72 results about "Near sight" patented technology

The invention discloses a video stream defogging method and system for 5G remote control, and relates to the technical field of image enhancement. The method comprises the following steps: acquiring a foggy video in real time; inputting the single-frame foggy video into a pre-trained monocular depth estimation neural network, and reasoning to obtain a depth map with the same size as the input single-frame foggy video; performing close-shot and long-shot region division on the scene of the current foggy video frame according to the depth map to obtain a region mask with the same size as the depth map; based on the region mask and the foggy video, calculating to obtain an atmospheric light parameter; calculating the transmissivity of each pixel according to the distance estimation result of each pixel in the depth map, and obtaining a transmissivity map with the same size as the depth map; based on the depth map, the atmospheric light parameters and the transmissivity map, adaptive calculation is performed on the foggy video to realize defogging, and a defogged clear video frame is obtained; the method can adapt to different depth-of-field fog effects, realizes different depth-of-field defogging, and outputs clear video frames.

The invention discloses an airport runway friction coefficient detection method based on a close-range camera technology. The method comprises the following steps: manufacturing a remote control trolley; setting driving parameters of the remote control trolley and camera shooting parameters of the high-definition camera; driving the remote control trolley to run on the detected road section, and acquiring road surface images or videos; the method comprises the following steps: firstly performing coordinate conversion on a plurality of obtained pavement images at different shooting angles in the same time period to obtain a pavement three-dimensional coordinate image of the road section, and then preprocessing the obtained three-dimensional coordinate image; performing image feature extraction on the preprocessed image to generate a pavement image texture model, importing the generated image texture model into Matlab, and performing three-dimensional texture reconstruction of pavement surface texture based on a triangular network model to obtain a three-dimensional model of the airfield pavement; and calculating friction coefficients of different positions of the road surface according to the three-dimensional data of the road surface in the three-dimensional model through a neural network algorithm.

A telephoto lens (1), a camera module (10) and an electronic device (100). And when the telephoto lens (1) is focused at a distant view, the field angle FOV of the telephoto lens (1) is less than 60 degrees. The telephoto lens (1) includes a first lens group (G1) and a second lens group (G2) arranged from an object side to an image side, the first lens group (G1) having a positive refractive power, the second lens group (G2) having a negative refractive power, and a distance between the first lens group (G1) and the second lens group (G2) varies during a focusing process in which the telephoto lens (1) is switched between a distant view and a close view. The telephoto lens (1) not only can realize long-range shooting and high imaging quality, but also can have relatively strong close-range shooting capability, so that wide object distance imaging from long-range shooting to close-range shooting is realized.

The invention provides a three-dimensional geologic body rendering method and system which are used in the field of geospatial information visualization, and the method comprises the steps: obtaining a three-dimensional geologic body model; simplifying the three-dimensional geologic body model by a half folding algorithm based on secondary error measurement; blocking the simplified geologic body model through a planar quadtree, and constructing a spatial index tree according to the blocks to generate a geologic body blocking model of an LOD level; obtaining a viewpoint position and a vision field range of a user, calculating a screen space error, switching a geologic body partitioning model corresponding to an LOD level, and preferentially loading a high-precision close-range tile; the calculation shader based on the WebGPU performs parallel rendering on the geologic body block model tiles; and after the tiles of the geologic body block model are compressed through a Draco compression algorithm, the geologic body block model data stream is sent to the client. According to the scheme, the processing efficiency of the geologic body model in the links of loading, rendering, transmission and the like is improved, and the rendering efficiency of the geologic body model under mass requirements can be effectively guaranteed.

The invention belongs to the technical field of intelligent glasses, and discloses intelligent optical zoom glasses which comprise two glasses frames provided with lenses, a nose bridge frame is connected between the two glasses frames, the two glasses frames are provided with nose pads and pile heads, the two pile heads are respectively provided with a left glasses leg and a right glasses leg in a hinged mode, the lenses are optical waveguide glasses lenses, and the optical waveguide glasses lenses are optical waveguide glasses lenses. An electrochromic liquid crystal film is mounted on the front side of the lens; the nose bridge frame is provided with a distance sensor; according to the invention, under the combined action of the lens prism module and the optical waveguide spectacle lenses, rapid long-shot, close-shot, micro-distance and wide-angle zooming is realized, and intelligent zooming can be realized at any time through a distance sensor, a left brain-computer signal sensor and a right brain-computer signal sensor on a spectacle cross beam. Manual focusing can be carried out through a focusing key, and voice focusing can be carried out through a microphone.

The invention discloses an unmanned ship binocular vision positioning method in a near-shore environment based on far and near scene segmentation, and relates to the field of unmanned ship positioning, and the method comprises the steps: obtaining an image of the near-shore environment collected by an unmanned ship binocular camera; dividing the image into a long-shot area and a close-shot area by adopting an OTSU algorithm; extracting feature points of a long-shot area and a close-shot area in the image by adopting an ORB algorithm; performing feature point matching on the feature points of the two adjacent frames of images to obtain an initial pose of the unmanned ship; and screening and inserting a new key frame, and optimizing the initial pose and the local map to obtain the real-time pose of the unmanned ship so as to realize the positioning of the unmanned ship. According to the invention, accurate positioning of the unmanned ship can be realized.

The present invention provides a three-dimensional object detection method, system and storage medium for multi-dimensional fusion of near-view and far-view. The three-dimensional object detection method includes: Step 1: Use Centernet to detect the center points of the image and regress the basic attributes, and extract the eigenvalues through a fully convolutional encoding-decoding backbone network; Step 2: Use the estimated depth to divide the three-dimensional ROI region of the object, and then divide the detection task into near-view detection and far-view detection. The beneficial effects of the present invention are: The three-dimensional object detection method of the present invention combines the advantages of lidar, millimeter-wave radar and camera, realizes the technology of 3D object detection in the field of autonomous driving, can accurately identify and locate objects such as vehicles, pedestrians, cyclists, etc., and can be applied in actual scenarios.

This application relates to the field of image detection technology, specifically to a method, device, and system for pedestrian trajectory tracking across cameras. Specifically, it includes: dividing a surveillance image into far-field, mid-field, and near-field regions; analyzing the imaging characteristics of moving pedestrians in different regions to construct regional change coefficients for far-field and near-field regions at different time periods; constructing motion change influence coefficients for each region by analyzing the pixel feature changes of clusters within the surveillance image during pedestrian movement and the grayscale changes of each pixel over time; determining the tracking region in each far-field and near-field region based on the regional change coefficients and motion change influence coefficients; reducing the hard threshold for denoising in the tracking region and mid-field region to retain more image details, thereby performing pedestrian trajectory tracking; avoiding the problem that traditional denoising algorithms use the same denoising parameters for the entire image, which may smooth the pedestrian's features during denoising, affecting the accuracy of pedestrian recognition.

The application discloses an optical lens, a camera module and an electronic device. The optical lens comprises a first lens group and a second lens group arranged from an object side to an image side, the first lens group has positive refractive power, the second lens group has negative refractive power, and the first lens group and / or the second lens group is a focusing lens group. In the focusing process of switching the optical lens from a long shot to a close shot, the distance between the first lens group and the second lens group increases, and the effective focal length of the optical lens decreases. The optical lens satisfies the relationship F2 / EFL >-5, wherein F2 is the focal length of the second lens group, and EFL is the effective focal length of the optical lens. The optical lens has the characteristics of long focal length, micro distance and large aperture, and can be miniaturized.

1. The name of the design product: close-range coding terminal device (2.0). 2. The use of the design product: the product is used for coding identification in the application scenario of vehicle automatic driving. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: rear view.

The application discloses a kind of dynamic ring vision splicing method and system based on image overlap area feature perception, method includes based on the depth information and the feature information dynamic planning splicing path;According to the splicing path of planning, image splicing is executed, including the multi-band image fusion based on depth perception.The present application extracts local and global features of the overlapping area and calculates pixel-level depth information, constructs a geometric transformation model optimized by feature-depth combination, so that panoramic stitching can accurately identify the outline of close-range objects, and dynamically plan the optimal stitching line to bypass prominent objects;By searching the distance splicing template based on the real distance and automatically unifying the focal length of each camera, the field of view range is kept consistent;Through the multi-scale depth estimation network guided by features and joint reprojection error optimization, high-precision registration can still be maintained in scenes with few feature points or low overlap areas.

The embodiment of the invention provides a lens assembly, a camera module and electronic equipment, and relates to the technical field of cameras, and the lens assembly can comprise a first lens unit and a second lens unit which are arranged from an object side to an image side along an optical axis, and the second lens unit can move towards the image side along the optical axis. Therefore, the lens assembly can be switched between the infinity state and the micro-distance state, and the lens assembly can realize the telephoto and micro-distance functions. And the focal power of the first lens unit and the second lens unit and the focal power of the first lens closest to the object side are reasonably distributed, and the light incoming amount is increased, so that the lens assembly has the characteristic of large aperture, the macro function of long-focus telephoto, high definition and high magnification is realized, high-quality imaging of a long-shot scene is ensured, and the lens assembly has a wide application prospect. And high-definition imaging of a close-shot scene is realized.

The invention discloses a super-vision-field 3D stereoscopic double-shooting technology system (hyper3D for short), and belongs to the technical field of 3D image double-shooting. Aiming at the problems of insufficient three-dimensional depth caused by a fixed baseline parameter (6-6.5 cm) and parallax distortion brightness loss and dizziness caused by a 2D-to-3D technology in the traditional 3D double camera, the invention provides a technical system based on cooperative displacement of double camera units and later parameter control. Through horizontal lateral displacement (12-30 cm) and angle rotation (1-3 degrees) of a left shooting unit, an optical parameter control module with the forced focal length larger than or equal to 50 mm is combined, the double-shooting vision field is expanded by 20%-50%, a later-stage reverse displacement compensation algorithm (close shot 1-3%, middle shot 3-5% and long shot 5-10%) is utilized, a picture body is aligned to the natural focusing position of human eyes, and the focusing effect is achieved. The high-immersion imaging effect that the three-dimensional layering sense is enhanced by 2-3 times and distortion is avoided is achieved. Technical parameters of the scheme are independent, mutual conversion with a traditional 3D video cannot be achieved, content quality is ensured from the source, and the method is suitable for the fields of movies and televisions, VR, element universe and the like.

The present invention discloses a method for on-site rapid calibration of a large field-of-view camera applicable to complex working conditions, including: on-site measurement scene arrangement, camera placement and adjustment, left camera acquisition of cross calibration target images, decoding of cross calibration target images, orientation of cross calibration target images, optimization of camera internal parameters, calibration of right camera internal parameters, synchronous acquisition of a frame of measurement scene images by the left and right cameras, feature point detection, feature point matching, solution of initial external parameter values, total station measurement of point distances, correction and optimization of camera external parameters, and export of calibration data. The present invention combines close-range photogrammetry technology, binocular stereo vision technology and total station ranging technology, uses a small-sized cross calibration target and the distance between a pair of feature points that are far apart in the measurement scene obtained by the total station as a flexible calibration target, realizes the calibration of a large field-of-view camera under complex working conditions, and can complete the on-site rapid calibration of the internal and external parameters of a large field-of-view camera with a field of view of dozens of meters to hundreds of meters within a few minutes.

The utility model discloses a slag field earth-rock volume measuring and calculating device based on close-range photogrammetry, which comprises a base, a pull rope connected to the lower side of the middle of a mounting plate, a protective shell connected to the lower end of the pull rope, a camera rotatably connected in the protective shell, an adjusting mechanism arranged on the rear side of the camera, and a detection hole arranged on the lower side of the protective shell. A detection hole is formed in the outer shell, an alignment rod is inserted in the detection hole and perpendicularly fixed to the middle of the upper side of a floating plate, the floating plate floats at the bottom of the outer shell, and transparent observation windows are arranged on the bottom and the side face of the outer shell. And when the conductive ring group is not connected, it is indicated that the current angle of the camera is not accurate, an image at the moment cannot be recorded, shooting is carried out again after the angle is adjusted through the unmanned aerial vehicle, and the accuracy of the image is improved.

The invention relates to a method (100) for processing near and far images in an image acquisition system (40) of a vehicle (5), comprising: - Providing (101) at least one near- and at least one far-field image, each depicting an environment (6) of the vehicle (5), - Determine (102) at least one relevant image area of ​​the at least one distant image based on an evaluation of the at least one near image, - Initiating (103) processing of the at least one remote image limited to the relevant image area.

The invention discloses a Cesium-based environment map making and material mapping method, which is combined with a self-defined Shader dynamic mapping method, and comprises the following steps of: firstly, making a building model and making a light and shadow texture map to generate a GLB model; and the HDR camera is adopted to carry out long-range environment map shooting to obtain an HDR long-range environment map. Building models, shadow texture chartlets and HDR long-range environment chartlets are arranged, a three-dimensional scene is set up in three-dimensional modeling software, and a scene camera is used in the set-up three-dimensional scene to render a close-range environment chartlet around each building. The GLB model, the close-range environment map around the building and the HDR long-range environment map of the output result are used, a three-dimensional scene is built in a Cesium three-dimensional engine, and finally, the building surrounding environment map is mapped to the GLB model by using a CustomShader.

The invention discloses a multi-view stereo matching optimization method based on multi-view geometry. The method comprises the following steps: (1) acquiring wide baseline images of three frames of target areas and camera positioning data of the wide baseline images by using a monocular vision inertial positioning system; (2) carrying out stereo correction on the acquired three-frame wide baseline image by adopting a multi-view geometric method; (3) performing depth estimation on the corrected three frames of images by adopting an SGM algorithm; (4) calculating geometric constraints among the depth maps by adopting a multi-view geometric algorithm; and (5) a multi-view stereo matching optimization algorithm based on multi-view geometric constraints and an improved SGM algorithm is adopted, and the problem of close-shot shielding existing in wide-baseline stereo matching is improved while wide-baseline stereo matching large-range region high-precision measurement is ensured. The method can be used for solving the problem of serious close-range occlusion in a wide-baseline stereo matching algorithm, the accuracy of wide-baseline depth estimation is improved, and accurate surveying and mapping of large scenes with close-range occlusion such as runways are facilitated.

The utility model discloses a synchronous imaging teaching instrument with multiple cameras and multiple display screens, and belongs to the field of medical teaching instruments. Comprising a supporting seat, a host, a first lens module, a second lens module, a first display screen and a second display screen, the host, the first lens module, the second lens module, the first display screen and the second display screen are all installed on the supporting seat, and the first lens module, the second lens module, the first display screen and the second display screen are all in communication connection with the host. Compared with the prior art, the first lens module can be used for long-shot shooting, and the second lens module can be used for close-shot shooting, so that students can conveniently watch a panorama of anatomy teaching and can conveniently watch details of anatomy teaching; meanwhile, the first display screen and the second display screen can display shot pictures of the first lens module and the second lens module in a split-screen mode, and can shoot and display panoramic and local key parts of anatomy teaching at the same time.

The invention relates to an optical lens, a camera module and electronic equipment. The optical lens comprises a first group, a second group and a third group, and the focal power of the first group is opposite to that of the second group. The first group includes at least two lenses. The second group includes at least one lens. The third group includes light folding elements. The light emitted from the first group sequentially passes through the second group and the light folding element, and the light is reflected for multiple times in the light folding element. The first group is a first group, the second group is a focusing lens group, and in the focusing process of the optical lens from long shot to close shot, the distance between the first group and the second group is increased, and the distance between the second group and the third group is reduced. The optical lens meets the following condition: FOV is less than 50 degrees, FOV is a field angle of the optical lens in the first focusing state, and the working object distance in the first focusing state is the maximum working object distance of the optical lens. The optical lens provided by the embodiment of the invention has the characteristic of miniaturization, can realize shooting in a long-shot state and a micro-distance state, and has good imaging quality.

The utility model discloses a focusable optical system and endoscope, focusable optical system includes first lens group, second lens group and third lens group, have negative focal power, positive focal power and positive focal power in proper order, by moving the second lens group along the optical axis, realize the switching between wide angle observation state and close shot observation state. The second lens group comprises a positive meniscus lens, the focal power of the positive meniscus lens can be small, and large variation of aberration and large variation of a main optical angle can be avoided. The third lens group comprises at least two lenses, two continuous lenses arranged close to the image surface are negative lenses, light can be gradually deflected in the direction away from the optical axis, the main light angle theta of the optical system can be increased, and the main light angle theta of the optical system can meet the requirement that theta is larger than or equal to 18 degrees and smaller than or equal to 34 degrees. Therefore, the focusable optical system provided by the utility model can obtain a larger main light angle, and can better correct aberration.

The embodiment of the present application proposes a hemispherical camera, including a base, a body shell, a high beam module and a lens assembly. The body shell is mounted on the base. The high beam module is arranged on the side wall of the body shell, and the high beam module includes a plurality of LED lamps with focusing cups. The lens assembly is arranged on the body shell, and the lens assembly includes a lens and a low beam module fixed on the lens, the low beam module is configured to illuminate the close-up area of ​​the lens shooting angle, and the high beam module is configured to illuminate the middle view area and the distant view area of ​​the lens shooting angle. The hemispherical camera in the embodiment of the present application is provided with relatively independent high beam module and low beam module on the body shell, respectively. Compared with the hemispherical camera in the related art, the brightness of the illumination of the hemispherical camera in the embodiment of the present application will not be weakened when illuminating the distant view area.

The large-scale scene organization and dynamic scheduling method based on two-dimensional Gaussian sputtering provided by the present invention relates to the technical field of scene rendering, and includes: dividing a large-scale scene into several sub-scenes, where there is an overlapping area between two adjacent sub-scenes; recursively creating bounding boxes using the median partitioning algorithm to perform spatial segmentation on the sub-scenes and generate a BVH tree; dynamically selecting nodes in the LOD tree to generate an LOD structure according to the granularity of the camera distance to the corresponding node to achieve the scene performance of the close view or the medium view; based on the LOD tree, using the breadth-first traversal of the binary tree method, generating a hierarchical level of detail HLOD using a preset proportion of the leaf nodes of the LOD tree to achieve the scene performance of the distant view; merging several sub-scenes; and being able to improve the rendering efficiency while maintaining the visual quality.

The present application relates to the technical field of lighting control, in particular to a landscape lighting dynamic control method based on environment perception data, the present application collects the reflected light radiation of the lighting object at the preset observation position, and compares the reflected light radiation with the reference radiation to obtain the reflected light radiation deviation rate; the surface optical state of the lighting object is identified according to the time series change trend of the deviation rate, and the intermediate state hosting mode is entered during the state transition; after completing the surface optical state switching, the corresponding projection adjustment parameter is obtained by searching the state-projection geometric relationship database; at the same time, hierarchical power compensation control is performed based on the brightness contribution proportion of the near, middle and far layers to generate device control instructions. The method can improve the adaptability of landscape lighting to environmental changes, enhance the visual level coordination and lighting stability.

The application provides a kind of complex scene under the method for detecting occlusion and fuzzy face, it is related to near view face target identification and detection technical field.The application is repaired to the face image of occlusion and fuzzy by using the GFP-GAN of generating face priori, then the face detection of face image after repair is carried out using improved MTCNN.Improved MTCNN introduces depth separable convolution in model network, so that the amount of calculation is reduced, the model detection time is reduced, and the model detection speed is accelerated;Improved residual attention module combined with Involution convolution is introduced in the R-Net network of the model, the accuracy of face detection is improved through the module, while reducing the size and amount of calculation of the model, the detection accuracy is ensured;Finally, the Dropout layer is introduced at the full connection layer of the network, the problem of overfitting is alleviated, to a certain extent, the effect of regularization is achieved, a kind of weighted post-processing algorithm Confluence is used to screen the face detection frame generated by O-Net network, and the accuracy of face detection frame is improved.

This application discloses an optical lens, a camera module, and an electronic device. The optical lens includes a first lens group and a second lens group that are arranged from an object side to an image side, where the first lens group has positive focal power, the second lens group has negative focal power, and the first lens group and / or the second lens group are / is focusing lens groups / a focusing lens group. In a focusing process in which the optical lens switches from a long shot to a close-up shot, a spacing between the first lens group and the second lens group increases, and an effective focal length of the optical lens decreases. The optical lens satisfies a relational expression: F2 / EFL>−5, where F2 is a focal length of the second lens group, and EFL is the effective focal length of the optical lens.

The invention belongs to the technical field of computer vision and deep learning. The invention provides an infrared image generation network based on a GAN and a deep attention mechanism. The embodiment of the invention designs an infrared image generation network for target perception segmentation, an original visible light image is divided into a target area (human and vehicle) and a background area by adopting an image segmentation network, two generators are respectively adopted for processing, and a segmentation image obtained through the segmentation network is used as intermediate input to guide image generation. Meanwhile, image depth estimation is further fused into the model, an image depth map is used as an attention gating mechanism, generation of close shot and long shot can be effectively distinguished by the network, and for detail information, details of the close shot part are expected to be rich, and unimportant details of the long shot part are expected to be sparse. And the fidelity of the generated image is further improved, and the method has extremely high accuracy in the aspect of a generation task and has considerable practical value.

The invention discloses an AF module far and near scene linear region distribution detection method and system and a terminal, and solves the problem of burning failure caused by the fact that far and near scene codes of a product are not distributed in a motor linear region when far and near scene codes are burned in advance, and the method comprises the steps: respectively selecting distances corresponding to two strokes in the total stroke of a far scene and a near scene, and obtaining a middle scene stroke; selecting a corresponding medium-scene distance from the depth-of-scene table according to the medium-scene travel to serve as a test environment; performing center MTF peak value test on the middle-scene distance and the far-and-near-scene distance to obtain corresponding code values; respectively calculating the sensitivity of the voice coil motor according to the stroke value and the code value, judging whether the far and near scenes of the AF module are in a linear region or not, and adjusting the AF module according to the judgment result. Whether far and near scene codes are respectively distributed in a linear area or not is judged in advance in a station before PDAF burning, and corresponding measures can be taken for adjustment according to a result, so that the burning yield of a subsequent PDAF burning station is improved.

Figure 1.7 is an enlarged view of part A in Figure 1.1. Figure 1.9 is an enlarged view of part B in Figure 1.8. Figure 1.11 is an enlarged view of part C in Figure 1.10. Figure 1.13 is an enlarged view of part D in Figure 1.12. Figure 1.15 is an enlarged view of part E in Figure 1.14. Figure 1.17 is an enlarged view of part F in Figure 1.16. Figure 1.19 is an enlarged view of part G in Figure 1.18. Figure 1.21 is an enlarged view of part H in Figure 1.20. The design in question relates to a tire, and in particular to the tread pattern of a tire. 1.1) Front view; 1.2) Rear view; 1.3) Left side view; 1.4) Right side view; 1.5) Top view; 1.6) Bottom view; 1.7) Close-up view; 1.8) Perspective view; 1.9) Close-up view; 1.10) Perspective view; 1.11) Close-up view; 1.12) Perspective view; 1.13) Close-up view; 1.14) Perspective view; 1.15) Close-up view; 1.16) Perspective view; 1.17) Close-up view; 1.18) Perspective view; 1.19) Close-up view; 1.20) Perspective view; 1.21) Close-up view

The present application relates to a kind of pronghorn identification system, method, processing device and storage medium based on video tracking, the system includes: data input interface, for receiving the video or image data stream from one or more camera equipment;Data type judging module is used to judge the data type of video or image data stream based on the meta information of video or image data stream;Refinement individual analysis channel is used to accurately track the individual of pronghorn in video or image data stream, trajectory optimization and behavior analysis are carried out to near view data, and the refinement individual analysis result of pronghorn is obtained;Macroscopic group analysis channel is used to carry out population statistics and rhythm analysis to long-range or aerial data in video or image data stream, and the macroscopic group analysis result of pronghorn is obtained;Data output interface is used to output the refinement individual analysis result and macroscopic group analysis result of pronghorn, the present application can be widely applied in computer vision and intelligent monitoring technical field.