25results about How to "Improve stitching quality" patented technology

The invention discloses a method for improving video stitching quality based on an improved optimal stitching line algorithm, relates to the field of application of vehicle-mounted cameras, the field of panoramic image fusion, the field of reversing parking and the field of unmanned driving, and solves the common problems of image distortion, obvious stitching lines, discontinuous stitching areas and the like in the video stitching process. An improved optimal suture line algorithm is selected, and the method can automatically find the optimal overlapping area and the optimal suture line position, so that the distortion and discontinuity phenomena in the splicing process are reduced. By improving the splicing effect, the method can be widely applied to the fields of panoramic video production, virtual reality and the like, and the user experience is remarkably improved.

The invention discloses a loader front image splicing method based on splicing seam position calculation, and belongs to the technical field of image processing. The method comprises the following steps: 1) determining the position of a bucket in an image based on an angle sensor and two cameras; 2) determining a main image and an auxiliary image; 3) determining an overlapping region of the main image and the auxiliary image; 4) determining a splicing seam position of the main image, and estimating an initial splicing seam position of the auxiliary image; 5) comparing the features of the main image and the auxiliary image, solving registration parameters, and obtaining an optimal abutted seam; and 6) completing fusion and splicing of the two images. According to the brand-new image seam splicing method, the upper camera and the lower camera are combined, the images, blocked by the bucket, in front of the loading machine are observed, and perfect splicing of the blind areas blocked by the bucket in front is achieved based on seam splicing position calculation. Meanwhile, the boundary of the bucket is reserved in the splicing process, so that a transparent bucket can be formed in the image, and the visual field and the working efficiency of an operator are obviously enhanced.

The utility model discloses an auxiliary device for splicing door and window frames, which comprises a supporting table, a positioning and clamping assembly is arranged on the upper side of the supporting table, a reciprocating beating assembly is arranged on one side of the positioning and clamping assembly, the positioning and clamping assembly comprises a positive and negative screw rod, and the outer surface of the positive and negative screw rod is in threaded connection with a first clamping strip. A first motor is fixedly installed at one end of the forward and reverse screw rod, a second motor is arranged on one side of the forward and reverse screw rod, a connecting disc is fixedly installed at the output end of the second motor, a connecting rod is rotationally connected to the outer surface of the upper side of the connecting disc, a U-shaped frame is rotationally connected to one end of the connecting rod, and a second clamping strip is fixedly installed on the outer surface of one side of the U-shaped frame. And by starting the second motor and the first motor and cooperating with the reciprocating beating assembly, automatic splicing of the door and window frame is achieved, the practicability of the auxiliary splicing device is greatly improved, and then the splicing efficiency and quality of the door and window frame are improved.

This invention discloses a panoramic image stitching and restoration system and method based on feature matching, relating to the field of image processing technology. The system includes capturing images and collecting image capture information; numbering the images based on the collected image capture information; matching feature points of two adjacent numbered images to obtain image feature information; mapping the image numbers and image feature information to a Cartesian coordinate system in stages; setting a triple planning mechanism based on the staged mapping information; planning the image stitching order based on the triple planning mechanism; performing image stitching processing according to the planned image stitching order; and performing restoration processing on the stitched panoramic image. This achieves automated planning of the optimal image stitching order, reduces the accumulation of image stitching errors, and minimizes the impact of poor image stitching quality on subsequent image restoration effects.

The invention discloses cutting equipment for a profile frame and a forming method for a supporting frame for doors and windows. The cutting equipment comprises a rack, a feeding positioning mechanism, a cutting positioning mechanism, a first cutting mechanism and a second cutting mechanism. A positioning sliding groove and a shearing groove are formed in the top of the cutting positioning mechanism, the shearing groove comprises first inclined faces arranged in a mirror symmetry mode, and a second inclination angle is formed between each first inclined face and the vertical direction. The first cutting mechanism comprises a cutting saw blade and a rotary transmission wheel set, saw teeth of the cutting saw blade are obliquely arranged downwards, the cutting saw blade is wound on the periphery of the rotary transmission wheel set, the cutting saw blade is driven by the rotary transmission wheel set to transmit, and a first inclination angle is always kept between the cutting saw blade and the vertical direction so as to cut the rib plate; the second cutting mechanism comprises a second lifting driving mechanism, a movable die and a horizontal telescopic driving mechanism, the movable die and the horizontal telescopic driving mechanism are driven by the second lifting driving mechanism to be close to or away from the cutting positioning mechanism, the movable die is provided with a stamping knife, and the horizontal telescopic driving mechanism is used for driving the stamping knife to reciprocate along the shearing groove and shear the profile body; the cutting track of the cutting saw blade intersects with the cutting track of the stamping knife. The splicing effect of the supporting frame can be improved, and meanwhile the assembling efficiency of the supporting frame can be improved.

The invention discloses a plate welding part detection method based on improved image stitching and a deep learning network. The method comprises the following steps: performing image stitching fusion on plate welding part images shot by a camera group at the same time through a fusion method of an improved SURF algorithm; the structure of the YOLOv8 neural network is improved, and the improved YOLOv8 neural network used for target detection and recognition of the plate welding part is formed; carrying out integration to obtain an NWD-Inner-IoU loss function; and importing the spliced and fused image data set into an improved YOLOv8 neural network, training the structure-improved YOLOv8 neural network in combination with an NWD-Inner-IoU loss function, and applying the trained model to detection of plate welding parts. According to the method, the detection accuracy and the detection precision of the plate welding part and the robustness of the model can be effectively improved, the practicability is enhanced, the method can be deployed in edge equipment, and the performance requirement of the required equipment can be reduced.

The invention discloses a radar holder adaptive superhelix control method based on RMSE speed planning, and belongs to the technical field of laser radar mapping and motion control. According to the method, the problem that the laser radar dynamic mapping effect is unstable due to the fact that the splicing effect of the point cloud data is not considered and the robustness is poor in an existing radar holder control method is solved. Firstly, feature points in two adjacent frames of point cloud data are roughly matched, then fine matching is completed based on generalized ICP, and RMSE is calculated based on a fine matching result; designing a rotating speed regulation rate according to a fitting equation of the rotating speed of the radar pan-tilt and the RMSE, and observing the disturbance suffered by the radar pan-tilt by using a super-spiral expansion state observer; and finally, designing a self-adaptive super-spiral controller based on a disturbance observation result and a rotating speed regulation rate based on RMSE, driving a radar holder to rotate by using the output of the controller, and continuously scanning to obtain point cloud data of a next frame until the whole scanning process is finished. The method can be applied to a laser radar dynamic mapping task in a complex large scene.

The utility model relates to a kind of for precast concrete building component splicing mould, including splicing joint outer fixed plate, splicing joint inner filling fixed part, the splicing joint outer fixed plate is oppositely set, the splicing joint inner filling fixed part is provided with several groups, several groups splicing joint inner filling fixed part are arranged between opposite side splicing joint outer fixed plate, and the both ends of splicing joint inner filling fixed part are fixed on splicing joint outer fixed plate.This mould is through the unique structural design, realizes the dense filling and stable fixation of splicing joint, improves the splicing quality and the structural performance of building, strengthens the effect of splicing, simultaneously, using internal reinforcing member as the fixed structure of external formwork component, external formwork can be reused, reduce cost, it is suitable for use.

The present invention belongs to the technical field of plate welding, and specifically relates to an integrated plate automatic cutting and welding device, including a feeding device, an input positioning device, a cutting and welding auxiliary device, a roller conveyor device with a liftable anvil, an integrated cutting and welding device for cutting and splicing plates, and a traction device for driving the plates to move; the feeding device, the input positioning device, the cutting and welding auxiliary device, and the roller conveyor device are arranged in sequence along the moving direction of the plates; the integrated cutting and welding device includes a working end and a driving unit for driving the working end to move above the cutting and welding auxiliary device and the roller conveyor device. The integrated plate automatic cutting and welding equipment provided by the present invention can improve the splicing and welding quality, reduce the process difficulty, improve the processing efficiency, and save energy consumption.

This utility model specifically relates to a fiberglass cloth splicing device, belonging to the field of fiberglass cloth processing technology. It includes a splicing table and a fiberglass cloth sheet disposed above the splicing table. It also includes a hot pressing component located above the splicing table, a lifting frame located behind the splicing table and fixedly connected to it, a connecting frame slidably mounted on one side of the lifting frame, a transverse frame fixedly mounted on one side of the connecting frame, a hot pressing frame slidably mounted inside the transverse frame, and a tension frame symmetrically hinged at the lower end of the hot pressing frame. The hot pressing component is located on the lower surface of the tension frame, and a tension element is disposed inside the lower end of the tension frame to provide tensile tension to the fiberglass cloth sheet during hot pressing. Through the tension element and the hot pressing component, this device can continuously stretch the splicing area while hot pressing the fiberglass cloth sheet, thereby improving the hot pressing effect and reducing the possibility of hot pressing wrinkles.

The present invention discloses a corner forming machine, which is suitable for forming a foamed material plate into a corner product, including a hot pressing device and a forming device; the hot pressing device includes a first driving mechanism and a hot pressing die, the output end of the first driving mechanism is connected to the hot pressing die and drives the hot pressing die close to the material plate, hot pressing the material plate to form a template that can be folded into a corner; the forming device includes a second driving mechanism and a forming die, the output end of the second driving mechanism drives the template into the forming die, and forms the template into the corner. The corner forming machine of the present invention can be produced automatically, has high production efficiency and good forming quality. In addition, the present invention also discloses a corner forming method.

The invention provides a floor support plate typesetting method under a supporting beam non-coplanar condition, which comprises the following steps of: firstly, establishing a floor elevation control net through an NURBS curve, covering the floor elevation control net with an NURBS curved surface to form a smooth and continuous surface model of a floor, and then, carrying out high-quality plane unit division on the floor by utilizing a finite element grid division technology, so as to form a smooth and continuous surface model of the floor. And finally, extracting a unit number, a node number and a node coordinate of each unit plate according to the grid division scheme to serve as a typesetting and splicing scheme of the floor support plate. According to the method, floor support plate abutted seams and elevation differences caused by manual typesetting are reduced, material waste is reduced, the quality problems such as slurry leakage in the concrete pouring process of the floor support plate are avoided, and it is guaranteed that a curved-surface-shaped floor formed through pouring is continuous and attractive.

The invention relates to the field of data modeling, in particular to a three-dimensional modeling method and system based on surveying and mapping multi-source data. The method comprises the following steps: analyzing acquired surveying and mapping multi-source data, dividing a target area, and generating a three-dimensional disturbance layering graph; analyzing the three-dimensional disturbance layering atlas, and extracting disturbance level attributes and modeling complexity data; analyzing the disturbance level attribute and the modeling complexity data to generate a modeling subarea attribute set; constructing a modeling boundary relationship and a modeling priority based on the modeling subarea attribute set, analyzing the modeling boundary relationship in combination with the modeling priority, and generating a modeling control structure; analyzing the target area based on the modeling control structure to generate candidate point cloud data; processing the candidate point cloud data to generate a modeling candidate data set; and generating a three-dimensional modeling fusion result according to the modeling control structure and the modeling candidate data set. According to the invention, the continuity and regional adaptability of three-dimensional modeling can be improved.

The present invention provides a long-distance single-photon radar large field of view high-resolution rapid imaging method, which can be applied to the field of laser radar technology. The method includes: constructing a long-distance single-photon large field of view scanning imaging system based on a two-dimensional dual-axis turntable and an array single-photon detector, and configuring the long-distance single-photon large field of view scanning imaging system; using the long-distance single-photon large field of view scanning imaging system to control the array single-photon detector to scan the panoramic pixel points of the target object in a raster-type continuous scanning manner by setting the motion mode of the two-dimensional dual-axis turntable, and obtaining the large-field panoramic single-photon detection data of the target object; performing real-time data compression processing based on resampling on the large-field panoramic single-photon detection data to complete the image restoration of the target object, and obtain the real image of the target object. The method provided by the present invention can adapt to different imaging scenes, and has the ability to detect large imaging fields and rapid detection capabilities.

The application relates to the technical field of computer vision, and discloses a picture splicing method based on multiple cameras. The method comprises the following steps: acquiring multiple image frames collected synchronously; correcting and aligning the images; constructing an initial pixel-level difference cost map; combining a previous frame splicing seam to generate a space-time guide weight map; fusing to obtain a dynamic optimization cost map; performing shortest path search in a limited area to determine a current best splicing seam; and generating a seamless splicing image by weighted fusion according to the current best splicing seam. By introducing a space-time guide mechanism and an edge preservation factor, the calculation complexity is reduced, the splicing continuity and visual quality are guaranteed, and the method is suitable for high-dynamic scenes such as vehicle-mounted surround view and unmanned aerial vehicle aerial photography.

The invention provides an image splicing method and device, electronic equipment and a storage medium, and relates to the technical field of microscopic imaging. Comprising the following steps: determining an overlapping region between a first cell image and a second cell image to be spliced; performing intra-region division on the overlapped region to obtain a target processing region; according to the target processing area, the pixel information of the first cell image and the pixel information of the second cell image, accumulated cost information of each pixel point in the target processing area is determined; determining associated pixel points corresponding to the pixel points according to the accumulated cost information of the effective neighborhood pixel points corresponding to the pixel points; and determining a target splicing position in the overlapping area according to the end point pixel point in the target processing area and the associated pixel points corresponding to the pixel points. Based on the method, a better splicing position can be determined, seamless splicing of the image is satisfied, cells are not cut, and the problem of repeated cells in the spliced image caused by cell movement is reduced.

The application discloses a multi-scale lining image splicing method in a tunnel detection parallax scene and a computer program product, relates to the technical field of computer vision, and solves the technical problem that a lining image splicing method has poor adaptability and is difficult to guarantee the quality of lining image splicing. The method comprises the following steps: performing feature map matching on a first image and a second image to obtain an initial global homography matrix; performing multi-scale matrix iterative optimization and global alignment feature map generation to obtain an optimal global homography matrix; performing local non-rigid adjustment on the second image through a TPS interpolation algorithm; extracting an initial mask from the registered first image and the second image, and generating a joint mask; searching for optimal seam lines in the joint mask to generate a content mask; and performing fusion processing on the registered first image and the second image based on the content mask, and outputting a panoramic image. The application has good adaptability to scenes with different parallax degrees, and the quality of lining image splicing is greatly improved.

The present application provides a panoramic image stitching method, device and panoramic camera, the method comprising: obtaining a first image obtained by scanning a photosensitive chip of a first camera along a first direction and a second image obtained by scanning a photosensitive chip of a second camera along a direction opposite to the first direction, calculating the optical flow of the overlapping area of ​​the first image and the second image, obtaining the optical flow vector of the feature point of the overlapping area, any feature point on the first image is recorded as x, and the pixel value is recorded as l(x), the feature point on the second image after being mapped by the optical flow vector is recorded as x', and the pixel value is recorded as l'(x'), and according to l(x) and l'(x'), the pixel value x after x and x' are fused is calculated c , according to the fused pixel value x c , the two images are stitched together to obtain a panoramic image. The present application can reduce the exposure difference in the overlapping area of ​​the images in the stitching of the panoramic image.

The present application relates to a kind of deburring bamboo board splicing methods, bamboo strip jacking mechanism will be several long width face up bamboo strip upward jacking, push material mechanism will be pushed to the bamboo strip on the conveying belt;Conveying belt will be sent to the bamboo strip to splicing workbench, glue spraying mechanism is sprayed to the long width face of bamboo strip;Bamboo strip rotating mechanism is from left and right two ends to clamp several bamboo strips on splicing workbench, then will be moved to the side pressure plate above, bamboo strip rotating mechanism will be rotated 90 degrees to the bamboo strip in the same direction, so that the long narrow face of bamboo strip is up, bamboo strip rotating mechanism will be placed after rotating the bamboo strip on splicing workbench;Shaping mechanism is from left and right sides to arrange several after rotating bamboo strip, so that the end face of several bamboo strips is in line;Bamboo strip rotating mechanism front and rear side pressure plate moves towards, several bamboo strips are clamped from front and rear direction and pressure is kept.The method effectively improves the splicing efficiency and splicing quality of bamboo board, and is conducive to the continuous operation of bamboo board splicing.

The invention belongs to the technical field of building template manufacturing, and particularly relates to a wood composite template processing technology which comprises the following steps: selecting wood for blanking; wood is rotationally cut into wood chips; airing the wood chips; the aired wood chips are subjected to edge trimming and splicing in sequence through an edge trimming and gluing device and then glued and bonded into battens; the battens are cut off, stacked together and pre-pressed to obtain prefabricated slabs; the prefabricated plate is sequentially subjected to red pressing, hot pressing, sizing edge cutting and edge sealing to obtain the wood composite plate, the machining sequence is changed through the technology, and the recycling purpose of wood waste is expanded; mechanical operation of wood chip trimming and splicing steps is achieved through the trimming and gluing device, the manual participation degree is reduced, the production efficiency and the wood chip splicing quality are effectively improved, and the formwork quality can be improved.

The utility model discloses a tooth punching tool for a Teflon belt. The tooth punching tool comprises an operation table and a fixing frame which extend in the same direction. A cutting die is arranged in the extending direction of the operation table. A plurality of clamping assemblies are arranged in the extending direction of the fixing frame, the clamping assemblies are matched with the operation table to clamp and fix the Teflon belt, a die pressing assembly is arranged on the side, opposite to the operation table, of the fixing frame, and the die pressing assembly can be switched between the storage position and the working position and can slide in the extending direction of the fixing frame. Thus, the Teflon belt is pressed into the cutting die through the die pressing assembly, the precision of the tooth punching position and size is ensured, the cutting position and depth are accurately controlled, the Teflon belt is fixed and pressed through the clamping assembly, the stability and accuracy of the cutting process are ensured, the problem of poor tooth punching caused by movement or dislocation of the Teflon belt is solved, and the cutting efficiency is improved. And the uniformity and consistency of tooth punching are improved, so that the splicing quality is improved, the loss of waste materials is reduced, the material cost is saved, and the production efficiency is improved.

The invention relates to an automatic bare soil covering and dense mesh splicing device which comprises a vehicle body, walking auxiliary balance devices, a splicing device, walking wheels, a cabin door and a rock falling opening, the cabin door is arranged at the top of the vehicle body, the rock falling opening is formed in the bottom of the vehicle body, the walking auxiliary balance devices are arranged on the two sides of the vehicle body, and the splicing device is arranged at the front end of the vehicle body. The splicing device comprises a supporting body, a cabling machine, a cabling track, a cabling plate and a net collecting hole, the supporting body is connected with the vehicle body, the cabling track is arranged at the front end of the supporting body, the cabling machine is arranged at the top of the cabling track, the cabling plate is arranged at the bottom of the cabling track, and a gap is formed between the cabling plate and the bottom of the cabling track; the device has the beneficial effects that the device is simple in overall structure and convenient to operate, labor is saved, construction time is saved, construction efficiency is improved, covering and splicing quality is effectively guaranteed, bare soil covering construction quality is guaranteed, construction cost is saved, and the device has a very high application prospect.

A delivery fastener applicator includes a fastener nest for receiving a fastener, a head operable to drive a staple leg of the fastener, and a rotatable anvil including an initial curved portion and a final curved portion. The rotatable anvil has a first rotational position in which an initial bending portion of the anvil is positioned below the fastener, the initial bending portion of the anvil configured to partially bend the staple leg when the leg is driven downward from the bore of the fastener lower plate by the head. The anvil is rotatable relative to the fastener nest from a first rotational position to a second rotational position. The head is configured to be operable to drive the staple leg and press the partially curved staple leg against the final curved portion of the anvil to secure the staple leg to the lower plate of the fastener.

The present invention provides a method and system for real-time image stitching of a drone, which relates to the field of image stitching technology and includes the following steps: obtaining state parameters of at least two cameras on a drone; analyzing camera state parameters and environmental parameters; obtaining real-time images from the cameras on the drone; segmenting the real-time images frame by frame to obtain frame-by-frame images, and performing stitching processing on the frame-by-frame images to obtain the frame-by-frame images after stitching processing. The present invention provides a method and system for real-time image stitching of a drone, which rapidly processes images captured by the drone camera through frame-by-frame segmentation and stitching processing of real-time images, significantly improves the efficiency of image stitching, and optimizes the errors in the stitching process using a comprehensive energy function and a gradient descent method, thereby reducing stitching errors, improving stitching quality, and generating continuous, seamless, high-quality images.