Product
Patsnap Eureka
For R&D, Patsnap Eureka makes reading and utilizing patents & technical documents easy.
Patsnap Eureka AIR
Designed for self-driven R&D workflows. Generate viable solutions, solve complex R&D challenges, empower your innovation with AI.
Patsnap Eureka Materials
Designed for material experts only. Revolutionize your material R&D, from search, analyze, to developing new materials.
Feature
TechResearch
Generate reliable direction feasibility study reports for your R&D in just a few steps.
TechSeek
Discover and master advanced knowledge NOW. Basics, ideas, possibilities, all at once.
TechMind
As an expert in R&D Theories, TechMind can generates customized viable solutions instantly.
TechRisk
Analyze your overall solution with one click, know your potential R&D risks in advance.
TechMonitor
Get weekly tech updates, stay abreast of the latest tech innovations and key insights.
Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
102results about How to "Extract complete" patented technology
Efficacy Topic
Property
Owner
Technical Advancement
Application Domain
Technology Topic
Technology Field Word
Patent Country/Region
Patent Type
Patent Status
Application Year
Inventor
Webpage core block determining method based on DOM (Document Object Model) node text density
InactiveCN102184189AImmune to styleImprove visualizationSpecial data processing applicationsHyperlinkVolumetric Mass Density
The invention relates to a webpage core block determining method based on a DOM (Document Object Model) node text density, which comprises the following steps of: 1, analyzing an HTML (Hyperlink Text Markup Language) webpage, and generating a DOM tree to ensure that each HTML label corresponds to one node in the DOM tree, wherein character contents in the webpage are leaf nodes of the DOM tree; 2, adding statistical information including number of all text characters included by the nodes, number of all labels included by the nodes, number of all hyperlink text characters included by the nodes and number of all hyperlinks included by the nodes, and defining text density of the nodes according to the statistical information; and 3, determining a webpage core block according to the text density of the nodes in the DOM tree. According to the invention, on the premise of being free of being influenced by webpage coding styles and remaining the original webpage DOM structure, the core content block in the webpage is completely extracted.
Owner:BEIJING INSTITUTE OF TECHNOLOGYGY
Visual SLAM method based on point-line fusion
PendingCN110782494AExtract completeFast extractionImage analysis3D modellingImage extractionRadiology
The invention discloses a visual SLAM method based on point-line fusion, and the method comprises the steps: firstly inputting an image, predicting the pose of a camera, extracting a feature point ofthe image, and estimating and extracting a feature line through the time sequence information among a plurality of visual angles; and matching the feature points and the feature lines, tracking the features in front and back frames, establishing inter-frame association, optimizing the pose of the current frame, and optimizing the two-dimensional feature lines to improve the integrity of the feature lines; judging whether the current key frame is a key frame or not, if yes, adding the key frame into the map, updating three-dimensional points and lines in the map, performing joint optimization on the current key frame and the adjacent key frame, and optimizing the pose and three-dimensional characteristics of the camera;and removing a part of external points and redundant key frames; and finally, performing loopback detection on the key frame, if the current key frame and the previous frame are similar scenes, closing loopback, and performing global optimization once to eliminate accumulated errors. Under an SLAM system framework based on points and lines, the line extraction speed and the feature line integrity are improved by utilizing the sequential relationship of multiple view angle images, so that the pose precision and the map reconstruction effect are improved.
Owner:BEIJING UNIV OF TECH
Detection method for moving target in infrared image sequence under complex background
InactiveCN1885346AImprove real-time performanceSegmentation targets are accurateImage enhancementImage analysisImaging processingHistogram
The invention relates to a method for testing the motion target in infrared image sequence at complex background, belonging to the digit image processing technique, wherein the inventive method comprises: first, calculating the differential images between last frame and present frame, present frame and next frame, in infrared image sequence; then extracting change testing template from two differential images; combining two templates (selecting the crossed region), to obtain the motion target template; at last, marking the connecting area on the motion target template, to obtain the motion target; the step that extracting change testing template comprises: evaluating the differential image noise parameter (average value and variance) to extract initial change testing template, that iterating from two ends of histogram to the middle; using small unit to corrode and expand to process closed calculation with large unit. The invention can divide target accurately, have better real-time property, and strong robustness.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA
Multispectral remote sensing image land feature classification method based on spectrum and textural features
ActiveCN103559500AExtract completeComplete texture featuresCharacter and pattern recognitionSvm classifierClassification methods
The invention discloses a multispectral remote sensing image terrain classification method based on spectrum and textural features. A quadtree partitioning technology is adopted in the method to carry out multistage partitioning processing on images, spectrum and textural features of land features are extracted in an image block mode, an SVM classifier is adopted to conduct land feature classification on image blocks, and classification marginal regions of the image blocks are processed through a region growing method. Compared with the prior art, the multispectral remote sensing image terrain classification method has the advantages that anti-noise performance of spectral signatures and textural features in land feature classification is improved, the problem of sizes of textural feature extraction windows is avoided, and classification result regions are high in consistency and low in noise.
Owner:BEIHANG UNIV
Multi-feature fusion cascaded classifier-based fire fighting access vehicle detection method
InactiveCN106295636AEliminate the effects ofImprove accuracyCharacter and pattern recognitionImaging processingEdge extraction
The invention proposes a multi-feature fusion cascaded classifier-based fire fighting access vehicle detection method, and belongs to the field of image processing and machine learning. The method comprises the steps of extracting a road region in a fire fighting access image by utilizing a significance detection method so as to eliminate the influence of an object, in a non-lane region, on a detection result; extracting a vehicle bottom shadow region by adopting a vehicle bottom shadow adaptive threshold segmentation method based on a characteristic that a grayscale value of the road region obeys normal distribution; obtaining a lower edge of the vehicle bottom shadow by adopting a pixel change rate-based shadow edge extraction method, and constructing a region of interest of a vehicle according to the extracted lower edge of the vehicle bottom shadow; and extracting an Haar feature of the region of interest of the vehicle, training a feature value after fusion and an HOG feature value of ''integrogram'' calculation by utilizing an Adaboost algorithm to generate a strong classifier, and finally verifying whether the region of interest of the vehicle is the vehicle or not by using a cascaded classifier obtained by training.
Owner:CHONGQING UNIV
Face expression identification method based on partially shielded image
ActiveCN105825183AAchieve coarse classificationReduce mistakesAcquiring/recognising facial featuresSvm classifierImage matching
The invention discloses a face expression identification method based on a partially shielded image. The method comprises the following steps that 1) face images of N types of expressions included in a known type of face expression library are preprocessed; 2) the similarity between a sample to be tested in a test set and a training sample in a training set is calculated, and an image in the same type with the sample to be tested and nearest to the sample to be tested is obtained; 3) a shielded part of the sample to be tested is reconstructed; 4) PWLD features are extracted from the reconstructed sample to be tested and the training sample in the training set; and 4) an SVM classifier is used to identify all the samples to be tested in the test set in a classified manner. According to the invention, an image matching method is used to effectively reconstruct the shielded part of the image, the problem that features cannot be represented completely when only non-shielded parts are extracted is solved, a three-layer pyramid structure is used to extract global and local features of the image, and the accuracy of feature representation is improved.
Owner:HEFEI UNIV OF TECH
Multifeature-based target object contour detection method
ActiveCN102201120AEasy extractionExtract completeImage analysisCharacter and pattern recognitionUltimate tensile strengthVisual perception
The invention belongs to the field of computer vision technology and discloses a multifeature-based target object contour detection method. The multifeature-based target object contour detection method comprises the steps of filtering processing, extraction of local features of images, calculation of inhibiting weight under all features, making of inhibited contour images and binarization processing. In the multifeature-based target object contour detection method, a group of filters with different orientations is adopted to carry out filtering processing on input images to obtain an orientation information distributing image in all orientations; then the local orientation, brightness and contrast features of the images are respectively extracted, the inhibiting weights of a nonclassical receptive field to central pixel points are respectively calculated under each feature, and finally the inhibiting weights under all the features are combined to obtain a final inhibiting weight; and the inhibiting strength of pixels in the corresponding nonclassical respective field to the inhibiting weight of each pixel point is adjusted according to the inhibiting weight so as to obtain an inhibited contour image. In the multifeature-based target object contour detection method disclosed by the invention, multifeature information of the images is comprehensively input, and the capability of extracting an object contour quickly and completely from a complicated scene is effectively improved.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA
Cartoon animation fabrication method based on video extracting and reusing
The invention discloses a cartoon animation producing method based on video extraction and reuse. The invention comprises the following steps of: firstly dividing a cartoon video into cartoon frames; next, extracting cartoon roles via the foreground dividing technique based on the boundary; then, calculating the distance between any two frame roles and obtaining a distance matrix via using the role similarity on the boundary; then constructing an undirected graph via using the manifold learning arithmetic according to the obtained distance matrix; finally, appointing any two points in the undirected graph and establishing the shortest path between the two points which comprises the frames to form a smooth cartoon frame sequence. The method has the advantages of: effectively extracting the cartoon roles from the complex background via designing a foreground extraction technique based on boundary; making the similarity between roles more accurately by calculating the movement attribute of the roles in every frame via optical flow tracking technique; and constructing a cartoon data reuse platform to facilitate a user to produce cartoon animations conveniently and rapidly via manifold learning.
Owner:ZHEJIANG UNIV
A method and a system for extracting leaf growth parameters of fruit trees based on clustering segmentation
ActiveCN109166145AAccurate growth parametersAccurate extraction of growth parametersImage enhancementImage analysisFruit treeCluster algorithm
The invention provides a method and a system for extracting leaf growth parameters of fruit trees based on clustering segmentation, the method comprising the following steps: superclustering the pointcloud data of the canopy branches and leaves of a target fruit tree, and performing LCCP clustering on a plurality of adjacent voxel blocks in the obtained voxel block set to obtain a first clustering set; applying Kmeans clustering to any point group in the first clustering set to obtain the second clustering set; according to the point cloud data corresponding to each point group in the secondclustering set, obtaining the growth parameters of each leaf based on boundary extraction. After LCCP clustering is used to segment the point group obtained by superbody clustering, and the Kmeans clustering algorithm based on dynamic K-value is further adopted. The improved clustering Kmeans algorithm can automatically obtain the K value, the shortcoming of manual setting the K value in the traditional algorithm is overcome, the point cloud data segmentation of the canopy branches and leaves of the target fruit trees becomes more complete and more thorough, and then extracted leaf growth parameters are more accurately.
Owner:CHINA AGRI UNIV
Image local feature extracting method on basis of Hilbert curve and LBP (length between perpendiculars)
ActiveCN102663399AImprove performanceReduce complexityCharacter and pattern recognitionNeighbor relationLength between perpendiculars
The invention relates to an image local feature extracting method on the basis of a Hilbert curve and the LBP (length between perpendiculars). In the method, the Hilbert curve is adopted to arrange LBP features of an image according to a certain sequence and the sequence can ensure an obtained feature vector to comprise a space neighboring relation among feature points so as to improve the identification performance of a mode identification system. The local feature extracting method particularly comprises the following steps of: a first step of acquiring an input object and carrying out pretreatment of filtering, removing noise and the like; a second step of extracting the LBP features; and a third step of selecting the Hilbert curve and extracting a high-order feature. The method disclosed by the invention is easy to implement, only relates to simple difference, binaryzation and feature ordering and has low complexity, i.e. low calculation complexity of both difference and binaryzation.
Owner:北京博研高科技术有限公司
Foreground extraction method under complicated background
The invention provides a foreground extraction method under a complicated background. An infrared lamp is used for polishing over foreground objects. Then, an infrared camera and a color camera are used for extracting images. The infrared camera obtains the images, the foreground objects are illuminated because of polishing of the infrared lamp. A certain threshold value is used for carrying out binarization processing over the infrared images so that a mask image which corresponds to the foreground objects can be digged out. The mask image is carried out fine adjustment up and down of the pixel positions so as to be consistent with the images acquired by the color camera. Finally, the mask image after the fine adjustment is used for completely extracting the foreground image and well keeping the edges of the foreground images. Great improvements on the accuracy of extracting the foreground are obvious.
Owner:QINGDAO SMARTEYE TECH
Method and device for extracting urban heat island range
ActiveCN108332859AExtract completeClimate change adaptationPyrometry for hot spots detectionUrban heat islandThermal infrared remote sensing
The invention discloses a method and a device for extracting an urban heat island range. The method comprises the steps of acquiring a thermal infrared image corresponding to a research urban, and performing a surface temperature inversion of the thermal infrared image by using a single window algorithm to obtain surface temperature and generate a surface temperature image; dividing the surface temperature image into segmentation units according to the surface temperature, shape, and texture of the surface temperature image, and calculating average surface temperature inside the segmentation units; according to the average surface temperature, and calculating surface temperature aggregation features of the segmentation units by using a local autocorrelation algorithm; according to the surface temperature aggregation features, performing clustering classification on the segmentation units, and extracting size and boundary information of a research city heat island. The method can achieve fast and complete extraction of urban heat island range information based on the thermal infrared remote sensing image.
Owner:GUANGZHOU UNIVERSITY
Improved latent Dirichlet allocation-based natural image classification method
InactiveCN103870840AComplete feature information extractionImprove average classification accuracyCharacter and pattern recognitionVisual dictionaryClassification result
The invention discloses an improved latent Dirichlet allocation-based natural image classification method, and mainly aims to solve the problems that the existing entire-supervision natural image classification method has a long classification time and the classification accuracy is degraded on the premise of shortening the classification time. The improved latent Dirichlet allocation-based natural image classification method has the implementation steps: performing dense grid sampling on each natural image to get grid sampling points thereof; extracting SIFT (scale-invariant feature transform) features of each grid sampling point; performing K clustering on the SIFT features to generate a visual dictionary; performing quantification on the natural images into visual documents by virtue of the visual dictionary; constructing a two-layer space pyramid for each visual document to obtain five visual documents; inputting the five visual documents of each natural image into an LDA model to obtain five latent semantic theme distributions; connecting the latent semantic theme distributions of all the natural images in sequence and then inputting to an SVM classifier for classification, to obtain the classification result. Compared with the classical classification method, the improved latent Dirichlet allocation-based natural image classification method has the advantage that the classification accuracy is increased while the average classification time is shortened. The improved latent Dirichlet allocation-based natural image classification method can be used for target recognition.
Owner:XIDIAN UNIV
Dynamic object detection and static map reconstruction method of dynamic environment hybrid vision system
ActiveCN112132857AAchieve adaptiveAchieving Adaptive Threshold TargetsImage enhancementImage analysisPoint cloudMotion vector
The invention provides a dynamic object detection and static map reconstruction method of a dynamic environment hybrid vision system. The method comprises the steps that S1, conducting external parameter calibration to acquire coordinate transformation parameters between a panoramic camera and a three-dimensional laser sensor; S2, projecting the t-th frame of point cloud as a feature point to thet-th frame of image, obtaining a pixel motion vector of the feature point, and estimating an artificial motion vector of the feature point caused by the motion of the trolley to perform background motion compensation, thereby obtaining a dynamic point in the point cloud; S3, performing cluster segmentation on the point cloud of the current frame; S4, judging through the proportion of dynamic points in the cluster by utilizing the unique characteristic of each point index in the point cloud data and combining the dynamic point detection result and the segmentation result, and extracting a dynamic object; S5, reconstructing the static map by using an octree map tool and the laser radar odometer under the frame; Dynamic object extraction and static three-dimensional map reconstruction can beperformed robustly and more completely.
Owner:FUZHOU UNIV
High-spectrum image texture analysis method based on V-GLCM (Gray Level Co-occurrence Matrix)
The invention discloses a high-spectrum image texture analysis method based on a V-GLCM (Gray Level Co-occurrence Matrix), comprising the following steps of: selecting high-spectrum image data needing to be subjected to texture analysis; carrying out gray level range conversion on an original image; normalizing a gray level value to a certain range; selecting a suitable movable cubic window size and an angle parameter; taking statistical index information inside a movable cubic body as a texture characteristic of a cubic center picture element; utilizing a picture element relation in a movable cubic window to establish a co-occurrence matrix; carrying out index quantification counting on the established co-occurrence matrix and backfilling to the central position of the current movable cubic window, namely replacing the texture characteristic of the position; and continuously moving the cubic window and carrying out texture calculation and extraction on the whole image to obtain a V-GLCM texture image. According to the high-spectrum image texture analysis method disclosed by the invention, the image texture extracted by the method considers a relation between adjacent wave sections of a high-spectrum image, contains the texture characteristic of the adjacent wave sections, and can sufficiently represent the special properties of the high-spectrum data.
Owner:HOHAI UNIV
Remote sensing image fishpond extraction method based on row-column self-attention full convolutional neural network
ActiveCN111860351AAccurate extractionExtract completeScene recognitionNeural architecturesEngineeringComputer vision
The invention discloses a remote sensing image fishpond extraction method based on a self-attention full convolutional neural network, and the method comprises the basic steps: calculating an NDWI through employing a remote sensing image, carrying out the preliminary classification through employing the NDWI, and then manufacturing a remote sensing image sample set and a calibration sample set; constructing a self-attention full convolutional neural network and training a model, and predicting the remote sensing image by using the trained model to obtain probability distribution of each pixelcategory of the remote sensing image; and finally, fusing the probability distribution of each pixel category with an NDWI classification result to obtain a classification result. On the basis of using a full convolutional neural network, a row-column bidirectional GRU self-attention model is designed, the method has the characteristics of high accuracy and complete edge for fishpond extraction ina remote sensing image, and the method can be applied to fishpond automatic extraction, change detection and the like.
Owner:CHINA UNIV OF PETROLEUM (EAST CHINA)
Method of quickly extracting rivers from remote sensing image
InactiveCN101576957AHigh speedReal-time processingCharacter and pattern recognitionElectromagnetic wave reradiationImage extractionLarge size
The invention discloses a method for quickly extracting rivers based on seed point scanning, and relates to the technical field of multispectral remote sensing image processing. The method mainly solves the problem that the prior method cannot extract rivers quickly from a large-size remote sensing image. The method comprises the following steps: carrying out manual selection of original seed points and automatically dividing a river into two parts according to the different positions of the original seed points; and scanning the two parts from the original seed points, constantly updating the seed points and a scanning mode thereof according to the flow direction of the river, and carrying out the scanning till the river scanning is completed. During the scanning, if a branch of the river appears or the flow direction changes or the river is cut off by a bridge, subsequent seed points and scanning mode are determined through windowed searching. The method can extract rivers quickly from an image with the size exceeding 50 million pixels, and has the advantage of high precision; therefore, the method can be used for quick extraction of river areas, bridge detection and water target detection from a multispectral remote sensing image.
Owner:XIDIAN UNIV
An edge detection and extraction algorithm for standard water model of magnetic resonance imaging
ActiveCN109461163ASuppress false edgesImprove recognition rateImage enhancementImage analysisWater modelIdentification rate
The invention discloses an edge detection and extraction algorithm for magnetic resonance standard water model, Firstly, the pixel difference between true and false edges is enlarged by spatial enhancement, Then the initial adaptive canny algorithm and area sorting are used to eliminate most of the pseudo-edges, After that, the edge of each layer of spherical shell is obtained by closed region reduction and quadratic adaptive canny algorithm. Finally, the final image is classified by convolution neural network, and the edge required by the final result is obtained. After edge detection by canny operator, pseudo-edge can be suppressed effectively, By using the secondary canny operator to detect the edges after restoring the image in the closed area, the edges that the first canny can not detect can be detected effectively. This method has the advantages of high recognition rate and short time, and can reduce the error in the calculation process, so that the edge information of the standard water model can be extracted completely and clearly.
Owner:HENAN NORMAL UNIV
Single-photon laser height measurement data denoising method and device based on density statistics
ActiveCN111665517AImprove denoising effectImprove accuracyCharacter and pattern recognitionElectromagnetic wave reradiationComputational physicsLaser altimetry
The invention relates to a single-photon laser height measurement data denoising method and device based on density statistics. The method comprises the following steps: 1) obtaining single-photon laser altimetry data; 2) performing data preprocessing; 3) carrying out photon density calculation by using an ellipse search area whose direction and size change adaptively with the gradient; and 4) adaptively calculating segmented threshold separation signal photons and noise photons by adopting an OTSU method, and removing discrete noise points by adopting a K nearest neighbor distance. Compared with the prior art, the method is not influenced by non-uniform background noise photons, and is high in adaptability, the direction of the search ellipse can be changed adaptively along with the slope, the size can be changed, signal photons and noise photons of each segment are separated by using the segmentation threshold, and a small number of discrete noise points left after rough denoising are further removed; and the method is complete in signal extraction, high in denoising precision and high in efficiency.
Owner:TONGJI UNIV
Resource recommendation method, device and equipment and storage medium
PendingCN111737573AImprove effectivenessImprove accuracyDigital data information retrievalGeometric image transformationRecommendation modelResource information
The embodiment of the invention discloses a resource recommendation method, a recommendation model training method and device, and a storage medium, and belongs to the technical field of recommendation. The method comprises the following steps: acquiring resource-related characteristics of a target resource, wherein the resource-related characteristics comprise a resource identifier, resource attribute information and resource environment information; determining a resource recommendation probability of the target resource through a resource recommendation model based on the resource related characteristics, wherein the resource recommendation probability is used for indicating a possibility that the user accepts resource recommendation, and the resource recommendation model is a network model adopting a self-attention mechanism; and performing resource recommendation based on the resource recommendation probability. According to the embodiment of the invention, when resource recommendation is carried out, the resource attribute information and the resource environment information are fused, so the resource recommendation model can more completely extract the resource information based on a self-attention mechanism by combining the information, and the prediction accuracy and the effectiveness of resource recommendation are further improved.
Owner:BEIJING SANKUAI ONLINE TECH CO LTD
Quality control method for Huang's sound pill
ActiveCN104483399AControl feeding stabilityAdd content determination methodComponent separationQuality controlContent determination
The invention relates to a quality control method for Huang's sound pill. The quality control method employs liquid chromatography to perform content determination, and the content processes comprises determination processes for free anthraquinone, combined anthraquinone, forsythol, ferulic acid and glycyrrhizic acid. The quality control method is sensitive and fast, is capable of timely accurately reflecting the quality of nonvolatile main compositions in Huang's sound pill and improving production monitoring level, and has more practical application meaning in industrial production.
Owner:WUXI JIMIN KEXIN SHANHE PHARMA
JPEG image robustness steganography method based on serial concatenated code and DCT coefficient relative relationship
InactiveCN108024028AResistance to JPEG compression attacksImprove error correction performancePictoral communicationComputer hardwareHigh probability
The invention relates to the technical field of information hiding, and particularly relates to a JPEG image robustness steganography method based on a serial concatenated code and a DCT coefficient relative relationship. The JPEG image robustness steganography method based on the serial concatenated code and the DCT coefficient relative relationship comprises the steps of determining a modification area; designing a corresponding distortion function; encoding a to-be-embedded message by using an error correcting code; embedding the error correcting code through robustness steganography; performing JPEG compression; obtaining an information stream containing errors by using steganographic translated code, and correcting errors by using a translated code of the error correcting code; and obtaining the original embedded message. JPEG compression attacks of different quality factors can be defended in a larger range, the embedded message can be extracted completely at a higher probability, further, for same secret information, the method has higher detection resistance.
Owner:PLA STRATEGIC SUPPORT FORCE INFORMATION ENG UNIV PLA SSF IEU
Intravenous grain extraction method based on maximal intra-neighbor difference vector diagram
InactiveCN101751564AExtract completeEliminate distractionsCharacter and pattern recognitionFeature extractionHistogram
The invention discloses an intravenous grain extraction method based on maximal intra-neighbor difference vector diagram, which firstly calculates the maximal intra-neighbor difference and directions of all pixel points in an original intravenous image to construct a maximal intra-neighbor difference vector diagram, and further calculates the maximal intra-neighbor vector difference of the maximal intra-neighbor difference vector diagram, finally obtains proper thresholds from the histogram of the maximal intra-neighbor vector difference diagram and extracts the intravenous grain of the original image from the maximal intra-neighbor vector difference diagram. The invention avoids the image pretreatment process, also can eliminate the influence of intravenous image coherent features on grain extraction such as unevenness and fuzzy boundary, can completely extract the intravenous grain and ensures the grain to be even and have less noise points, and also greatly facilitates the following feature extraction.
Owner:SOUTH CHINA UNIV OF TECH
Multifeature-based target object contour detection method
ActiveCN102201120BEasy extractionExtract completeImage analysisCharacter and pattern recognitionVisual perceptionBrightness perception
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA
Floc detection method combining three-frame differential higher-order statistics (HOS) with OTSU algorithm
ActiveCN103632373AExtract completeReal-time extractionImage analysisPattern recognitionImage post processing
Disclosed is a floc detection method combining three-frame differential higher-order statistics (HOS) with an OTSU algorithm. Since a commonly used floc tracking method is susceptible to noise, light and floc movement speed and the like at present and complete characteristics of a floc movement target are hard to extract, a floc target detection method based on combining the three-frame differential HOS with a particle swarm optimization enhancing OTSU algorithm is put forward. The floc detection method includes: subjecting consecutive three-frame images to differential calculation; calculating fourth-order moment pixel by pixel and comparing the fourth-order moment with threshold values; utilizing particle swarm optimization enhancing OTSU algorithm to acquire a best threshold value; utilizing the best threshold value to perform image binaryzation and image post-processing to acquire a relatively clear floc target finally. Therefore, a foundation of automatic subsequent floc analysis is laid. The floc detection method has the advantages of being accurate and rapid, capable of effectively extracting the floc target and applicable to effective extraction of the floc target in water treatment.
Owner:EAST CHINA JIAOTONG UNIVERSITY
Domain name traffic feature extraction method, device and equipment and readable storage medium
ActiveCN111866196AHigh precisionExtract completeNeural architecturesTransmissionDomain nameFeature extraction
The invention discloses a domain name flow feature extraction method and relates to the field of network security. A structural feature and context dependency relationship of the domain name flow waveform sequence are learned by utilizing a neural network; the characteristics of abstract and high-level flow features can be extracted; statistics and normalization processing is performed on the domain name access flow data; an input sequence capable of performing feature extraction is gennerated; the abstract and high-level flow characteristics of the domain name are automatically extracted through a trained sequence-to-sequence model. Compared with a traditional manual feature extraction means, the method has the advantages that the context dependence relation and the structural features corresponding to the time sequence formed by the domain name flow features can be extracted completely, and domain name flow feature extraction accuracy is improved. The invention further discloses a domain name traffic feature extraction device and equipment and a readable storage medium, which have the above beneficial effects.
Owner:SANGFOR TECH INC
Production method of traditional Chinese medicine fertilizer
InactiveCN107382474ASafe, non-toxic and harmlessAchieve the same purposeWaste based fuelOrganic fertilisersChinese drugDisease injury
The invention relates to a production method for traditional Chinese medicine fertilizer. A traditional Chinese medicine extract solution or traditional Chinese medicines is or are added to fertilizer, and the traditional Chinese medicine fertilizer is prepared. With addition of the traditional Chinese medicine extract solution or the traditional Chinese medicines, the fertilizer has bacteriostatic and bactericidal effects and can control crop diseases.
Owner:王娟
Papaya anti-oxidant serum and preparation method thereof
InactiveCN101444562AEasy to carry and storeImprove extraction efficiencyAntinoxious agentsPharmaceutical non-active ingredientsGramHealth food
The invention discloses a papaya anti-oxidant serum, which comprises cyclodextrin inclusion compound of pawpaw natural antioxidant and carriers accepted in pharmacy. The papaya anti-oxidant serum consists of enzymolysis water extract and alcohol extract of the papaya, in which vitamin C and polyphenols are contained. Vitamin C contained in papaya anti-oxidant serum is no less than 2 milligram per gram, and the total polyphenols calculated in gallic acid is no less than 22 milligram. Therefore, the papaya anti-oxidant serum has the characteristics of high efficiency, durability and stable inoxidizability and easy storage and carrying, can be served as health food to eat, can used as cosmetics, animal food, food or medical mediate, and further can used for preparing antioxidation products; the invention also discloses a preparation method for the papaya anti-oxidant serum, which comprises seven steps of pulping, enzyme digestion, filtering, alcohol extraction, concentration, inclusion and pelletization, so as to be a system for efficiently, sufficiently and completely extracting papaya anti-oxidant, and the antioxidative activity can be effectively preserved.
Owner:ARMY MEDICAL UNIV
Electric power inspection intelligent defect detection method based on deep learning
PendingCN111862078AShort processing timeHigh precisionImage enhancementImage analysisNetwork modelEngineering
The invention discloses an electric power inspection intelligent defect detection method based on deep learning. The method comprises the following steps: obtaining a plurality of original images of different insulators and dividing the original images into a training set and a test set; carrying out enhancement processing on the original image of the training set to obtain an enhanced set image;dicing the original images in each enhancement set image and each test set to obtain a plurality of sub-block images and masks thereof; carrying out the semantic segmentation on each sub-block image and the mask thereof, and extracting an insulator region; obtaining a communication area of each insulator; rotating the connected region by using principal component analysis to obtain a normalized insulator image; inputting the normalized insulator image of the enhanced set image into a neural network model for training to obtain a training model; predicting insulator coordinates in the normalized insulator images of the test set through the training model; and carrying out the inverse transformation on the insulator coordinates to restore to the original image coordinates. The method can achieve the recognition and segmentation of the insulator string, is short in processing time, is high in precision, and is high in robustness.
Owner:JIAYING UNIV
Method and system for extracting lines from image
The invention discloses a method and a system for extracting lines from an image so as to solve the problem that the lines are incompletely and inaccurately extracted in the prior art. The method comprises the following steps of: obtaining a difference image according to color difference between the image and a preset sample point; performing binarization on the difference image to obtain a candidate line area image; determining framework lines which accord with preset conditions in the candidate line area image; expanding the framework lines which accord with the preset conditions in a preset width; performing reference filtering on the expanded framework lines by taking the image as a reference; and performing binarization on the filtered image to obtain the lines in the image. Due to the technical scheme, the accuracy of extracting the lines is improved.
Owner:方正国际软件(北京)有限公司 +1
Who we serve
- R&D Engineer
- R&D Manager
- IP Professional
Why Patsnap Eureka
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com