks and the like, optimal grabbing points, path points and target points are selected from the set, pixel coordinates of the optimal grabbing points, the path points and the target points and the target direction of an end effector are output, and three-dimensional coordinates under a coordinate system of a coordinate robot are connected to generate a planned path of the end effector; and finally, in combination with the target direction and the planned path, solving the joint position and generating a continuous trajectory to complete operation planning. Compared with the prior art, the method has the advantages that the reasoning time is shortened, and the method can be efficiently and widely applied to various scenes.

The invention discloses a method and a system for automatically testing a virtual cabin of a whole vehicle based on digital twinning, and relates to the technical field of vehicle testing, a three-dimensional working condition space containing an eye point coordinate set and a polarization angle set is constructed by simulating the height and head micro motion of a driver, and a head-up display image in the three-dimensional working condition space is acquired through a camera; constructing a correction coordinate system and a gray brightness mapping table by combining a standard calibration board, and converting the image into a physical brightness matrix; marking elements in the physical brightness matrix through a brightness threshold value, extracting two parallel straight lines of a main track and a ghosting track through Hough transform and ODR sub-pixel refitting, and calculating a visual angle difference and a ghosting brightness ratio; and generating a test result table containing the three-dimensional working condition index based on a double-threshold joint judgment rule. The system comprises an image acquisition module, a feature extraction module, a data calculation module and a test judgment module, realizes automatic and accurate quantitative automatic test based on the system, and improves the virtual cabin test efficiency and reliability.

The utility model belongs to the technical field of optical instrument manufacturing, and particularly relates to an imaging brightness and chrominance and visual angle analyzer which comprises a shell, a rotating disc is arranged on the upper surface of the shell, a CCD electronic lens and a conoscopic lens penetrate through the position, deviating from the circle center, of the rotating disc, the distance between the CCD electronic lens and the circle center of the rotating disc is equal to the distance between the conoscopic lens and the circle center of the rotating disc, and a range finder is arranged in the center of the rotating disc. A spectrograph and an RGB camera are arranged in the shell, a through hole matched with the RGB camera and the spectrograph is formed in the upper surface of the shell, the distance between the through hole and the central axis of the rotary disc is equal to the distance between the CCD electronic lens and the central axis of the rotary disc, and a filter switching device matched with the through hole is arranged between the through hole and the RGB camera in the shell. A turntable driving device for driving the turntable is also arranged in the shell; the problem of how to test the optical characteristics of the spectrum, the brightness, the chromaticity and the visual angle of the display screen on one test device is solved.

The invention discloses an intelligent cooperative control system and method for vestibular function inspection. The invention discloses an intelligent cooperative control system and method for vestibular function examination, and belongs to the technical field of medical instruments. The system comprises a cold and hot air perfusion module, a dynamic visual target module, a head posture detection module, a sight distance detection module, an eye movement acquisition module, a voice interaction module, an adjustable examination chair and a central controller. The central controller realizes closed-loop control based on multi-source sensing data: during a cold and hot test, if the elevation angle of the head deviates from 30 degrees + / -2 degrees, stimulation is paused and voice prompt is performed; during visual target testing, visual target parameters are automatically compensated according to the actually measured sight distance of 1.5 m + / -0.1 m so as to maintain a constant visual angle, and the upright state (90 degrees + / -2 degrees) of the head is verified; abnormities such as eye closing more than or equal to 2 seconds or head movement more than or equal to 5 degrees per second are monitored in the whole process, graded voice intervention is implemented, and the abnormities are paused. Body position conversion is guided by voice, multi-source data are synchronously stored, and peripheral and central function correlation analysis is supported. According to the invention, the accuracy and compliance of inspection can be improved.

A 3D bounding volume within which a graphical element is to be rasterized for a given eye is determined. A set of light-emitting elements from which light is directed toward the given eye along a direction intersecting the 3D bounding volume is identified. For each light-emitting element in the set, its perceived shape and angular size from a perspective of the given eye is determined. The graphical element is rasterized onto the set of light-emitting elements based at least on relative perceived positions of the light-emitting elements from the perspective of the given eye, and their perceived shape and their perceived angular size. A colour value is computed for each light-emitting element in the set based on the rasterization of the graphical element. A synthetic light field is then produced displaying the graphical element, using computed colour values for the light-emitting elements in the set.

The invention provides a method for controlling a mowing robot to return to a charging station and a mowing robot, and the method comprises the steps: firstly obtaining an original image, then processing the original image through employing a first processing method to obtain a first binary image, and processing the original image through employing a second processing method to obtain a second binary image, and combining the advantages of the first processing method and the second processing method, carrying out logic AND operation on the first binarized image and the second binarized image to obtain a fused binarized image, and then determining a central reference line based on a visual identifier in the fused binarized image to guide the mowing robot to return to the charging station. According to the method, the stability of the preset threshold value and the flexibility of the adaptive threshold value are combined, the recognition precision of the visual identification under different view angle conditions is remarkably improved, the visual identification can be effectively distinguished under the conditions of strong light, shadow or color interference, and therefore the success rate of returning the mowing robot to the charging station is improved.

The invention belongs to the technical field of fruit quality detection, and particularly discloses a honey pomelo granulation nondestructive detection method based on X-ray and two-dimensional fruit shape feature fusion, and the method comprises the following steps: obtaining two-dimensional fruit shape features of honey pomelo based on an X-ray image sequence and an RGB image sequence of the honey pomelo, the X-ray image sequence is obtained by shooting an X-ray image of the honey pomelo in a rotating mode, and the RGB image sequence is obtained by shooting a surround video of the honey pomelo at a horizontal view angle; inputting the X-ray image sequence and the two-dimensional fruit shape features to a honey pomelo granulation detection model, and obtaining a granulation degree category output by the honey pomelo granulation detection model; wherein the honey pomelo granulation detection model is used for predicting the granulation degree category of the honey pomelo based on the X-ray image sequence and the two-dimensional fruit shape features. According to the method, the granulation degree of the honey pomelo can be efficiently and accurately detected under the condition that the honey pomelo is not damaged.

The utility model discloses a refrigerator display cabinet of evaporimeter upper placement, including the cabinet, the tray frame of setting in the inner chamber of cabinet and the supporting leg of setting in the bottom four corners of cabinet, the inside upper end of cabinet is provided with the evaporimeter, the bottom and left and right two ends of evaporimeter all are provided with the air outlet, the side of cabinet is provided with the auxiliary frame, the vertical swing joint of auxiliary frame has the lifting link, the bottom of lifting link is provided with the universal wheel, the upper end of auxiliary frame is provided with the drive assembly for driving the lifting of lifting link, this refrigerator display cabinet of evaporimeter upper placement, will evaporimeter remove to the food top of cabinet, and its air inlet and outlet are located above, effectively prevent the health hidden danger of food residue falling into and polluting the evaporimeter, significantly improve the cleaning convenience and health grade of equipment, simultaneously, the cabinet bottom case structure can be thinned, makes the display food can be placed in the lower position, optimizes the customer visual angle, strengthens the product display effect.

The utility model discloses an intelligent support with a face recognition function, the intelligent support comprises a base and a support frame arranged on the base, the support frame is provided with a camera and a connecting mechanism used for being connected with external display equipment, the intelligent support further comprises a movement mechanism, a driving mechanism and a main controller, the camera is used for shooting a face image and transmitting the face image to the main controller, the main controller obtains the angle and height, needing to be adjusted, of the intelligent support through the face image and transmits angle and / or height data to the driving mechanism, and the driving mechanism controls the movement mechanism to move so as to adjust the angle and / or height of the connecting mechanism. According to the invention, after the face is input for one time, the subsequent screen height and angle adjustment can be automatically completed through the intelligent support, manual adjustment is not needed, convenience is provided for people in need, and the optimal visual angle is provided for an operation user all the time.

The invention discloses a speech emotion recognition method based on multi-modal multi-view pseudo-tag fusion, and relates to the technical field of emotion recognition, and the method comprises the steps: firstly, constructing a visual-view pseudo-tag generation module, extracting high-quality facial expression features from a video through employing an optimized OfficientNet-V2 network and a RetinaFace face detector, and carrying out the recognition of the high-quality facial expression features; generating a visual pseudo label through time sequence aggregation; secondly, adopting a semi-supervised learning framework FixMatch to generate a reliable voice pseudo tag for the non-tag voice; secondly, fusing emotional knowledge from multiple visual angles and modes of vision and voice based on a pseudo-tag fusion function, and generating a more accurate and robust fused pseudo-tag; and finally, training a voice emotion recognition model based on a Conformer encoder to perform voice emotion recognition by combining the fused pseudo label data with the labeled voice data. According to the method, the problem of error accumulation in traditional semi-supervised learning is effectively solved, cross-modal emotion complementarity is utilized, and the accuracy and generalization ability of speech emotion recognition are remarkably improved.

The utility model discloses a powder box with an angle-adjustable lens. The powder box comprises a box bottom, a box cover, a lens seat and the lens, the lens seat and the box cover are connected in a rotating fit mode through a shaft column and a shaft seat, the lens seat is arranged in the lens seat installation groove in a rotating mode, the shaft column is regularly arranged on the back face of the lens seat, the shaft seat is regularly arranged in the lens seat installation groove, and the shaft column and the shaft seat are matched with each other, clamped into a shaft column clamping groove of the shaft seat and connected with the shaft column clamping groove in a rotating mode. The shaft column or the shaft ball is arranged on the back face of the lens seat, the shaft seat is regularly arranged on the inner side of the box cover, the lens seat is matched with the shaft seat through the shaft column or the shaft ball to rotate, and the visual angle of the lens is adjusted through the lens seat. Compared with a traditional powder box, the powder box is simple in structure and convenient to assemble. Meanwhile, after the box cover of the powder box is opened and fixed, the visual angle of the lens can be further adjusted through the lens seat, and more lens visual angles are provided for consumers.

The utility model relates to the technical field of vehicle-mounted display screens, in particular to a novel separated vehicle-mounted display screen framework which is characterized in that a sliding groove and a limiting groove are formed in a mounting plate; an electric push rod and a connecting rod are arranged on the connecting plate; an electric telescopic rod pushes a limiting block to be in sliding fit with a connecting rod and a moving rod along a limiting groove, so that a connecting plate can rotate along a rotating shaft on the connecting rod, the angle of the display panel is changed by changing the included angle between the connecting plate and a mounting plate, and the visual angle of the display panel is increased; a worker pulls a pull rod to enable a sliding block to slide along a through groove, a clamping block fixedly connected to the sliding block slides along the through groove and is separated from a clamping groove, the installation plate and the display panel can be detached from the vehicle-mounted host, maintenance personnel can carry out maintenance work on the vehicle-mounted host, and the maintenance efficiency of the vehicle-mounted host is improved. And the whole vehicle-mounted host does not need to be disassembled, so that the maintenance cost and the maintenance time are reduced, and the practicability is high.

The invention relates to the technical field of three-dimensional display, and discloses a three-dimensional display visual angle adjusting method, device and equipment, a medium and a program product, the method acts on a liquid cylindrical lens, and comprises the following steps: determining a current applied signal according to an expected visual angle range, so as to generate a driving force based on the current applied signal, enabling the liquid cylindrical lens to deform based on the driving force so as to change the curvature radius; and after the curvature radius of the liquid cylindrical lens is changed, the light beam emitted by the display is controlled to penetrate through the liquid cylindrical lens and is refracted, so that the actual visual angle range of three-dimensional display reaches the expected visual angle range. The curvature radius of the liquid cylindrical lens is changed by applying a signal, and light beams emitted by a display can be refracted at different transmission angles, so that the optimal viewing angle range of an observer is enlarged, the flexibility of a three-dimensional display viewing angle is improved, and the viewing experience is improved.

The invention discloses a sea surface target tracking method based on a visual guidance radar under a high sea condition. The method comprises the following steps: synchronously collecting millimeter wave radar point cloud, image, positioning, direction and speed data when an unmanned ship sails; detecting an image target by using a YOLOv8 model subjected to slideloss weighted training and model distillation optimization to obtain a target detection frame; preprocessing the radar point cloud; point cloud proposal and classification are carried out based on an image detection frame, clustering clusters are generated through DBSCAN clustering, and visual angle information is fused in proposal clusters; and the clustering cluster is converted to a world coordinate system, dual matching with a historical target is realized through a Hungary algorithm, the unmatched target is initialized, and the state of the matched target is updated by using extended Kalman filtering and the future position is predicted. According to the method, a close-range sensing blind area is filled, the precision and stability of sea surface target tracking under the high sea condition are remarkably improved, and reliable guarantee is provided for safe navigation of the unmanned ship.