72results about How to "Solve occlusion" patented technology

The invention relates to the field of electric power inspection, in particular to an unmanned aerial vehicle-based electric power inspection method and system, the method comprises two aspects of data initialization and electric power inspection, and the data initialization completes data acquisition, inspection parameter calculation and inspection model training and compression processing; the electric power inspection can realize off-line and real-time automatic inspection in mountains and other non-signal areas by receiving the inspection parameters and the inspection model calculated and completed by the server, and the inspection has the functions of automatic route planning, obstacle avoidance reminding and inspection center keeping. Detection results of six angles in the annular direction of the tower and seven angles in the center of the top of the tower are inspected and collected for comprehensive judgment, the problem of component shielding in fault detection can be solved, meanwhile, the problem of component missing can be detected by comparing historical angle pictures, and finally, exposure abnormity repair is provided based on the historical angle pictures, so that the fault detection accuracy is improved. The inspection picture quality is optimized, and the inspection precision is guaranteed.

The invention provides an AIGC video creation method for cultural relic explanation, and belongs to the technical field of AIGC video creation. S2, cultural relic video processing; s3, audio data processing; and S4, carrying out fusion synthesis treatment. According to the method, the high-precision video processing effect and the audio authenticity processing effect can be increased, meanwhile, the interestingness can be increased by adding 3D personnel, and cultural relics can be known from multiple angles through cultural relic 3D restoration.

The invention discloses a video virtual fitting framework in a real scene based on a controllable image diffusion model. The video virtual fitting framework comprises a data preprocessing module used for constructing feature expressions of figures and clothes; the single-stage image fitting network is used for generating a virtual fitting image based on the preprocessed data and performing implicit deformation and fusion of clothes by using a cross attention mechanism; the diffusion guiding module is used for enhancing the time sequence consistency of the generated video, a self-supervision model Video MAE and a self-supervision image model DINO-V2 are introduced, the video is masked by the self-supervision model Video MAE, and a complete video is reconstructed through inter-frame information; and the long video generation module is used for generating a long video and keeping the consistency of the video, splitting the complete video into a plurality of short video clips, and reducing the jitter at the clip joints by using a time sequence common noise reduction strategy. The diffusion guiding module is designed, efficient time sequence consistent video virtual fitting is achieved through self-supervised learning, and the method has the advantages of being excellent in shielding processing, stable in long video generation and the like.

The invention relates to an electrically controlled liquid crystal 2-bit reconfigurable folding reflective array antenna, and belongs to the technical field of communication antennas. The antenna comprises a main reflective array, an auxiliary reflective array and a microstrip patch antenna array feed source located in the middle area of the main reflective array. Wherein a nylon column is arranged between the main reflective array and the auxiliary reflective array for supporting. The main reflective array comprises a plurality of reflective metasurface units which are periodically arranged and mutually independent; and the reflective metasurface units are used for carrying out polarization conversion and phase compensation on incident electromagnetic waves. The sub-reflection array comprises sub-reflection units which are arranged periodically and used for transmitting polarized waves which are not consistent with the feed source and reflecting the polarized waves which are consistent with the feed source. According to the invention, the aperture efficiency and the regulation and control precision of the antenna array are improved, and the advantages of low profile, easy integration and the like can be realized while high aperture efficiency and excellent two-dimensional scanning capability are maintained. In addition, a folded array structure is realized, and the problem of feed source shielding during antenna radiation is solved.

The present invention discloses a 2D / 3D detection and segmentation method, device and terminal equipment for orthopedic surgery registration, the method steps include: step S01. obtaining preoperative CT data and intraoperative perspective X-ray images; step S02. inputting the preoperative CT data into a 3D detection network for three-dimensional landmark detection and bone tissue segmentation, wherein the 3D image coding features of the input CT data are first detected, edge information is extracted, and decoding is performed to obtain preliminary detection and segmentation results, and the confidence map is calculated to obtain the intermediate segmentation results, which are input into a 3D prompt encoder to generate prompt information; step S03. inputting the intraoperative perspective X-ray image into a TransUNet network for two-dimensional landmark detection and bone tissue segmentation; step S04. outputting the landmark detection and segmentation results for registration. The present invention has the advantages of simple implementation method, strong real-time and interactivity, high detection and segmentation accuracy and efficiency.

The invention discloses a mixed reality eye movement interaction system and method based on dense map semantic segmentation, and relates to the field of man-machine interaction, and the system comprises a head-mounted device which integrates a communication module, a mixed reality display module, an eye movement tracking module and a three-dimensional environment perception module. The three-dimensional environment perception module obtains images and depth data of the surrounding environment and carries out three-dimensional reconstruction to obtain a dense map, the dense map is divided into different blocks, and semantic information of the blocks is marked; the eye movement tracking module emits rays to the dense map in the gazing direction, and a gazing target is determined according to the real object information of the collision point position of the rays and the dense map; the communication module communicates with the real object of the watching target according to the watching target and the label of the real object information; and the eye movement tracking module obtains a control instruction according to the eyeball movement and a preset control rule, and sends the control instruction to a real object watching the target. The problem that the immersion feeling is insufficient due to the fact that natural eye movement interaction with a real object cannot be carried out in an existing mixed reality scene is solved.

The invention discloses a viewpoint planning method based on deep reinforcement learning. The method comprises the following steps: acquiring a sheltered image and a camera attitude and obtaining state information; inputting the state information into a trained DQN network to obtain a camera motion action at the next moment; and controlling the mechanical arm to move to adjust the camera position. In order to improve the training efficiency of the DQN network, a training strategy of global trend guide learning is creatively provided. The global trend guided learning strategy refers to a method for forming a global trend by recording viewpoint areas of successfully detected fruit stems in the same scene in a learning process, and then guiding a network learning process by using the global trend. According to the invention, viewpoint planning is realized by a brand-new technical means, so that the problem of shielding of the grape picking robot in the picking operation process is solved. According to the method, the viewpoint planning time is greatly shortened under the condition that the high picking success rate is kept, and a new technical scheme is provided for the field of active vision technology application.

The invention relates to the technical field of computer vision and mechanical arm automation, and discloses a mechanical arm control method and system based on improved Swinin-Unet. Comprising the following steps: S0, acquiring an RGB color image and a depth image of a target working scene; s1, converting the depth image into a normal vector angle image, and splicing the depth image and the normal vector angle image to obtain a space image; s2, analyzing the RGB color image and the space image by using a grabbing model to obtain grabbing parameters; the capturing model is constructed based on improved Swinin-Unet, and comprises a coding part which is used for respectively extracting features of a plurality of set scales of the RGB color image and the space image and carrying out feature fusion; in the decoding part, the fusion features of the minimum scale are sampled upwards step by step to the input resolution, the fusion features of the corresponding scales are fused through jump connection, and then capture parameters are obtained through prediction; and S3, the mechanical arm is controlled to grab the target object based on the grabbing parameters. According to the method, the position information of the target object can be accurately judged in a multi-object disordered scene, and grabbing is completed.

The embodiment of the invention provides a surgical navigation positioning system and method, electronic equipment and a storage medium, and relates to the technical field of medical instruments.The system comprises a double-face tracer, a mechanical arm, a binocular camera and a controller, the double-face tracer is two-face spherical markers installed on the two sides of the tail end of the mechanical arm respectively, and the binocular camera is a binocular camera; each side comprises four spherical markers, and the double-sided tracer is placed in the field of view of the binocular camera; the binocular camera collects first position coordinates of the double-sided tracer and sends the first position coordinates to the controller; the controller determines a Z-axis included angle between a first coordinate system where the double-sided tracer is located and a second coordinate system where the binocular camera is located based on the first position coordinates; sending an adjusting instruction to the mechanical arm based on the Z-axis included angle, wherein the adjusting instruction is used for indicating the mechanical arm to adjust the position until the first coordinate system is parallel to the Z axis of the second coordinate system; and the mechanical arm drives the double-sided tracer to adjust the position according to the adjusting instruction. According to the embodiment of the invention, the efficiency of surgical navigation positioning is improved.

The invention relates to the technical field of optical fiber communication and sensing, in particular to an indoor positioning system based on distributed ultrasonic sensing, which is characterized in that a detection pulse is modulated to light for emission, the detection pulse is amplified by an erbium-doped optical fiber amplifier and then is sent to a to-be-detected optical fiber through an optical circulator; and the received Rayleigh backscattering signal is coupled with the local oscillator after being amplified, and finally demodulation processing is carried out. According to the invention, limitation of sensing distance and sensing frequency can be broken through in a traditional DAS system, effective sampling and recovery of high-frequency signals are realized, a distributed ultrasonic sensing technology is applied to indoor positioning, and a new application direction is developed for the field of real-time detection.

The utility model discloses a notching equipment for synthetic resin plate processing relates to synthetic resin plate processing technical field, including support frame, notching mechanism and adsorption mechanism, notching mechanism is provided on the support frame, adsorption mechanism includes adsorption table, vacuum generator, ball and spring, adsorption table is provided on the top of support frame, and the vacuum generator is provided on the top of support frame. The adsorption table is located below the grooving mechanism, a cavity is formed in the adsorption table, the air suction end of the vacuum generator is communicated with the cavity, an adsorption hole and a groove are formed in the top of the adsorption table, the adsorption hole is communicated with the cavity, and an electric control valve is arranged in the adsorption hole. When a plate needs to be fixed, the vacuum generator extracts air from the cavity in the adsorption table through the air suction end, so that negative pressure is formed in the cavity, then the electric control valve is opened, the negative pressure in the cavity is allowed to be transmitted to the adsorption holes, the plate is fixed, the shielding problem caused by a traditional abutting mode is effectively solved, and the engraving precision and efficiency are improved.

The invention discloses an image processing method and device and electronic equipment, and belongs to the technical field of image processing. The image processing method is applied to the head-mounted display device, the head-mounted display device comprises an infrared light source, an infrared camera, a motor and a light machine, the motor is used for adjusting the position of the light machine, and the relative position of the light machine and the infrared camera is kept unchanged; the image processing method comprises the steps of obtaining at least two frames of eyeball images of an optical machine at different positions through an infrared camera under the condition that an infrared light source is turned on; wherein each frame of eyeball image comprises at least one light spot, and the positions of the light spots in the eyeball images are at least partially different; according to the positions of the light spots in the eyeball images, fusing the eyeball images to obtain a target eyeball image; and executing a preset operation according to the target eyeball image.

The invention discloses a method for identifying kitchen waste occluded objects and judging occlusion relations, constructs an occlusion data set of kitchen waste images, and marks the visible mask, complete mask, and category of each target object therein; first, uses the occlusion data set to train an instance segmentation model and a mask restoration model; the input of the instance segmentation model is the kitchen waste image, and the output is the category and visible mask of each target object; the input of the mask restoration model is the visible mask and the kitchen waste image output by the instance segmentation model, and the output is the complete mask; secondly, the kitchen waste image to be identified and judged is used to obtain the category, visible mask, and complete mask of each target object using the trained instance segmentation model and mask restoration model; finally, the visible mask and the complete mask are combined to judge whether each target object is occluded and the occlusion relationship. The invention performs instance segmentation and mask restoration on recyclable garbage, and can obtain object category, position, occlusion relationship and complete shape information.

This invention relates to the field of energy storage technology, specifically to a base support device and a power supply system. The base support device includes a seat assembly, an electrical connection assembly, and a protective assembly. The electrical connection assembly is disposed on the seat assembly; the protective assembly is rotatably disposed on the electrical connection assembly; a transmission assembly includes a first transmission unit and a second transmission unit; the first transmission unit is disposed on the electrical connection assembly and is drively connected to the protective assembly; the second transmission unit is disposed on the electrical connection assembly and is drively connected to the protective assembly; the first transmission unit can be moved under force to drive the protective assembly to rotate away from the electrical connection assembly; the second transmission unit can drive the protective assembly to rotate close to the electrical connection assembly, and the protective assembly and the electrical connection assembly together form a shielding space. This solves the problem that existing base supports requiring an active drive source to operate the shielding mechanism for blocking electrical outlets need to be activated to block or de-block.

The invention discloses a kitchen fire hazard intelligent processing system, and relates to the technical field of smart home, the system comprises a multi-sensor module, an intelligent data analysis module and an automatic control module; the multi-sensor module is used for acquiring kitchen environment parameters; the intelligent data analysis module is used for determining whether abnormal conditions exist or not according to the kitchen environment parameters; the automatic control module is connected with the gas main valve; according to the intelligent gas stove disclosed by the invention, the intelligent data analysis module and the automatic control module are arranged, so that the fire risk caused by forgetting to close the gas stove can be prevented, and the family safety is comprehensively enhanced.

The invention discloses a human body upper limb pose estimation method and system based on an inertial position tracker and a depth somatosensory camera, and the method comprises the steps: collecting a human body upper limb skeleton point pose signal through the depth somatosensory camera, and collecting a human body elbow joint pose signal through the inertial position tracker; constructing a global coordinate system, and converting the upper limb skeleton point pose signal and the elbow joint pose signal into the global coordinate system; and acquiring elbow joint and wrist joint positions in the global coordinate system according to a double Kalman filtering method, and calculating an elbow joint angle by combining a shoulder joint position obtained according to the upper limb skeleton position posture signal. According to the method, the data of the inertial position tracker and the depth somatosensory camera are fused by adopting a dual Kalman filtering algorithm, and signal resources in multiple aspects of the sensor are fully utilized, so that the requirements of high robustness and high estimation precision in a dynamic environment are met, and meanwhile, the shielding problem of the depth somatosensory camera is solved.

The semi-transparent BIPV assembly comprises a perovskite top cell, a glass backboard and a plurality of silicon substrate cells, the perovskite top cell, the silicon substrate cells and the glass backboard are arranged in sequence from top to bottom, and adhesive film layers are arranged between the silicon substrate cells and the perovskite top cell and between the silicon substrate cells and the glass backboard; gaps exist between the adjacent silicon-based batteries, the adjacent silicon-based batteries are connected through copper alloy wires, and the outer sides of the silicon-based batteries are connected with the junction box through the copper alloy wires so as to realize electric energy output. The crystalline silicon semi-transparent BIPV assembly has the advantages of being compact in structure, convenient to disassemble and assemble, high in light utilization rate, large in effective power generation area, low in cost and the like, and solves the problems that an existing conventional crystalline silicon semi-transparent BIPV assembly is small in effective power generation area and low in assembly efficiency.

The invention belongs to the technical field of yarn detection, and particularly relates to a monofilament number detection device and method. The device comprises a tow clamp, a workbench, an operation display device, a high-speed camera shooting assembly, an electrostatic adsorption device, a pneumatic device and a master control device. The electrostatic adsorption device is arranged and mounted on the workbench; the pneumatic equipment is electrically connected with the general control device; the high-speed camera shooting assembly is installed on the workbench and electrically connected with the master control device. The master control device is installed on the workbench and used for processing and analyzing the tow images uploaded by the high-speed camera shooting assembly. The operation display device is installed on the workbench and electrically connected with the master control device. According to the technical scheme, the number of monofilaments can be accurately detected in the radial direction of tows, and accuracy is improved. Multiple components cooperate to adapt to various tow types and production environments. The device is simple in structure, convenient to operate and high in function expansibility, has a reinspection process, effectively ensures that the detection result is reliable, and assists spinning quality control and production management optimization.

The utility model discloses a car lamp module which comprises two light-emitting module units arranged in the first direction and a first lens, the first lens is arranged on the light-emitting sides of the light-emitting module units, one light-emitting module unit is a low-beam light-emitting module unit, the other light-emitting module unit is a high-beam light-emitting module unit, and the other light-emitting module unit is a high-beam light-emitting module unit. The virtual images of the light emitted by the low-beam light-emitting module unit and the virtual images of the light emitted by the high-beam light-emitting module unit are located at the focus position of the first lens. According to the car lamp module provided by the utility model, the low-beam light-emitting module unit and the high-beam light-emitting module unit are integrated, and the virtual images of the light emitted by the low-beam light-emitting module unit and the virtual images of the light emitted by the high-beam light-emitting module unit are positioned at the focal positions of the first lenses, so that the complete imaging of the light pattern can be ensured, and the problem of light shielding is avoided; therefore, the light utilization rate can be improved. Meanwhile, a shielding piece in the prior art is omitted, and the light utilization rate is increased. The utility model further discloses a car lamp.

The invention relates to an optical fiber rotary connector, in particular to a spatial dual-channel optical fiber rotary connector, and aims to overcome the defects that an existing dual-channel optical fiber rotary connector is poor in environmental adaptability or high in assembly difficulty or difficult to complete two-way signal transmission. The spatial dual-channel optical fiber rotary connector comprises a rotor sleeve, a stator shell, a first optical assembly and a second optical assembly, the rotor sleeve and the stator shell are coaxially and rotatably connected, and the first optical assembly comprises a first input assembly, a beam expanding assembly, a beam shrinking assembly and a first output assembly which are sequentially arranged along an optical path; the first input assembly and the beam expanding assembly are arranged in the rotor sleeve, input a first optical signal and form a collimated hollow light beam; the beam shrinking assembly and the first output assembly are arranged in the stator shell, and the collimated hollow light beam is shrunk and output; and the second optical assembly is arranged in a second input assembly and a second output assembly in the stator shell and is used for transmitting a second optical signal.

The present application discloses a passenger flow counting method, device, equipment and medium based on image generation, which relates to the field of passenger flow counting. The method comprises: detecting and counting complete pedestrians in a passenger flow area image to obtain a foreground passenger flow; removing complete pedestrians in the passenger flow area image to obtain an occluded image; detecting occluded objects in the occluded image, and removing the occluded objects in the occluded image to obtain a remaining image area; inputting pixel values ​​in the remaining image area into an image generation model, reconstructing the area where the occluded objects are located, and determining the occluded image after the occluded objects are reconstructed as a reconstructed image; the image generation model is constructed based on a deep learning model; detecting and counting complete pedestrians in the reconstructed image to obtain the passenger flow in the occluded area; determining the final passenger flow in the target scene according to the foreground passenger flow and the passenger flow in the occluded area. The present application can effectively solve the occlusion problem and improve the accuracy of passenger flow counting.

The embodiment of the present invention provides an image processing method and device, a storage medium, and an electronic device, the method comprising: determining M image regions of the first image based on an edge detection map of the first image, wherein M is a positive integer; determining a texture feature value of the first image based on the pixel value in each of the M image regions; and determining whether the first image is blocked based on the texture feature value of the first image. The present invention solves the problem of determining video blockage in the related art, thereby achieving the effect of improving the accuracy of video blockage detection.

The invention discloses a rock slag feature recognition method based on multi-sensor fusion. The method comprises the steps that 1, an industrial camera and a laser radar synchronously collect rock slag images and point cloud data; 2, an improved YOLO11n-seg network is constructed; 3, performing feature extraction on the rock slag image through an improved YOLO11n-seg network; 4, constructing total loss; 5, training the improved YOLO11n-seg network on the basis of the total loss; 6, segmenting a subsequent rock slag image to be processed to obtain a rock slag area segmentation map; 7, projecting the original point cloud data to obtain the point cloud of each rock slag area; and obtaining the rock slag volume based on the point cloud of each rock slag area. The method is simple in step, instance segmentation of the rock slag target area is achieved through the improved YOLO11n-seg network, the contour shape and the volume of the rock slag are obtained, and therefore a basis is provided for parameter optimization of follow-up TBM tunneling.

This utility model relates to the technical field of optical device coating equipment, and particularly to a coating fixture for an open-aperture lens. The open-aperture lens has a coating surface at one end along the axial direction and a positioning part at the other end. The open-aperture lens also has a through hole along the axial direction. The coating fixture includes a fixture body and a fixing member. A positioning groove is provided at the bottom of the fixture body, and a through hole communicating with the positioning groove is provided at the top of the fixture body. The positioning groove is used to position the positioning part. The fixing member passes through the through hole and the through hole in sequence. One end of the fixing member has a limiting part that can be engaged in the through hole, so that the positioning groove positions the positioning part. The coating fixture for the open-aperture lens provided by this utility model can reliably fix and accurately position the lens without obstructing the coating surface.

The invention provides a virtual camera control method and device, electronic equipment and a computer storage medium, and the method comprises the steps: determining a target cube grid of an entity according to the offset of the entity in game control input data and the coordinates of the world where the entity is currently located; determining a first target cube lattice of the virtual camera according to the offset of the entity and the current world coordinate of the virtual camera; if the mark of the first target cube lattice of the virtual camera in the scene space set is shielded, selecting a cube lattice closest to the first target cube lattice of the virtual camera in the virtual camera following position set according to the target cube lattice of the entity as a second target cube lattice of the virtual camera; and frame-by-frame interpolation processing is performed on the virtual camera from the current world coordinate, and the virtual camera is transited to a second target cube grid of the virtual camera. Therefore, the problem that a lens is blocked or interspersed when the virtual camera moves along with an entity (an object, a role and the like) is effectively solved.

The invention discloses an industrial robot image processing method based on image fusion, relates to the technical field of image fusion, effectively solves the problem of object occlusion in storage through multi-angle real-time image acquisition and a deep learning hierarchical fusion algorithm, and improves the image fusion accuracy by evaluating the visibility coefficient Zjy of an occlusion area. Identification errors caused by shielding are avoided; the image denoising coefficient Ncj is calculated in real time, so that the image quality is ensured, and the subsequent recognition accuracy is improved; for shape and color diversification, a color diversification characteristic coefficient Ssc and a shape complexity coefficient Xtx are calculated, and a color histogram and geometric morphology analysis are combined, so that the recognition capability of a complex object is comprehensively improved; through comparison with the preset threshold value, the diversification conformity of the object is judged, objects of various shapes and color combinations are effectively recognized, the data labeling requirement is reduced, and the recognition efficiency and accuracy of an automatic warehousing system are improved.

The utility model discloses a kind of light guide structure for improving light guide uniformity, including light guide body, LED light source;The light guide body includes light guide tooth, light guide diffusion pattern surface, light guide diffusion pattern surface every pattern has two direction adjustment diffusion surface.The utility model uses the form of corn kernel pattern at light guide near light end, the diffusion of every pattern can be controlled, the diffusion size of two directions of corresponding pattern is adjusted to solve the problem of bright spot and dark area generated in simulation design process, simultaneously, since the controllable corn kernel diffusion can also effectively solve the shielding problem of dark area, can be reduced from 40mm shielding to 35mm, improve light guide utilization rate, also can improve the area size of window, meet the modeling design requirement.

The utility model provides a material rack which comprises a rack body, the rack body comprises a first face and a second face which are opposite, and the first face is provided with an operation area; the material accommodating part is formed between the first surface and the second surface and is used for accommodating a plurality of material rolls; the recycling part is arranged below the material containing part and used for recycling waste of the material coil. Wherein the waste is introduced into the recovery part from the lower part of the operation area. Due to the fact that the material rack is provided with the material containing part and the recycling part, the material roll is placed in the material containing part, the recycling part is located below the material containing part, waste of the material roll can enter the recycling part along with use of the material roll, and the problem that the waste naturally droops, and the lower-layer material roll is blocked is solved; in addition, due to the fact that the first face of the frame body is provided with the operation area, operation of operators is facilitated, and production efficiency can be improved.

A main body tube is provided with an axially through catheter inner cavity, an optical fiber body comprises at least one optical fiber core, the optical fiber body is arranged in the axial direction of the main body tube, the optical fiber body is used for generating image data after shape demodulation, and the image data is configured to be readable in an image display device. According to the microcatheter and the catheter system, the microcatheter is combined with the inventive concept that the optical fiber technology is used for navigation and positioning, the optical fiber is added into the microcatheter to achieve the more accurate positioning function, the optical fiber is used for sensing the position distribution of the microcatheter in the three-dimensional space, a three-dimensional image is formed, and the view field is widened. Therefore, the problem of sight blocking in the operation can be effectively solved, and clinicians can judge and operate conveniently. The application of the optical fiber technology in the micro-catheter can solve the technical problems that the existing clinical radiation medical imaging technology is large in radiation damage, and the two-dimensional imaging is not favorable for judging the position of the catheter.