35results about How to "Complete outline" patented technology

The invention discloses a method for extracting the contour of a building by integrated images and LiDAR data. The method includes (1) pretreatment of LiDAR point cloud data; (2) establishment of a contour extraction area; (3) extraction of a building line segment; and (4) screening of the building contour. The invention can effectively extract the contour of a building with complex structure, the positioning accuracy is in proportion to in proportion to the LiDAR scheme, the rates of accuracy and integrity are high, and the robustness is favorable.

The invention discloses a video multi-vehicle outline detection method, which comprises the following steps of: drawing a virtual fence to define a monitoring area, and drawing a virtual detector to acquire parameters such as vehicle width and minimum vehicle length; obtaining a vehicle body outline set through a series of steps such as background subtraction, binarization, morphological opening operation and outline extraction; tracking a vehicle by tracking pixel blocks forming the vehicle, thus obtaining an initial outline of the vehicle; merging the outline of the vehicle by taking the initial outline of the vehicle as the basis, thus ensuring the outline of the vehicle which runs into the monitoring area to be more complete; and selecting a large outline from the residual body outline to serve as an initial outline of a new vehicle, and merging the body outline of the new vehicle, thus timely detecting the vehicle which newly runs into the monitoring area. The method solves the problems of camera shake, image interference on the two sides of a road, disconnected vehicle foreground and adhesion of outlines of adjacent vehicles.

The invention discloses an unmanned aerial vehicle visual detection method for high-voltage line insulators. The method comprises the following steps of: obtaining a high-voltage line insulator image; converting the high-voltage line insulator image from an RGB model to an HSV model; detecting a brightness component of the HSV model by adoption of a cross gradient search method, and obtaining a direction gradient and a spatial brightness similarity so as to generate a foreground image; carrying out connected domain segmentation on the foreground image by adoption of a 8-neighborhood connected domain segmentation method so as to obtain an insulator connected domain image; carrying out edge extraction on the insulation connected domain image by adoption of an LOG operator so as to obtain edge information of the image; carrying out 8-neighborhood edge classification on an image edge point by adoption of a 8-neighborhood edge tracking method; for pixel class groups obtained through the 8-neighborhood edge classification, carrying out edge linear detection on the groups, and displaying a recognized insulator profile. According to the method, complete insulator connected domain images can be obtained, so that the insulation recognition correctness is effectively improved, the operation speed is effectively improved and the calculation complexity is reduced.

The invention discloses a Bayesian wide-residual neural network-based injurious insect image recognition method. The method comprises the following steps of: 1, preprocessing an injurious insect imagerecognition data set and carrying out injurious insect edge detection on a greyscale map by using a Rich-Edge detection algorithm; 2, constructing a Bayesian wide-residual neural network (NWResNet);3, inputting an injurious insect edge image obtained in the step 1 into the NWResNet constructed in the step 2, and obtaining a total error function of the NWResNet by utilizing an injurious insect edge image training set obtained in the step 1; 4, training the network through a Block-cg algorithm by utilizing the error function obtained in the step 3; 5, updating a hyper-parameter according to the network optimized in the step 4; and 6, repeating the steps 4 and 5 to obtain a final network, and inputting an injurious insect image verification set into the network to ensure that the classification correctness is higher.

The invention provides a manufacturing method for a grid electrode, including the following steps: a substrate is provided, at least one grid electrode material layer is formed on the substrate; a pattern mask layer is formed on the grid electrode material layer; the transverse size of the pattern mask layer is microcopied by the wet-type etching process; the pattern mask layer is taken as the mask to remove a part of the grid electrode material layer, thereby forming a grid electrode with a preset size.

The invention discloses a crop leaf abnormal image extraction method based on video monitoring. The method is characterized by including the steps that 1, a crop growth video is collected, and frame processing is conducted on the crop growth video; 2, a crop growth key frame image is extracted through a self-defined similarity clustering method; 3, a binary image of a salient region of the crop growth key frame image is extracted through a self-defined image binaryzation method; 4, an RGB image in which the complicated background is removed and a crop leaf abnormal region is contained is extracted through an improved image complicated background removing method. The method is capable of rapidly extracting a key frame image with few redundancies from a large number of crop growth frame images and accurately extracting crop leaf abnormal images, and therefore a scientific and effective basis is provided for automatic identification and prevention of follow-up pest and disease damage.

The invention relates to a zipper slider and a zipper head with the same. The zipper slider comprises a cap, a self-locking element, an elastic element and a slider body, wherein the slider body comprises an upper wing plate and a guide post; the self-locking element is arranged on the upper wing plate of the slider body; one end of the elastic element is fixed between the slider body and the cap; the other end of the elastic element is elastically pressed against the self-locking element; the cap is fixed on the upper wing plate of the slider body; the self-locking element comprises a horse hook part, a pivot joint part and a pressing part; the pivot joint part is positioned between the pressing part and the horse hook part; a pivoting part is arranged in the cap; the self-locking element is in pivot joint with the pivoting part of the cap through the pivot joint part; a through hole is formed on the slider body corresponding to the pressing part; and the pressing part is communicated with the outside of the slider body through the through hole. The invention has the advantages that: the zipper slider and the zipper head have simple structures and are convenient to assemble.

The embodiment of the invention provides a 3D point cloud densification method and device based on a binocular RGB image. The method comprises the following steps: step 1, generating a depth image from a binocular RGB image; estimating an approximate three-dimensional coordinate point of each pixel point of the depth image under a LiDAR coordinate system according to the depth information of the depth image; step 2, performing ground segmentation on the point cloud by using a cyclic RANSAC algorithm, and extracting a non-ground point cloud; step 3, inserting the extracted non-ground point cloud into a KDTree, searching a predetermined number of neighbor points in the KDTree according to the approximate three-dimensional coordinate points of each pixel point under the LiDAR coordinate system, and performing curved surface reconstruction by utilizing the neighbor points; and step 4, exporting accurate coordinate points of the approximate three-dimensional coordinate points by a computational geometry method according to a curved surface reconstruction result and calibration parameters of the LiDAR and the camera, and fusing the accurate coordinate points with the original LiDAR pointcloud to obtain a densified point cloud.

The invention relates to a ball and a high-speed rotational motion parameter detection method and aims to solve problems of complexity in spherical surface feature point recognition and calculation, unreliability in calculation result and the like. The ball is provided with spherical surface marks, and a spherical surface is divided by a first large circle, a second large circle and a third large circle in perpendicularity. Trisection points of the first large circle and two intersection points of the second large circle and the third large circle are five datum points of the spherical surface marks, the spherical surface marks include at least five mark lines equal in length, the mark lines are not parallel and shorter than 1 / 3 of the circumference of the first large circle, and midpoints of the mark lines are coincided with the five datum points respectively. The mark lines at least include a first mark line, a second mark line, a third mark line, a fourth mark line and a fifth mark line.

The invention discloses a special-shaped blank design method for a high-temperature alloy F-type section ring. The special-shaped blank design method includes the steps: designing a blank according to an F-type ring volume distribution characteristic curve; setting equal overall heights of an ideal blank and the ring according to metal axial flow characteristics in the rolling process of the ring; determining the height between two steps according to the height of the corresponding step of the ring. A mathematical model is built according to selected rolling ratio and space conservation laws, and the radial size of the ideal blank is determined. The size of the ideal blank is simplified according to the volume splitting complementary principle. The size of a fillet in the blank is optimized according to section filling conditions obtained by finite element simulation, and the shape and the size of an actual blank are finally determined. By the method, the shortcoming of filling incompleteness in the rolling process of the F-type ring can be effectively overcome, production cost is saved while material utilization rate is increased, and damage of machining to the internal streamline of a forging is avoided.

The invention discloses a method for extracting the outline of a carved character, and the method comprises the steps: carrying out histogram analysis of an optical character image, and obtaining a high-gray-value region and a low-gray-value region of the character through a multi-phase activity outline model and a double-parallel-set method; carrying out local histogram analysis of the obtained high-gray-value region and low-gray-value region, determining the gray value transformation coefficient according to the gray value distribution and corresponding relation of the obtained high-gray-value region and low-gray-value region, and converting a multi-phase image into a double-phase image; employing an activity outline model which can employ local information, extracting the outline of a single-parallel-set method, and enabling the outline to serve as the outline of the carved character. The extracted character outline is complete and accurate, and the features of the character can be conveniently extracted through the results.

The invention provides a moving target detection method and device, equipment and a readable storage medium. The method comprises the steps of obtaining a video sequence frame image; performing movingtarget detection on the current frame image by adopting a Gaussian mixture model to obtain a first moving target binary image and a current frame background image; performing moving target detectionaccording to the current frame of background image and the five frames of images corresponding to the current frame of image by adopting an improved five-frame difference algorithm to obtain a secondmoving target binary image; performing OR operation on the first moving target binary image and the second moving target binary image to obtain a total moving target binary image; and performing morphological processing on the total moving target binary image to obtain a moving target. The moving target detection accuracy of the improved five-frame difference algorithm is improved, the existing smear problem is basically eliminated, or operation is carried out on the moving targets detected by the two algorithms to obtain all the moving targets detected by the two algorithms, and the detectedmoving targets can have complete contours.

The invention provides a temperature detection method and device, electronic equipment and a readable storage medium. The method comprises the following steps of: acquiring a target visible light image and a target thermal induction image including a human face; when a shielding object exists in a face area in the target visible light image, determining a first image area of an unshielded face from the target visible light image; based on the position of the first image area in the target visible light image, determining a second image area representing an unshielded face from the target thermal induction image; and acquiring a detection result representing the human face temperature according to the second image area. In the scheme, the temperature of the human face is detected by judgingwhether the human face has a shielding object or not and then determining the second image area of the unshielded human face from the thermal induction image, so that the shielding object is prevented from influencing the detection of the temperature of the human face, and the accuracy of temperature detection is improved.

The invention discloses a bevel gear-elliptical gear eccentric wheel rocking bar combined-type wide and narrow row spacing transplanting mechanism. In the prior art, seedling taking of a conventional seedling taking and transplanting integrated vegetable pot seedling transplanting mechanism is realized via direct clamping of seedlings, so that seedling roots are damaged; seedling taking and transplanting are realized via two mechanisms, and the mechanisms are complex. Each of the two ends of a center shaft of the bevel gear-elliptical gear eccentric wheel rocking bar combined-type wide and narrow row spacing transplanting mechanism is provided with a first-grade planetary gear train, and the two first-grade planetary gear trains are arranged to be symmetrically; the two ends of each first-grade planetary gear train is connected with a second-grade upper planetary gear train and a second-grade lower planetary gear train; each second-grade upper planetary gear train is provided with an upper transplanting arm, and each second-grade lower planetary gear train is provided with a lower transplanting arm; each first-grade planetary gear train comprises a first-grade gear case, a center elliptical gear, a first-grade upper planetary elliptical gear, a first-grade lower planetary elliptical gear, a first-grade upper planetary shaft, a first-grade lower planetary shaft, an upper eccentric wheel, an upper connecting rod, an upper rocking bar, a lower eccentric wheel, a lower connecting rod, and a lower rocking bar. The bevel gear-elliptical gear eccentric wheel rocking bar combined-type wide and narrow row spacing transplanting mechanism can be used for realizing in-pot soil-taking-type seedling taking, and in seedling taking period, translational motion of seedling claws is maintained, so that pot soil completeness is maintained as far as possible.

The invention relates to an improved pathogen detection method and device based on an improved cyclic steering filtering algorithm and a storage medium. The method comprises the following steps: S1, constructing a pathogen training set for detecting spores and bud spores; s2, training an artificial intelligence target detection model for detecting pathogens of spores and blastospores based on the pathogen training set of spores and blastospores constructed in S1; s3, carrying out filtering operation on the original image based on an improved cyclic steering filtering algorithm; and S4, detecting the pathogen in the input image: carrying out filtering operation on the input image according to the step S3, and then inputting the filtered image into the artificial intelligence target detection model of the pathogen trained in the step S2 for pathogen detection to obtain the position information of the pathogen. According to the invention, the algorithm based on improved cyclic guide filtering is adopted, and the processed image is input into the deep learning target detection model based on the convolutional neural network, so that the detection rate of pathogens is effectively improved.

The invention relates to the technical field of image processing, in particular to an edge fine matting method and system for clothes, and the method comprises the following steps: S110, carrying outfeature recognition of a photographed clothes image or clothes video, classifying the clothes in the clothes picture or the clothes video into a preset type; S120, correcting the preliminary clothes boundary extracted from the clothing picture or the clothes video according to the clothes boundary parameters of the classified type to obtain an accurate clothes boundary; and S130, segmenting a clothes image from the clothes picture or the clothes video according to the accurate clothes boundary. According to the invention, that the contour of the extracted main image is complete and the main image can be extracted thoroughly.

The invention discloses a pest image recognition method based on a Bayesian width residual neural network. The method comprises the following steps: step 1, preprocessing the pest image recognition training data set, using a rich edge detection algorithm (Rich-Edge) Pest edge detection is performed on the grayscale image. Step 2. Construct a Bayesian Wide-Residual Network (BWResNet). Step 3: Input the pest edge image obtained in Step 1 into the BWResNet constructed in Step 2. Use the pest edge image training set obtained in step 1 to obtain the total error function of BWResNet. Step 4: Use the error function obtained in Step 3 to train the network. Here we propose the block-conjugate (Block‑cg) algorithm to train the network. Step 5. Update hyperparameters according to the optimized network in step 4. Step 6: Repeat steps 4 and 5 to obtain the final network, and the classification accuracy rate obtained after inputting the verification set of pest images into the network is higher.

The present invention relates to an improved pathogen detection method, device and storage medium based on an improved cycle-guided filtering algorithm, comprising the following steps: S1: constructing a pathogen training set for detecting spores and blastospores; S2: constructing spores and blastospores based on S1 Pathogen training set of blastospores, training artificial intelligence target detection model for detecting spores and blastospore pathogens; S3: Filtering the original image based on the improved loop-guided filtering algorithm; S4: Detecting pathogens in the input image: the input image according to S3 performs a filtering operation, and then inputs the filtered image to the pathogen artificial intelligence target detection model trained by S2 to detect the pathogen, and obtain the location information of the pathogen. The invention effectively improves the detection rate of pathogens by adopting an improved loop-guided filtering algorithm and inputting processed images into a deep learning target detection model based on a convolutional neural network.

The utility model discloses an adherent white blood cell segmentation method based on nucleus-marked watershed transformation. The adherent white blood cell segmentation method comprises the following steps: firstly inputting original RGB (Red, Green, Blue) images and discovering the generally difficult-to-solve problem of peripheral white blood cells and bone marrow white blood cells in the image processing process; secondly, carrying out HIS (Hue-Saturation-Intensity) and LUV color space and grayscale space conversion on the original images and analyzing the characteristics of each channel component image; thirdly, respectively carrying out threshold value segmentation and image subtraction on components B and grayscale images to obtain white blood cell images containing a part of impurities; fourthly, obtaining a target taking a white blood cell nuclei as a marker through an image enhancement technology; fifthly, carrying out morphological operation and watershed transformation on the white blood cell nuclei and the white blood cell images containing the impurities to remove the impurities, obtain accurate white blood cell images and solve the problem of cell adhesion; finally, cutting the targeted white blood cells, converting the targeted white blood cells into an LUV space, clustering the white blood cell images from the view of space and color and obtaining a white blood cell nucleus.

The invention relates to the technical field of mold manufacturing, and discloses a washable suit. The washable suit comprises a suit body and buttons, the suit body is made of a cotton cloth which iswoven from crease-resistant cotton fiber, fixation devices are fixedly connected to two sides of the suit body, the suit body is fixedly connected to two symmetrically-arranged sleeves through the fixation devices, the buttons are symmetrically arranged on two sides of the front part of the suit body, and two symmetrically-arranged pockets are fixedly connected to the front part of the suit body.According to the washable suit and a manufacturing method thereof, the suit can be prevented from being deformed during washing, in the machining process, all garment pieces composing the suit are serged to prevent rough selvedges of the ready-made suit during washing, meanwhile, the suit can be guaranteed to keep a tall and straight appearance after being washed, and when the suit is broken, groove plates and bump plates are separated from the suit, a zipper can be recycled, and the practicability of the zipper is improved.

The invention provides a three-dimensional display method aiming at 360-degree multi-view three-dimensional human body microwave image data. The method specifically includes the following steps that 1, the 360-degree multi-view three-dimensional human body microwave image data is acquired; 2, the acquired 360-degree multi-view three-dimensional human body microwave image data is processed; 3, theprocessed data is displayed. Compared with previous three-dimensional data display methods, the three-dimensional display of the 360-degree multi-view three-dimensional human body microwave image datahas the advantages that the edge resolution is higher, the outline of a human body is clearer and more complete, and the slice information of the height direction and the distance direction can be displayed in a layered mode.

The invention discloses a carbon block filler adsorption device. The carbon block filler adsorption device comprises a material suction cylinder, wherein the top of the material suction cylinder is fixedly connected with a flexible cloth cylinder, a connecting cylinder is fixed to the top of the flexible cloth cylinder, a mounting support is fixed to the back of the connecting cylinder, a discharging bent pipe is connected to the top of the connecting cylinder, and connecting buckles are fixed to two sides of the material suction cylinder respectively. Each sling is fixed to the corresponding connecting buckle, the tops of the slings are fixedly connected with a hanging plate, and the top of the hanging plate is fixedly connected with a steel strand of a winding device through connecting wires. The adsorption device is simple in structure, a carbon block filler can be rapidly adsorbed into a filler tank from the ground, the height of the material suction cylinder can be gradually reduced along with reduction of the height of the filler, and the distance between the material suction cylinder and the filler is reduced.

The invention provides a 3D printing and cutting combination method. The 3D printing and cutting combination method comprises the following steps of acquiring model parameters of a to-be-printed object, determining first size information of a bounding box according to the model parameters, and acquiring second size information of a 3D printer; setting the number and the size of cutting blocks according to the first size information and the second size information, and dividing the bounding box into a plurality of independent modules according to the number and the size of the cutting blocks; and determining splicing surfaces of the multiple independent modules, wherein the splicing surfaces comprise splicing structures, the splicing structures are mutually matched structures of the multiple adjacent independent modules, and the multiple adjacent independent modules can be tightly connected through the matched structures. The invention further provides a 3D printing and cutting combined device, a 3D printer and a storage medium. According to the 3D printing and cutting combination method and device, the bounding box is divided into the multiple independent modules, the multiple independent modules are tightly connected through the splicing structures of the splicing surfaces, and it is guaranteed that the printed independent modules can be stably spliced.

The invention discloses a method and a system for detecting a contour of an urban building. The method comprises the following steps of: S1, processing cloud data of a light detection and ranging (LiDAR) point, so as to extract an initial contour of the building; S2, performing data registering on the cloud data of the LiDAR point and aerospace remotely sensed image data; S3, performing image orientation according to a data registering result; S4, detecting an image edge according to the image orientation; S5, building a buffering region around the initial contour, searching the image edge in the buffering region, and identifying a contour of an image side by using a geometric constraint condition; and S6, performing contour vectorization and normalization on the contour of the image side to obtain the final contour of the building. By the method, the contours of a large range of urban buildings can be precisely detected.

The invention discloses a special-shaped blank design method for a high-temperature alloy F-type section ring. The special-shaped blank design method includes the steps: designing a blank according to an F-type ring volume distribution characteristic curve; setting equal overall heights of an ideal blank and the ring according to metal axial flow characteristics in the rolling process of the ring; determining the height between two steps according to the height of the corresponding step of the ring. A mathematical model is built according to selected rolling ratio and space conservation laws, and the radial size of the ideal blank is determined. The size of the ideal blank is simplified according to the volume splitting complementary principle. The size of a fillet in the blank is optimized according to section filling conditions obtained by finite element simulation, and the shape and the size of an actual blank are finally determined. By the method, the shortcoming of filling incompleteness in the rolling process of the F-type ring can be effectively overcome, production cost is saved while material utilization rate is increased, and damage of machining to the internal streamline of a forging is avoided.

The invention discloses a method for edge extraction and edge fitting of optically carved characters based on structural features. Marking, using the K nearest neighbor algorithm to remove the false edge of the sample according to the marked feature vector; constructing different connection modes according to the stroke structure characteristics of the character and the distance between the endpoints of the intermittent edge line segment to form a pixel scene; according to each pixel scene The positional relationship of each edge line segment, the gray information of the line segment, and the distance between the line segments determine the connection mode of the end points of the line segment, and perform edge fitting to form a character outline. The edge extracted by the invention is accurate, and the profile of the character obtained by fitting is complete, which brings great convenience to the subsequent character feature extraction.

The invention discloses a video foreground and background separation method and system in the technical field of video processing. The method comprises the following steps: collecting video information; and inputting the acquired video information into the constructed robust principal component analysis model based on robust Gaussian scale mixing and generalized nuclear norms, and extracting a corresponding video foreground and a corresponding video background. The invention includes describing a sparse foreground in the video by adopting a Gaussian scale mixed function; describing a low-rank background in the video by adopting a generalized nuclear norm with relatively good approximation; introducing a noise item into the method to model the noise in the nature, so that the anti-noise robustness of the method is enhanced. The invention has the advantages that the foreground and background separation effect of the video is better, the extracted foreground is clearer, the contour is more complete, and the anti-noise capability is high; the invention is applied to video foreground and background separation, and it is proved that the video foreground and background separation method based on robust Gaussian scale mixing and generalized nuclear norms has the good video foreground and background separation effect and the anti-noise advantage.