53results about How to "Speed up splitting" patented technology

The invention discloses a feature identification and fast search method for a hotspot path based on route big data, relates to a network path searching method and in particular relates to a method for performing hotspot path identification and search in a road network according to historical route big data. The invention provides a set of core features of a hotspot path and the feature identification method of the hotspot path. The method comprises a vector markup method for path heat and a ranking mechanism based on the path heat; an index mechanism is provided, so that the frequency of query on the route big data and the cost of input / output (I / O) of a disk in the process are reduced. Moreover, the method for rapidly querying the hotspot path based on the time interval comprises the following steps of: determining the starting point, destination and interested time interval and calculating the hotspot path, which meets the conditions, according to the historical route big data. The reasonable definition of the hotspot path is given, the hotspot path query at any time interval is provided, and the method is an important supplement to traditional path query methods and can well provide core technical support for various services based on positions.

The invention discloses a traffic scene analysis method based on a multi-task network, comprising the following steps: a multi-task network is divided into an encoder, a partition decoder and a detection decoder, wherein the encoder extracts the features of the image and extracts the multi-scale information from the feature map; the segmentation decoder enlarges the size of the feature map and fuses it with the feature map; the detection decoder processes the input characteristic map and outputs a corresponding target detection result; using a deep learning framework Tensorflow to configure, train and test the above multitasking network. The multi-task network of the invention can extract abundant image features, make up for the loss of image detail information caused by downsampling in the encoder, and help to improve the segmentation and detection effect. The invention designs a multi-task network structure, which can realize semantic segmentation and target detection of traffic scene images through one-time back propagation, and has better real-time performance and higher accuracy rate.

The present invention discloses an image segmentation method and device, and relates to the field of the computer technology. The objective of the invention is to solve the problem that the tooth CBCT (Cone Beam Computed Tomography) image segmentation method is low in efficiency in the prior art. The method comprises: obtaining dental crown grid data and tooth CBCT image data, wherein the dental crown grid data and the tooth CBCT image data being two three-dimensional data with different types of the same one dentition; registering the dental crown grid data and the tooth CBCT image data, and determining the dental crown regions corresponding to each tooth in the tooth CBCT image data; performing the segmentation of the tooth CBCT image data according to the dental crown region, and obtaining the initial three-dimensional region of each tooth; and performing the segmentation of the initial three-dimensional region corresponding to each tooth according to a SVM (Support Vector Machine) algorithm to obtain the final three-dimensional shape of each tooth. The image segmentation method and device are applicable to the process of the dental orthodontics.

The invention relates to a contour tracing method based on shape-transmitting united division and image-matching correction. The contour tracing method comprises the steps of in a shape-transmitting united division part, combining a shape prediction image with original image and video frames, by virtue of relocation of position coordinates, providing a new image model construction method, constructing a new energy optimization function to carry out united division on an image model, in an image-matching correction part, carrying out image matching on a division result of a current image with a division result of a previous frame image, when the result of the image matching can not meet a limiting condition, regulating weighting parameters of an energy function in the shape-transmitting united division, and correcting inaccuracy of the division result. According to the contour tracing method based on the shape-transmitting united division and the image-matching correction, an accuracy rate of contour tracing of a target object in a video can be effectively improved.

The invention discloses an automatic three-dimensional model component category tagging method. The method comprises the following steps that in the training process, quick training is conducted according to a three-dimensional model tagging training set, and a quick tagging model used for classifying and tagging surface patches and mesh edges of an unknown three-dimensional model is obtained through training; in the tagging process, the quick tagging model, obtained through training, of the surface patches and mesh edges is utilized for classifying surface patches and mesh edges of a target model, classification probability distribution of the surface patches and mesh edges is obtained, a graph model is constructed, smoothing and optimizing of segmentation boundaries are conducted through multi-tag graph cut optimization, and therefore rapid automatic target three-dimensional model component category tagging is achieved.

The invention discloses a retinal fundus image preprocessing method. The retinal fundus image preprocessing method is characterized by including the steps of 1), reading an original image, namely reading the original image of the retinal fundus through a green channel; 2), removing blood vessel center light reflex; 3), removing salt-and-pepper noise; 4), smoothing noise; 5), conducting background extraction; 6), obtaining a shadow correction image; 7), obtaining a homogenization image; 8), solving a complementary image; 9), obtaining a blood vessel enhanced image; 10), outputting the enhanced image. The retinal fundus image preprocessing method has the advantages that the method enables the processed retinal image to be better in image quality and richer in information quantity, reserved enhancement of useful information such as blood vessels, optic disks and central foveae in the retinal image is achieved, and feature extraction, segmentation and recognition of the retinal image are improved.

The invention discloses a robot disordered grabbing method and system based on machine vision. The method comprises the steps that a kinect camera and a grabbing system of a robot are built, the kinect camera is used for collecting image data of the surface of a target object, and the image data are preprocessed to obtain three-dimensional point cloud data; target detection, target segmentation, target clustering, key point feature extraction, feature registration and target object recognition are performed on the acquired three-dimensional point cloud data to obtain pose information of the target object; and according to the calibration result of the hand-eye system, coordinate conversion is carried out on the pose information, and a control instruction is sent to a robot to grab the target object. By analyzing the defects of the algorithm in the disordered grabbing process of the robot and improving the existing algorithm, the complex environment where the robot is located is adapted, and therefore the disordered grabbing accuracy of the robot and the flexibility of the robot are improved.

The invention provides an improved object-oriented high-resolution remote sensing image classification method. The method comprises the following steps: firstly, carrying out initial segmentation on an image by using an SLIC superpixel segmentation algorithm; enabling an initial segmentation result to generate over-segmentation; subjecting the initial segmented image to region fusion to solve anover-segmentation problembut a local under-segmentation phenomenon occurs in the image subjected to region fusion, so that secondary segmentation is carried out on the local under-segmentation region toobtain a new segmented image, the over-segmentation and under-segmentation phenomena in the segmentation process can be solved, and the image segmentation and image classification precision is improved.

The invention provides a markov random field (MRF) iteration-based synthetic aperture radar (SAR) unsupervised change detection method to overcome the defects of low detection capacity, accuracy and speed in the conventional SAR detection technology. Statistical parameters of a mixed distribution model of a difference map are estimated stably by an improved expectation-maximization (EM) algorithm, independency among pixels is defined by using an MRF model, and a more reliable and accurate image change detection result is generated by map cut-based MRF division. The invention also provides a device based on the method. The device comprises an optical fiber link input module, a digital signal processor (DSP), a synchronous dynamic random access memory (SDRAM), a compact peripheral component interconnect (CPCI) output module and a display terminal; the method and the device have the advantages that various high-accuracy high-speed unsupervised change detection can be realized; the device has the engineering application capacity of batch processing of mass data and meets the image analysis requirement under the complex condition; and compared with the conventional method, the method has the advantages that the processing accuracy and the processing speed are improved obviously.

The present invention provides a building structure of two-floor mobile plate home whose member materials can be used for several times and circularly reused and can be made up by adopting industrial production mode and its installation method. It includes four ground beams, four corner uprights, uprights, trimmer beams, purlines, four ring beams, sandwich panels, windows, doors, roof, corridor and stairs, and its installation method has eight steps, said eight steps are provided by said invention specification.

The present invention provides a building structure of one-flour mobile phone home whose member materials can be used for several times and circularly reused and can be made up by adopting industrial production mode and its installation method. Its structure includes four ground beams, four corner uprights, several uprights, four ring beams, several sandwich panels, one roof, several doors and several windows. The installation method of said building structure includes seven steps, said seven steps are provided by said invention specification.

The invention provides a fast continuous edge extraction method for an image. The fast continuous edge extraction method comprises the steps of: adopting a differential detection operator for carrying out edge detection on an image I, so as to obtain an edge detection result; performing skeleton extraction on the edge detection result to obtain a skeleton extraction result; traversing each pixel point in the skeleton extraction result, if a value of the current pixel point is not zero, counting the number of pixel points with the value not being zero in a nine rectangle grid which takes the current pixel point as the center, marking the current pixel point if the number is 2, and placing the current pixel point into a breakpoint extraction result set; continuing a breakpoint extraction result to obtain a continuation result; expanding the continuation result to a width r; generating an embedded function initial data matrix of a partial differential equation; and solving the partial differential equation by iteration to obtain a continuous edge of the image, so as to obtain an image continuous edge extraction result. The fast continuous edge extraction method can obtain the edge with good continuity in the image processing process, so as to better improve the precision of later image segmentation, measurement, matching and the like.

The invention discloses a label anti-collision method and a reader. The label anti-collision method comprises the following steps: 1, the reader reads all labels and sends a collision order used for indicating all the labels to be divided into n branches; 2, the branch with a minimal value of a counter is taken as a current branch; 3, the reader reads the current branch, a step 4 is carried out under the condition that the labels inside the current branch are replied and collided, and otherwise, the step 2 is carried out again when the current branch is successfully replied, the labels of thenext branch of the current branch are divided into m sub-branches, counters of the labels inside the next branch and the sequential branches are all added with (m-2); 4, the reader sends a collision order used for indicating that the labels of the current branch are divided into p sub-branches, and each label inside the current branch respectively updates the value of the counter; 5, the sub-branch with the minimal value of the counter is taken as the current branch, and the step 3 is carried out again. By adopting the invention, forward separation is realized, and the counting efficiency is improved.

The invention discloses a non-woven cotton soft towel with a moisturizing function and a preparation method thereof, and belongs to the technical field of paper making. The non-woven cotton soft towelis prepared from a soft dry paper towel and a collegen liquid, wherein the collegen liquid is uniformly sprayed on the surface of the soft dry paper towel and prepared from the following components in parts by weight: 2 to 6 parts of jojoba oil, 1.5 to 3 parts of an aloe extracting solution, 50 to 65 parts of deionized water, 1 to 4 parts of glycerol, 0.3 to 0.6 part of oligopeptide, 1 to 3 partsof clove oil, 2 to 6 parts of collagen, 0.1 to 0.3 part of sodium hyaluronate and 2 to 4 parts of 1-3 propylene glycol. The preparation method comprises the following steps: S1, weighing and matching; S2, preparing an A phase raw material; S3: preparing a B phase raw material; S4: preparing the collegen liquid; S5, carrying out spraying treatment. The formulation disclosed by the invention is more scientific and reasonable and enables the cotton soft towel to have better moisturizing capability; in addition, with smaller molecular weight, the oligopeptide can be directly absorbed by the humanskin, so that the division speed of skin cells can be improved to a certain extent; besides, the antibacterial capability of the non-woven cotton soft towel with the moisturizing function can be improved.

The invention provides a slag slice image segmentation method based on a laser three-dimensional camera. The method comprises the following steps: a feature extraction unit of the slag slice spatial information is composed of a plurality of scanning lines generated by the laser three-dimensional camera; the properties of slag slices are judged comprehensively by comprehensive height gradient, height mean and height gradient mean, and the binary image of slag slices is established according to the properties of slag slices. The binary images of slag slices are eroded and dilated respectively, and the non-slag slice regions are filtered based on the morphological characteristics of slag slices. The morphological features between the filtered connected domains are obtained, the blocked slag slices are filtered, and the segmented images of the unblocked surface slag slices are obtained. As that slag slice distribution spatial information obtain by the laser three-dimensional camera can avoid the influence of the uneven illumination on the image segmentation result, the slag slice image segmentation is realize by judging the slag slice shielding area based on the spatial height information, the slag slices shielded are eliminated, and the surface slag slices segmentation and the statistical accuracy of feature extraction are improved.

The invention belongs to the technical field of citrus planting and in particular discloses a citrus scion rooting solution and a citrus seedling breeding and grafting method. The rooting solution comprises 20-30 parts of gamma-polyglutamic acid, 40-60 parts of indoleacetic acid, 30-40 parts of borax, 0.05-0.08 part of mono-nitration guaiacol sodium salt and 0.01-0.05 part of zinc sulfate. The rooting agent is relatively high in physiological activity, the gamma-polyglutamic acid and the mono-nitration guaiacol sodium salt have synergetic effects, the plant cell division speed can be increased, rooting and root strengthening can be improved, and the absorption property of root systems upon nutrients is improved; the rooting solution is applied to citrus seedling grafting and breeding, andthrough scion selection, scion treatment, stock treatment, combination and grafted seedling management, the survival rate of grafted nursery stock is increased, and the disease resistance is also improved.

The invention discloses a watermelon sweetness-increasing and high-yield planting method. The method comprises the following steps that (1) base fertilizer application: before transplanting, 1,300-1,500 kg of fertilizer A is applied per mu; (2) fertilizer application according to seedlings: after melon seedlings are transplanted by 18-20 days, 600-700 kg of fertilizer B is applied per mu, a piece of cotton cloth soaked by a nutrient solution is placed to the root of each melon seedling, and the size of the cotton cloth is 10*10 cm; (3) fertilizer application according to fruits: after the watermelons grow to 0.8-1.2 kg, 200-300 kg of fertilizer C is applied per mu. The watermelon sweetness-increasing and high-yield planting method is a method for preparing specific fertilizer according to nutrition requirements at different stages and applying the fertilizer, the effect is remarkable, the watermelon sweetness can be effectively increased, the planted watermelons are popular among consumers, the fruiting number of single plant watermelon and the piece weight of the watermelon are improved greatly compared with the past, and the income of the grower is increased greatly.

The present invention discloses one kind of fracture-treating externally applied medicine and its preparation process. The fracture-treating externally applied medicine is prepared with angelica, Chinese ligusticum, trogopterus dung, rhubarb, licorice and other three kinds of Chinese medicinal materials in certain weight proportion. The preparation process includes the steps of grinding the Chinese medicinal materials into fine powder, 800-mesh sieving, mixing the medicine powder with egg white to form medicine liquid, and soaking gauze inside the medicine liquid to form medicated bandage. When used, the medicated bandage is wrapped to the fracture part and the fracture part is fixed with splint. The fracture-treating externally applied medicine can speed the healing of fracture.

The invention discloses a culture medium for rapidly proliferating neural stem cells. The culture medium comprises components as follows: taking a DMEM / F12 culture medium as a base material, 15-25 ng / mL of an epidermal growth factor, 10-15 [mu]g / mL of transferrin, 15-25 ng / mL of an alkaline fibroblast growth factor, 15-25 ng / mL of fibronectin, 50-70 ng / mL of a tetratype collagen group, 0.05-0.13 mg / mL of antibiotics, 40-60 [mu]g / mL of bovine serum albumin, 8-10 [mu]mol / mL of non-essential amino acid and a B27 cell culture additive. The non-essential amino acid is a nutrient substance necessaryfor neutral stem cells, the cell differentiation phenomenon caused by incomplete absorption of nutrient substances is avoided, and meanwhile, the cracking speed affected by toxin in a classificationprocess is decreased by addition of antibiotics.

The invention provides a method used for analyzing the porous asphalt mixture void structure by adopting a CT technique, comprising the steps of: preparing porous asphalt mixture specimens, scanning the porous asphalt mixture specimen by an X-ray CT machine, reinforcing and cutting CT images, extracting void characteristic, calculating the characteristic parameter of the void structure, analyzingvoid plane information, analyzing void space information and memorizing the analysis result. The method adopts the CT scanning technique, can exactly and quickly obtain the digital images of the porous asphalt mixture, reinforces and cuts the images, extracts the corresponding characteristic parameters of the void, carries out the quantitative description on the structural characteristic of the void by a fractal theory by virtue of the parameters, comprehensively reflects the void structure information, and avoids the effects of all characteristics of the porous asphalt mixture on the void evaluation. The parameters lays the foundation for further researching the effect rule of the void structure on the splitting tensile strength and the permeability of the porous asphalt mixture and for researching the void decay behavior of the porous asphalt mixture.

The invention discloses a method for inducing and culturing hybrid sweetgum tetraploid. The method comprises the following steps: correspondingly performing bud primordium predifferentiation culture on leaves and petioles of separated sterile plants of hybrid sweetgum; and then performing colchicine inducing culture on the leaves and petioles subjected to the predifferentiation culture. With the adoption of the method, the hybrid sweetgum tetraploid inducing rate is high, and the yield of the hybrid sweetgum tetraploid is high; in addition, the artificial hybrid sweetgum allotetraploid is induced and obtained for the first time; the quantity of two allogenetic chromosomes is copied; the hybridization effect and the ploidy effect are combined, so that heteromorphosis can be realized greatly, and as a result, the probability of further improving sweetgum genetic gain and selective breeding is increased.

The invention relates to an image processing method and device, electronic equipment and a storage medium, and the method comprises the steps: carrying out the semantic category segmentation of a to-be-segmented rib image, and obtaining a semantic category segmentation result corresponding to the to-be-segmented rib image; performing instance segmentation on the to-be-segmented rib image to obtain an instance segmentation result corresponding to the to-be-segmented rib image; and according to the semantic category segmentation result and the instance segmentation result, determining a target marking result corresponding to the rib image to be segmented, the target marking result including a plurality of ribs in the rib image to be segmented and a rib identifier corresponding to each rib. According to the embodiment of the invention, the global semantic information and the local geometric information of the image can be comprehensively considered, and the rib segmentation and marking precision is improved.

The invention relates to a robust interactive medical image segmentation method. The segmentation method comprises the following steps: (1) pre-segmenting a window where a retinal lesion is located inan OCT image; (2) setting two fixed foot points in the window to form a to-be-segmented window; and performing iterative contraction by using the active contour model in the to-be-segmented window tocomplete segmentation. According to the image segmentation method based on the active contour model, the segmentation range of a retina OCT image with high speckle noise is reduced through manual windowing so as to eliminate the influence of speckle noise in a background area on the model and accelerate the focus segmentation process. Inaccurate segmentation caused by the lightproof characteristic of the focus can be overcome through the foot points; in combination with fractional differential enhancement and a local fuzzy energy functional, an active contour model is proposed, so focus segmentation can be completed robustly and efficiently.

The invention belongs to the technical field of Chinese herbal medicine planting, in particular to a roxburgh rose planting method. The roxburgh rose planting method comprises the following steps of land preparation, field planting and management. According to the roxburgh rose planting method, the microbial content in the soil is increased by adding fertilizer-adding liquid into soil, and bacterial quantity is the most in soil microorganism. The organic matter in the soil is increased and the soil fertility is improved. Therefore, the roxburgh rose has better growth and high fruit yield. Thecontent of soluble solids is up to 29.11%, soluble protein is up to 0.53 mg / g, soluble sugar is up to 48.01%, thus changing the taste of the roxburgh rose.

The invention discloses a level gauge distance measurement method based on image enhancement, and relates to the field of artificial intelligence. Comprising the following steps: S1, collecting bar code scale images by using a level gauge; s2, performing grid segmentation on the bar code scale image to obtain a to-be-processed grid image; s3, the enhancement necessity of each to-be-processed grid image is calculated according to the gray variance of each to-be-processed grid image; s4, clustering each to-be-processed grid image to obtain a bar code scale region, and enhancing the bar code scale image through adaptive segmentation of the bar code scale region; and S5, matching and positioning the bar code scale areas in the enhanced image and the template image to obtain a measurement result of the level gauge. According to the invention, the bar code scale area is determined according to the bipolar phenomenon of the surface gray value of the bar code scale, and image enhancement is carried out on the bar code scale area, so that the features of the bar code scale are more obvious, and the subsequent bar code scale image segmentation and bar code recognition accuracy are improved.

The invention relates to the technical field of seed treatment, in particular to a rhodomyrtus tomentosa seed soaking solution which is composed of a solution A and a solution B. The solution A contains copper sulfate, ferrous sulfate, salicylanilide, sodium chloride, and the solution B contains cellulase, jasmonic acid, polyethylene glycol, forchlorfenuron, urea, potassium dihydrogen phosphate and trace element fertilizer. Aiming at the characteristics of rhodomyrtus tomentosa seeds, the seed soaking solution can effectively kill bacteria, lower pest invasion of seed seedlings caused by the bacteria and promote nutritional substances to enter seeds so as to supplement energy needed for seed germination, and seed germination rate, seedling survival rate and seedling growth quality are improved.

The invention discloses rooting powder for flower cuttage. The rooting powder is prepared from the following raw materials in parts by weight: 3-4 parts of gamma-polyglutamic acid, 36-39 parts of vinegar residues, 48-54 parts of kelp residues, 1-3 parts of magnesium sulfate, 2-3 parts of manganese sulfate, 8-12 parts of fish emulsion, 27-30 parts of potassium humate, 4-5 parts of ferrous sulfate, 35-38 parts of green hawthorn fruits, 36-39 parts of fresh smoked plums, 70-75 parts of chaenomeles speciosa, and a proper amount of water. With the co-synergistic action of organic acid in the gamma-polyglutamic acid, the green hawthorn fruits, and the fresh smoked plums, the rooting powder disclosed by the invention can accelerate the plant root cell division speed, so that the cut of a twig cutting forms a callus tissue within a short period of time, and adventitious root primordia is promoted to explosively strike root; and the nutrition support is provided for the growth of a root system by virtue of added fermented materials, the formation of powerful secondary roots is facilitated, and the rooting powder is suitable for flower cuttage or transplantation.

The invention discloses a machine vision-based robot disorderly grasping method and system. By building the grasping system of kinect camera and robot, using the kinect camera to collect image data on the surface of the target object, and preprocessing the image data to obtain three-dimensional point cloud data; Detection, target segmentation, target clustering, key point feature extraction, feature registration, identifying target objects, and obtaining the pose information of the target object; according to the calibration results of the hand-eye system, the coordinate transformation of the pose information is carried out to The robot sends out control instructions to realize the grasping of the target object. By analyzing the lack of algorithms in the robot's disorderly grasping process, and by improving the existing algorithms to adapt to the complex environment in which the robot is located, the accuracy of the robot's disorderly grasping and the flexibility of the robot are improved.

The invention discloses a cucumber grafting cultivation method, which is characterized in that it includes the following aspects: (1) seed detoxification, mixing cucumber and loofah seeds, and using abamectin solution to carry out atomization detoxification treatment; (2) For storage, add ethanol to the selenium-enriched fermentation liquid, put the seeds in and store them at low temperature for 8-10 days, and then carry out germination treatment; (3) For seedling cultivation, mix loess and organic fertilizer to lay seedling boxes, adjust the seedlings after sowing Indoor carbon dioxide and pressure for cultivation; (4) Grafting, cucumber seedlings and loofah seedlings are grafted when they grow to 4-6 true leaves, the incision is soaked in a growth-promoting agent, and the incision is wrapped after docking; (5) Daily management, regular use The growth-promoting agent and the selenium-enriched fermentation liquid are sprayed and watered on the cucumber grafted plants, and the cultivation temperature is adjusted, and the seedlings are raised about 20 days after grafting and then transplanted.