140results about How to "Efficient and Accurate Acquisition" patented technology

The present invention provides an advertisement click rate estimation method based on a decision tree and a device. The method comprises a step of obtaining the related characteristic information of a specific historical advertisement in a predetermined historical time period, a step of obtaining the personalized characteristic information of a target user, a step of carrying out decision tree model calculation based on the obtained target user personalized characteristic information and the related characteristic information of the specific historical advertisement so as to determine a cross characteristic vector for estimating the click rate of the specific historical advertisement, wherein each leaf node of the decision tree model represents one cross characteristic, a step of calculating based on the cross characteristic vector and the model training parameter obtained from training in advance so as to estimate the click rate of the specific historical advertisement. The invention also provides an application software recommendation method and a device. According to the method or the device, through the decision tree model, the cross characteristic with a strong classification characteristic is generated, and thus the advertisement click rate is accurately and rapidly estimated.

The invention relates to a system for obtaining a defect part model of a damaged part. The system comprises a computer and a scanner electrically connected with the computer. A modeling method comprises the following steps of (1) performing pre-scanning preparation on a to-be-scanned damaged part; (2) calibrating the scanner; (3) performing parameter configuration by the system; (4) performing scanning through the scanner to obtain initial model data; (5) processing the initial model data to obtain point cloud model data; (6) processing the point cloud model data through a mesh reconstruction algorithm to obtain a curved surface reconstruction model; (7) performing registration alignment on the curved surface reconstruction model and a standard part model; and (8) obtaining the defect part model of the to-be-scanned damaged part: after the registration alignment of the models, performing subtraction Boolean operation on the curved surface reconstruction model and the standard part model to obtain the digital model of the defect part of the to-be-scanned damaged part. The modeling method is high in modeling efficiency and high in precision.

The invention provides an idle frequency spectrum test method, a device and a system, wherein, the idle frequency spectrum test method comprises the following steps: secondary equipment receives downgoing signals of a base station of main equipment; and the downgoing signals are decoded according to the identification information for identifying resource allocation information of the main equipment in the downgoing signals to obtain the idle frequency spectrum resource information of the main equipment. The above idle frequency spectrum test method, the device and the system can accurately andefficiently obtain the idle frequency spectrum resource information of the main equipment, and effectively increase the use ratio of frequency spectrum by decoding the downgoing signals of the main equipment,; meanwhile, the interference of the secondary equipment to the main equipment are better prevented by ways of ensuring that the transmitted power of the secondary equipment is less than thetransmitted power of the main equipment or setting a protective belt and the like.

The invention provides a template-based CCD-DR (charge coupled device-digital radiography) image splicing method, comprising the following steps of: 1) collecting the template and a real object DR image in a certain visual field range by using a CCD array; 2) pre-processing the acquired CCD-DR image; 3) extracting characteristic points from the template DR images; 4) performing registration to the characteristic points of the adjacent template DR images; and 5) splicing a plurality of real object DR images which are collected by the CCD array according to the registration relationship of characteristic points acquired in the step 4). The method performs registration to the characteristic points of the template DR images, determines the position change relationship between the adjacent DR images in the CCD array and stores the related image registration information; under the condition that the CCD-DR image collection conditions are unchanged, the image registration information is repetitively transferred to splice the real object DR images so as to highly-efficiently acquire panoramic DR images with high resolution and large visual field.

The invention belongs to the field of text data mining and provides a web text-based acquiring and screening method of seismic macroscopic anomaly information, applied to collecting and screening seismic macroscopic anomaly text information from the internet. According to the method, based on a Heritrix frame, by the use of a seismic macroscopic anomaly subject descriptor group, a crawling strategy from seismic macroscopic anomaly subject relevancy judging and link ordering to information extraction is customized for three information sources, namely common webs, post bars and social networks, and subject related webs crawled are subjected to information screening mainly from three aspects, namely subjective sentence judging, text subjectivity judging and seismic macroscopic anomaly matching. The method has the advantages that online collection of seismic macroscopic anomaly information is provided with a scientific, efficient and accurate technical means and information acquisition efficiency is greatly improved.

The invention belongs to the technical field of intelligent analysis, the invention relates to a NLP-based method for automatic extraction and analysis of enterprise supply relationship, including company name identification, product name recognition and supply-demand relationship judgment, improves the Stanford NER toolkit with rules and dictionaries, the recognition effect of the tool to the company abbreviation is effectively improved. Besides, multiple features are integrated into a conditional random field model and in combination with the rule post-processing, the product category wordsin the text can be identified effectively, which can effectively and accurately obtain the supply-demand relationship information between enterprises from the text, and provide valuable reference information for investors and consumers. The invention has important theoretical research and practical application value.

User portrait obtaining method, apparatus, and storage medium are provided. The method includes: obtaining M training samples according to a user behavior log. An initialized user parameter matrix W_m×k and an initialized tag parameter matrix H_k×n are modified according to the M training samples by using a data fitting model, to obtain a final user parameter matrix W_m×k and a final tag parameter matrix H_k×n. Further, a user portrait matrix P_m×n is obtained according to the final user parameter matrix W_m×k and the final tag parameter matrix H_k×n. In the present disclosure, a user and a tag are parameterized, and a user parameter matrix and a tag parameter matrix are modified by using the data fitting model, so as to fit a training sample.

The invention discloses an automatic search engine method and system based on network public opinion analysis, the system comprises a platform positioning analysis module, a public opinion analysis module, a theme screening module and a content pushing module, and the system can capture hotspot content according to customized keywords of a platform to complete content analysis of the system and anexternal platform; meanwhile, curves of propagation paths, operation data, emotional trends and the like of the content are displayed, and complete and scientific propagation analysis is displayed; by tracing the captured content, the source of the content is accurately tracked to realize accurate traceability analysis, and copyright protection is facilitated; through professional and intuitive big data visual display, a statistical report of the system is made according to large-screen display, so that a manager can perform data operation conveniently.

The invention discloses an unsupervised remote sensing image change detection method, a storage medium and computing equipment. The method comprises the following steps: constructing a multi-scale image convolutional neural network; respectively inputting the dual-temporal images into a multi-scale image convolutional neural network, extracting spatial features and inter-spectrum features, and jointly calculating to generate an initial pseudo label; cascading the two images of the dual-temporal image, inputting the two images into a multi-scale image convolutional neural network, and trainingthe multi-scale image convolutional neural network to generate a two-channel difference image; utilizing a metric learning module of the multi-scale graph convolutional neural network to update the initial pseudo label as a label of a two-channel difference graph, and training the generated two-channel difference graph; and comparing two channels of the trained two-channel difference image to obtain a binary change image with the same size as the original image, and completing image change detection. According to the method, the change detection graph of the pair of dual-temporal images can beefficiently and accurately obtained in an unsupervised manner.

This invention belongs to the field of electronic information, especially involving ultra-high-resolution camera imaging detection system and method. This invention provides the ultra-high-resolution camera imaging detection System. It has multi-channel data acquisition of massive image which connects separately with multi-channel photosensitive element of camera, also it has graphics workstation. Each data acquisition channel of massive image includes: Signal matching modules, interface circuit, I / O control modules, front-end data receiver module, mass storage equipment control module, massive image processor as fast. The graphics workstations connect with the data acquisition channel of massive image, which is used to control the whole system, display the full-channel image and do various functional operations. This invention also provides an ultra-high-resolution imaging camera detection method, it sends the data from the multi-channel data acquisition of massive image to the graphics Workstation at which processes the integrated treatment to the acquisition data, and achieves the goal of display and zoom of the overall image.

The invention relates to a three-dimensional shape and texture reconstruction method and system based on a single image and a storage medium, and solves the problem that three-dimensional face data cannot be efficiently and accurately obtained. The method comprises the following steps: acquiring face image information; analyzing the face image information through a normalized coordinate coding prediction network, and mapping the analyzed data information to a UV space to form UV space texture mapping information; analyzing the UV spatial texture mapping information through a filling network toobtain complete texture map information; analyzing the complete texture map information through a shape network to obtain UV position map information; searching face three-dimensional shape coordinate information corresponding to the current UV position map information from a preset corresponding relationship between reference UV position map information and reference face three-dimensional shapecoordinate information; and forming three-dimensional texture face information according to the searched face three-dimensional shape coordinate information. According to the invention, the requiredreconstructed three-dimensional face data can be efficiently and accurately obtained.

The embodiment of the invention provides a risk identification method, device and system. The risk identification method includes the steps: when a service request is received, sending interaction information used for judging a service risk to a user, determining interaction information sent subsequently based on a feedback result of the user for the sent interaction information, and then comprehensively judging the risk level of the service request in combination with historical data of the user and the feedback result for the interaction information. The risk identification method performs risk interaction with a user in the transaction process, and adopts a progressive interaction mode, so that key data used for judging the risk of a service request can be efficiently and accurately obtained in the interaction process, and carries out comprehensive judgment in combination with historical data of the user, thus being able to obtain a more accurate risk recognition result.

The invention belongs to the technical field of image processing and discloses an image segmentation method of field crop weeds. The image data of the field crop weeds is collected by an image acquisition module using a camera. The main control module enhances image definition by image processing software through image enhancement module. The weed image is recognized by using the depth stack network model through the image recognition module. A feature extraction module extracts and segments the weed image data according to the identification content by using the image processing software; Thewarning module uses the warning device to warn according to the extracted weed data; Storing the collected weed image data through a data storage module by using a memory; And the weed image data isdisplayed by the display module. Deep-stacking network adopt by that image recognition module of the invention not only has strong feature expression and classification ability, but also reduces traintime and recognition time, and enhances robustness of weed recognition.

The invention relates to an unsteady cavitation flow field visualization test data processing method, and belongs to the multiphase fluid mechanics technical field; the method can fully considers the cavitation flow unsteady characteristics, thus improving the visualization test data processing efficiency, and reducing cost; the method can make statistics analysis for an unsteady cavitation flow field structure, thus accurately and efficiently obtaining the unsteady cavitation flow field characteristic parameters like a cavity shedding period, a cavity area and a cavitation vortex strip average length, and improving the unsteady characteristic display effect.

The invention relates to a pyramid face image super-resolution reconstruction method based on a regression model, and relates to the enhancement or restoration of an image. The feature that images have a non-local similarity is used to search low-resolution face images in a test set in corresponding feature images thereof for similar blocks of reconstruction image blocks to obtain a position set of all similar blocks. Face image blocks in the position set of all low-resolution images in a training set are used as a low-resolution training set corresponding to the low-resolution face image blocks in the test set. The sum of the distances between feature image blocks corresponding to the low-resolution face image blocks in the test set and feature image blocks corresponding to the low-resolution face image blocks in the training set and the distances between feature image blocks corresponding to the face image blocks after the low-resolution images in the test set are interpolated and amplified and the feature image blocks corresponding to high-resolution face image blocks in the training set constitutes constraint conditions. The method overcomes many deficiencies in the face imagereconstruction process of the prior art.

The invention relates to a riverway sewage draining exit tracing method based on gridding water quality monitoring, which comprises the following steps of dividing a river into a plurality of river reaches, carrying out gridding water quality monitoring on each river reach, and obtaining monitoring data of each river reach, according to the monitoring data of each river reach, acquiring the discharge water quantity of the sewage draining exit of each river reach, and achieiving quantitative accounting of the pollutant discharge amount, and obtaining the river reach with sewage discharge according to the quantitative accounting result, encrypting the gridding water quality monitoring of the river reach with sewage discharge, and gradually narrowing the troubleshooting range to realize the source tracing of the riverway sewage draining exit. Compared with the prior art, the method solves the problem that on-site troubleshooting is difficult due to pollution discharge amount accounting, underwater discharge and the like of the sewage draining exit, and provides a relatively convenient traceability troubleshooting means for the riverway sewage draining exit.

The invention discloses a measurement method and a measurement device in a soil freezing-thawing process. The device comprises an operating rod, an ice water dielectric sensing module, a depth sensing module, a PVC plastic tube, a data transmission line, a data acquisition module, wherein the ice water dielectric sensing module is fixed at the bottom end of the operating rod and is used for measuring an ice water dielectric constant of a soil profile in real time; the depth sensing module is embedded inside the operating rod and is used for measuring current depth data; the PVC plastic tube is buried in the measured soil and serves as an operating guide rail of the depth sensing module; the data transmission line is used for transmitting the ice water dielectric constant and the depth data to a data acquisition module; and the data acquisition module is used for judging the current soil freezing-thawing condition according to the received data. The measurement device can rapidly, conveniently, accurately and effectively acquire ice water content of the soil, acquire the depth data of a measuring point and bring the accurate data to the analysis of the soil freezing-thawing condition. The mode of acquiring data is flexible and accurate and the cost is lower.

The invention provides a bubbling device and a measuring method for measuring Young modulus of a two-dimensional material, and the bubbling device comprises a pressurizing system and a pressure chamber connected in sequence, wherein the top of the pressure chamber is provided with an upper air port; a bearing substrate for carrying the two-dimensional material to be tested is fixed at the upper air port; and a through hole is defined at a position corresponding to the upper air port of the bearing substrate. The bubbling device provided by the invention implements a wide range of pressure regulation, controllable deformation and continuous strain loading on a two-dimensional material through the bubbling device designed and processed by itself, the in-situ characterization of the bubblingmorphology is carried out with an atomic force microscopy, and a Young modulus value of corresponding material is calculated based on the thin film theory.

The invention discloses a container damage detection method and system. The method comprises the following steps that: the target image and point cloud data of a to-be-detected plate are obtained, pixel point coordinates in the target image are in one-to-one correspondence with the point coordinates of the point cloud data; the depth map of the to-be-detected plate is obtained based on the point cloud data; the target image and the depth map are inputted into a trained neural network model, and the edge coordinates of a damaged area are outputted; based on the edge coordinates of the damaged area, a target image where the damaged area is marked is obtained; and based on the target image where the damaged area is marked and the point cloud data, the quantitative data of damage in the to-be-tested plate is obtained. With the method adopted, the accuracy of damage identification and positioning can be improved. Based on accurate damage positioning, the point cloud data can reflect the depth condition of the surface of the to-be-detected plate, so that the quantitative data of the damage can be accurately and efficiently obtained on the basis of the accurate damage positioning.

The invention discloses a reconstruction method of a large-scale scene three-dimensional reconstruction system based on a BDS location service, which is characterized in that the method comprises thefollowing steps: S1, obtaining scene images by positioning a scene image through a BDS location service module and establishing a panoramic image acquisition system; S2, carrying out data processing and storing the data into a database; S3. modeling in Grassmann space; S4. performing texture and color mapping. This method can obtain real-time scene image, match accurately, and quickly build 3D model. It can overcome the problem of large accumulated error and huge data overhead space in traditional 3D reconstruction of large-scale scene.

The invention provides a timing sequence full-polarization synthetic aperture radar image unsupervised change detection method considering statistical characteristics, and mainly solves the problems of insufficient polarization information utilization and underfitting of differential image probability statistical distribution in an existing method. According to the method, rich spatio-temporal information provided by long-time sequence PolSAR images is fully utilized, comprehensive test statistics and an improved minimum error threshold selection method are combined, and ground object change detection information of a research area is extracted. All parameters can be automatically acquired, the execution speed is high, rich space-time information of a time sequence PolSAR image can be fully utilized, and a change detection result is efficiently and accurately acquired. Meanwhile, the method is not only suitable for time sequence PolSAR image change detection, but also can be well applied to change detection of single-polarization and dual-polarization SAR data, and provides a basis for large-range and long-time change detection of an observation area.

The invention provides a method and a device for determining a rock softening curve in a bonding crack model, which can completely and accurately reflect the three-section change of the rock softeningcurve, can be simultaneously suitable for soft rock and hard rock, and are high in fitting precision and free of human factor interference in the whole process. The rock softening curve determinationmethod comprises the following steps: step 1, acquiring peak stress, peak strain, residual stress and residual strain of a rock to be measured; 2, dividing rock uniaxial tensile test data into soft rock and hard rock according to different rock brittleness indexes B; 3, normalizing the post-peak stage of the rock stress-strain curve; 4, fitting the normalized data by adopting a power-exponentialsoftening function, and determining fitting parameters suitable for soft rock and hard rock respectively; 5, substituting the fitting parameters into the power-index softening function to obtain a rock softening curve in the bond crack model;

The invention discloses a high-precision indoor wireless positioning method. The method comprises the following steps: A, obtaining distances from all N anchor nodes to a measured node and coordinates of the anchor nodes at a moment t; B, mining spatial correlation of adjacent moments through Kalman filtering, defining Kalman state equation coefficients on three coordinates of x, y and z at the moment t, and summarizing vectors and matrixes at all moments into a specified set form; C, performing factor decomposition on the global probability distribution by utilizing a total probability formula and a Kalman state equation; D, modeling by utilizing the factor decomposition obtained in the step C, and dividing the factor graph model into a lattice network part, a Kalman network part and a Kalman parameter estimation part; and E, carrying out message calculation on the lattice network part, the Kalman network part and the Kalman parameter estimation part through a message transmission algorithm. According to the method, the high-precision position information of the indoor to-be-tested node can be obtained efficiently and accurately.