49results about How to "Improve match quality" patented technology

A method for linking related data, such as metadata, from at least two data sources. The method includes formatting items of data of the data sources according to attributes. The method also executes a scoring algorithm for one or more of the attributes to generate a score for one or more sets of the formatted items of data, each of the sets includes an item of data from one data source and an item of data from another data source. Finally the method identifies related items of data of the separate data sources based upon the generated scores to facilitate linking related data of the two data sources. The method may also provide a link between data items of the data sources.

A method for linking related data, such as metadata, from at least two data sources. The method includes formatting items of data of the data sources according to attributes. The method also executes a scoring algorithm for one or more of the attributes to generate a score for one or more sets of the formatted items of data, each of the sets includes an item of data from one data source and an item of data from another data source. Finally the method identifies related items of data of the separate data sources based upon the generated scores to facilitate linking related data of the two data sources. The method may also provide a link between data items of the data sources.

The invention relates to the technical field of computer vision. The embodiment of the invention provides a depth image acquisition method and system based on speckle structured light, wherein the method comprises the following steps: acquiring speckle reference image and speckle scene image, wherein the speckle reference image is related to a preset reference plane and the speckle scene image isrelated to a target plane; The obtained speckle reference image and speckle scene image are extracted by ellipse LBP to obtain the corresponding first feature texture image and the second feature texture image. Stereoscopic matching between a speckle reference image and a speckle scene image based on the first feature texture image and the second feature texture image to output a corresponding parallax image; And optimizing the parallax image and combining the internal parameters of the camera to determine a corresponding depth image. Thus, the influence of noise and light is avoided, and theoverall effect of the obtained depth image is improved.

The invention discloses a cell characteristic vector based lithium ion cell configuration method, which comprises the following steps of: carrying out one-time 0.2C discharge on cells to be configured, followed by 2C charge and discharge, finally standing for a while, recording the 2C discharge capacity C0.5, the ratio CI of constant voltage capacity to constant current capacity during the 2C charge, the ratio V0 of recovery voltage to discharge lower limit voltage after the final step of standing, and the correlation coefficient rYS of the characteristic vector formed by the 2C constant current charge and 2C constant current discharge voltage values of the cells and the standard characteristic vector, determining an appropriate grading numerical range for grading according to the C0.5, CI, V0 and Rys, and configuring individual cells within the specified range. The method provided by the invention is used to fully present the difference between cells and screen out cells with similar consistency for assembling and usage, thus raising the performance of battery packs.

The invention relates to a lithium ion battery group matching method, which includes: S1. subjecting a plurality of lithium ion batteries to electric discharge to cut-off voltage, and then performing standing; S2. conducting constant current charging on the plurality of lithium ion batteries to cut-off voltage, recording the charging capacity C1, the charging mean voltage V1 and the first voltage change value delta E1 before and after charging, and then conducting laying; S3. Performing constant-current discharge on the plurality of lithium ion batteries to cut-off voltage, measuring the discharge capacity C2, the discharge mean voltage V2 and the second voltage change value delta E2 before and after discharge, and then conducting laying; S4. conducting charging on the plurality of lithium ion batteries, then performing laying, and measuring the voltage drop delta V and internal resistance R; and S5. according to C1, V1, delta E1, C2, V2, delta E2, delta V and R, carrying out group matching based on grading standards. Through the lithium ion battery group matching method provided by the invention, group matching can better coincide with the actual state, and the use performance of a battery module can be better, thereby realizing safe and efficient use of the battery module.

The invention discloses a matching method of lithium ion batteries, relating to lithium ion batteries and being used for more fully reflecting the difference between unit batteries to be matched and improving the battery matching quality and the integral property of a battery pack after matched. The matching method comprises the steps of: discharging a to-be-matched battery in a full-energy state to reach a low limit voltage in a first constant current, recording a discharge capacity Q1; charging to reach an upper limit voltage in a second constant current, recording a charge capacity Q2 and a charge voltage of the battery at intervals of T; charging to a stop current by using the upper limit voltage in a constant voltage form , recording a charge capacity Q3; calculating a capacity quality coefficient lambda; calculating similarity of a characteristic vector of the unit lithium ion battery, meeting the formula of lambda>=lambda0, and a standard characteristic vector to obtain a correlation coefficient; and according to a value of the correlation coefficient, grading according to a grading numerical value range. The invention is used for matching the lithium ion battery.

The invention relates to a matching link of a plate blank of an iron and steel enterprise, belongs to the technical field of automation and particularly relates to a plate blank and contract transferring and matching method for improving the utilization rate of the plate blanks of the iron and steel enterprise, comprising the following steps of: downloading plate blank information and contract information; establishing a mathematic model; carrying out parameter setting on the mathematic model; grouping the plate blanks and contracts; carrying out tabu search algorithm on each group of contract data and each group of plate blank data based on a multi-neighborhood to obtain a matching relation again; and checking a result of the step five; if the result meets the requirements, delivering and executing. Aiming at the condition that a plate blank and contract matching relation is unreasonable, the invention provides the plate blank and contract transferring and matching method, namely the plate blanks in all the contracts are taken off to break through the previous plate blank and contract matching relation; and on the precondition that a unit capacity is balanced, optimization and matching of all the plate blanks capable of being matched and the contracts capable of being matched can be carried out again, so that the matching quality of the plate blanks is effectively improved, the cutting loss amount of the plate blanks is reduced, the residual material amount of the contracts is reduced, the integrity of the contracts is improved and the delivery just-in-time of the emergent contracts is guaranteed.

The present invention utilizes an integrated information structured abstract service system and service method to provide structured abstract service for network integrated information service. Said system includes structured abstract template of integrated information, structured abstract service function templete and integrated information structured abstract release and searching information storage database. The structured abstract template can be used for guiding the user to release and search various information abstracts, and the system server can be used for making automatic matching of the integrated information structured abstract information storage data by adopting various forms so as to raise and improve accuracy, simple and clear property and univer sality of integrated information release and searching service.

The present invention provides a graph pattern matching method for supporting a fuzzy constraint relation, which relates to the technical field of graph information query and is used for solving the problem, in the prior art, of failing to support matching of a fuzzy constraint relation on the basis of meeting all accurate constraints. The method comprises: by means of an input layer, acquiring a query graph Q input by a user and a stored target data graph G, wherein the query graph contains the accurate constraint relation and the fuzzy constraint relation; by means of an accurate matching layer, performing accurate map matching for the accurate constraint relation in the query graph input by the user in the target data graph, and constructing a guide graph with a merging node in the target data graph; and by means of a fuzzy search layer, finding out and outputting K matching graphs that have the best fuzzy constraint matching quality on the basis of meeting all the accurate constraints in the guide graph. The graph pattern matching method for supporting the fuzzy constraint relation is applicable for performing hybrid query supporting the fuzzy constraint relation and the accurate constraint relation in massive target data graphs.

The invention discloses a light field depth estimation method based on a splitting iteration algorithm, and belongs to the field of computer vision and light fields. The method comprises the following steps: A, correcting an input light field image to obtain a corrected light field image; B, obtaining a sub-aperture image I (x + [delta]x) of the other view angle of the corrected light field image I (x) after the displacement [delta]x occurs based on the Fourier phase shift theorem; C, performing stereo matching on all the sub-aperture images, and constructing an energy function to obtain an initial depth map; D, performing confidence calculation on the initial depth map, finding out inaccurate pixel points in the initial depth map, and optimizing the initial depth map; and E, obtaining an optimized depth map. According to the method, the problem that the baseline between the sub-aperture images is too narrow is effectively solved, the matching quality is improved, and the accuracy degree of the depth map is ensured as much as possible in the calculation process of the initial depth map.

The invention discloses a power lithium ion battery grouping method. The method comprises 1, selecting cells belonging to the same batch, same type and same material system, carrying out a charging and discharging test on the cells and removing the cells unqualified in platform time and a charging constant current ratio, 2, selecting the cells with capacities greater than the lower limit capacity as qualified cells, dividing the cells into N grades according to a certain capacity range M and dividing the cells into P grades according to a certain charging constant current ratio, 3, standing the uncharged cells in a T DEG C environment for t1 days and standing the cells at the room temperature for t2 h, 4, testing voltage and internal resistance, removing the cells of which voltage and internal resistance exceed a certain range and dividing the cells with the same grade of capacities and the same grade of charging constant current ratios into Q grades according to a certain voltage scope, and 5, grouping the qualified cells with the same grade of capacities, the same grade of charging constant current ratios and the same grade of voltage. The method utilizes dynamic and static combination, takes into full consideration of single cell difference, improves grouping accuracy and greatly improves battery performances.

The invention relates to the field of hot rolling pickling plates, in particular to a boracic low-carbon hot rolling pickling plate and a preparation method thereof. Boron with the content of 0.0010-0.0030% is added into a low-carbon hot rolling pickling plate, wherein the thickness is larger than or equal to 2.0 mm and less than or equal to 6.0 mm, and thus based on the ideal structural performance, multiple kinds of measures for controlling the surface quality can be taken, so that steel plate mixed crystals are effectively avoided, and the problem of rolled edge mixed crystals is especiallysolved. The boron is added into the low-carbon hot rolling pickling plate to reduce the transformation temperature from austenite to ferrite, so that structural transformation does not happen at thefinishing rolling temperature of 840-860 DEG C at a hot rolling stage, that is to say, no mixed crystal defects exist, finally, the even structural performance is obtained, and meanwhile, an advantageis created for inputting a large amount of descaling water to overcome scale defects at the hot rolling stage.

The invention discloses an image splicing method based on an improved ORB feature algorithm. The method comprises the following steps: constructing a Hessian matrix, and determining feature points ofeach to-be-spliced image; extracting a feature descriptor of each image to be spliced by using a BRIEF binary feature descriptor; performing rough matching by using a Hamming distance, and distinguishing correct matching point pairs from wrong matching point pairs by using a grid motion feature algorithm; removing error matching feature point pairs by using an improved random sampling consistencyalgorithm; searching an optimal suture line; dividing an overlapping region; and performing segmented fusion to complete image splicing. Through the multi-scale space theory, the Hessian matrix and the Gaussian pyramid are used for improving the registration method, the purpose of scale invariance is achieved, and meanwhile the matching precision of the matching point pairs is improved.

The invention discloses a visual odometer implementation method based on ORB feature points and a GMS matched filter. The visual odometer implementation method comprises the steps: employing an RGBD camera or a binocular camera for collecting video streams; for each frame of image, based on the current pose of the camera, obtaining feature points (including two-dimensional pixel coordinates, three-dimensional world coordinates and a brief descriptor of angular points corresponding to the feature points) in the image; matching the feature points between the current frame and the previous framein combination with a GMS matching filter; solving a PnP problem according to the matched feature points of the previous frame to obtain a camera pose of the current frame; simultaneously optimizing the camera pose and the three-dimensional world coordinates of the feature points of a plurality of nearest continuous frames by using a Bundle AdJustment; calculating a bag-of-word vector of the current frame by using a bag-of-word model, and judging whether the current frame needs to be stored as a key frame or whether loopback occurs or not; if loopback occurs, optimizing the related key framesby using fast loopback optimization. According to the method, under the condition of ensuring that the re-projection error of the camera is small enough, compared with a traditional singular value decomposition method, operation is simpler and more convenient, and real-time performance is ensured.

The invention discloses a friend relation chain matching method and a friend relation chain matching equipment in an instant messaging tool, belonging to the field of instant messaging. The method comprises the following steps of: analyzing data of material information of a user; mining data of material information of other users except the user according to an analysis result; and performing data matching of a mining result and the analysis result of the user. The equipment comprises a data analysis module, a data mining module and a data matching module. By the scheme, other users in which the user is interested can be automatically matched, manual intervention is not required in the whole matching process, and the application threshold of a friend relation chain matching system is reduced; moreover, matching is performed by taking user information as the center, the relevance of a matched user and the user is strong, matching quality is high, and improvement of the success rate of building a friend relation is promoted.

The invention discloses a power transmission line three-dimensional modeling method and system based on an oblique image. The method comprises the steps of obtaining the oblique image and the POS information of an unmanned aerial vehicle during a surveying and mapping scanning process; constructing a three-dimensional map model according to the oblique image; detecting and identifying a tower in the oblique image, and matching the identified tower with a tower model in a tower standing book system according to the POS information; and implanting the matched and returned tower model into a corresponding position in the three-dimensional map model to generate a three-dimensional model of a power transmission line. According to the power transmission line three-dimensional modeling method andsystem based on the oblique image provided by the invention, the establishment of the power transmission line three-dimensional model is realized through a terrain model and a tower model classification processing and synthesis method, the reuse rate of the tower model is improved, and the flight complexity and the surveying and mapping and modeling workload are reduced; and meanwhile, the problems of large deformation, shadow and shielding of the multi-view image inclined image are solved, and the matching quality is improved.

The invention discloses a self-adaptive panoramic stitching method based on pictures captured by a dome camera. The method comprises the following steps: S10, controlling the dome camera to capture aplurality of images of which the front parts and the rear parts are partially overlapped in a 360-degree range at preset rotation time and rotation angular speed; s20, preprocessing the image and calibrating a feature point selection region; s30, feature points are extracted from the calibrated feature point selection area, and feature point matching is carried out; s40, performing mapping correction and transition reduction on the spliced image; and S50, panoramic image splicing is carried out. According to the self-adaptive panoramic stitching method based on the dome camera snapshot pictures, the images snapshot by the dome camera can be spliced in a self-adaptive manner; by calibrating the feature point selection area, the feature point detection area is reduced to the image overlapping area from the whole image, the matching quality of the feature points is optimized while the detection efficiency is improved, the splicing efficiency and the splicing quality are greatly improved,and 360-degree dead-corner-free monitoring of a scene is realized.

The invention relates to the field of optical measurement and target recognition, and particularly relates to a two-dimensional image matching effect evaluation method. The method is an evaluation index system containing six parameters, and the larger a simple signal to noise ratio is, the lower the possibility that detection of a main peak is missed is; the larger the cross-sectional area of the bottom of the main peak is, the easier the peak can be found; high matching accuracy can be obtained conveniently through appropriate main peak standard summit cross-sectional area and a larger main peak smoothing coefficient, the shorter the average computing time is, the quicker matching is, and the smaller the average value of matched absolute error vector models is, the higher the matching accuracy is. Due to the fact that six evaluation indexes are used flexibly, an optimal related measure function, an optimal search algorithm, an optimal sub-pixel interpolation method and the like can be determined by using small-range experimental data, in an auxiliary mode, low-efficiency matching operation is reduced to the maximum, and the matching quality is optimized.

The invention relates to the field of hot rolling pickling plates, in particular to a boracic low-carbon hot rolling pickling plate with the thickness lower than or equal to 1.6 mm and a production method thereof. By means of the boracic low-carbon hot rolling pickling plate with the thickness lower than or equal to 1.6 mm and the production method thereof, surface quality control and structure property matching are effectively achieved, and boron is added into a low-carbon hot rolling pickling plate to reduce the transformation temperature from austenite to ferrite, so that structural transformation does not happen at the finishing rolling temperature of 840-860 DEG C at a hot rolling stage, that is to say, no mixed crystal defects exist, finally, the even structural performance is obtained, and meanwhile, an advantage is created for inputting a large amount of descaling water to overcome scale defects at the hot rolling stage.

A re-matching method of slab-order for improving slab utilization in iron and steel plants, comprising the steps of Step 1: formulating a mathematical model according to information of slabs and orders, so as to quantitatively describe the optimization problem of slab-order matching; Step 2: dividing the slabs and orders into groups; Step 3: generating a new slab-order matching relationship for each group of slabs and orders by using a multiple neighborhoods based tabu search algorithm on the basis of the formulated mathematical model, tries to re-optimize the matching relationship with consideration of the original slab-order matching relationship and the load balance of all equipment units, so as to improve the slab-order matching relationship, decrease the slab cutting loss, reduce the redundant slab weight of each order, and improve the integrity of an order, thus improving slab utilization, ensuring to deliver an urgent order on time, and improving the customer satisfaction.

The invention discloses a method for producing a digital elevation model by using three-line array three-dimensional satellite images, which comprises the following steps of: constructing an adjustment block for multi-degree overlapped and covered remote sensing satellite three-line array three-dimensional images in a production area, finishing block adjustment, and realizing high-precision orientation of all the images; then, according to index values such as an overlapping ratio and a same-name light intersection angle between any two images, constructing a plurality of pairs of double-image stereoscopic images from the multi-degree overlapped and covered three-line-array stereoscopic images, and respectively extracting a digital elevation model (DEM) by adopting an image dense matching algorithm; and finally, calculating a weight value of each double-image stereoscopic image according to indexes such as an intersection angle and relative precision of the double-image stereoscopic images, time phase difference forming the images and the like, and carrying out weighted fusion on DEMs extracted from all the double-image stereoscopic images in the region based on the weight values to generate a final DEM product. According to the method, the information advantage of the three-linear-array stereo image is fully utilized, and the matching success rate, the resolution ratio and the product quality of the DEM can be effectively improved.

The invention discloses a reverse-flow full-automatic battery grouping method and a reverse-flow full-automatic battery grouping system based on dynamic optimization. The battery grouping system comprises a battery performance detection device, a conveying device, and a main control computer. In the conveying device, an input conveyor belt and one side of an annular conveyor belt are parallel to and connected with each other, the annular conveyor belt and an output conveyor belt are perpendicular to and connected with each other, and the output conveyor belt and an outlet conveyor belt are perpendicular to and connected with each other. A baffle, a push-in cylinder and a photoelectric door are arranged at the tail end of the input conveyor belt. A push-out cylinder and a photoelectric door are arranged at the joint of the annular conveyor belt and the output conveyor belt, and a sensor is arranged beside the push-out cylinder. The distance between the push-out cylinder and the push-in cylinder is equal to the spacing between two adjacent batteries on the annular conveyor belt. The battery grouping method comprises the steps of battery input and grouping optimization calculation, battery pushing-out, and battery grouping output. By adopting the battery grouping method and the battery grouping system, real-time grouping optimization selection can be carried out according to the current state, and the quality and success rate of grouping are improved.

The invention discloses a single-phase direct-current motor driving chip and a chip drawing method. The chip comprises a digital layout area and an analog layout area, wherein the digital layout areais separated from the analog layout area, a digital module is arranged in the digital layout area, an analog module is arranged in the analog layout area, the digital module and the analog module areconnected with each other through a signal line, and the signal line is provided with ground wire shielding. The chip drawing method comprises the following steps: step 1, determining the area of thechip; 2, determining the shape of a protection ring; 3, planning the position of the modules; 4, carrying out layout and drawing; step 5, carrying out DRC / LVS verification; step 6, carrying out post-imitation; and step 7, carrying out delivery and manufacturing.

The invention discloses a video incidental music recommendation method based on a neural network technology, and the method comprises the steps: inputting a video set containing a plurality of to-be-incidental music videos and a music library containing a plurality of to-be-selected incidental music videos into a pre-trained style label prediction model for analysis, obtaining video style label information corresponding to the video set and audio style label information corresponding to each segment of to-be-selected music; according to the video style label information corresponding to the video set and the audio style label information corresponding to each segment of the to-be-selected incidental music, calculating a matching value between the video set and each segment of the to-be-selected incidental music; and according to the video set and the matching value of each segment of the to-be-selected score, screening out a target score from the music library to serve as the score of the video set. Therefore, the method can enable the pre-trained style label prediction model to have a better fusion capability for the characteristics of two different modes, namely audio and video, so that the matching quality of the finally recommended dubby music and video can be improved. The invention also relates to the technical field of blockchains.

The invention discloses a distributed data exchange network system and method based on an edge data gateway, relates to the technical field of data exchange, and solves the technical problem that a data transmission mode and uplink data cannot be accurately analyzed in a data interaction process of an enterprise in the prior art. When the demand block chain carries out original account data uplink, the original account data uplink is monitored, and the situation that the memory of the demand block chain is occupied due to invalidation or repetition of the original account data, and consequently the efficiency of business rationality analysis is affected is prevented; the premise of fixed length setting is that original account data updating exists in the demand block chain in the original account data backup process, so that the situation that no original account data updating exists in the demand block chain after the demand block chain backs up all stored original account data to a supervision chain is prevented; namely, in the data exchange process, the data of a supervision chain supervises no updated original account data, so that the supervision efficiency is reduced, and unnecessary waste of cost is caused.

The invention provides a feature matching method based on an attention mechanism and neighborhood consistency. The method comprises the following steps: extracting a picture key point and a descriptor thereof by using a convolutional neural network; key point positions are coded, and position information is fused into descriptors; using a self-attention network and a cross-attention network to enhance the characterization capability of the descriptors and searching a corresponding relation between the descriptors of the two pictures; constructing a graph by using a k-nearest neighbor algorithm according to the attention weight, evaluating neighborhood consistency of nodes by using a graph neural network, and fusing consistency information into attention aggregation; and according to the enhanced descriptor, calculating a matching result by using a Sinkhorn algorithm and mutual neighbor screening. The invention provides a method for integrating neighborhood consistency into an attention mechanism, the neighborhood consistency in the attention aggregation process is dynamically evaluated while global context information is aggregated by the attention mechanism, and local information and global information are combined to jointly enhance the characterization capability of a descriptor, so that higher-quality matching is obtained.

The invention discloses a method for group matching of power lithium-ion rechargeable batteries. The method comprises the following steps: one-step charge and discharge detection is conducted to the power lithium-ion rechargeable batteries and the batteries with capacities larger than lower limit capacity are taken as acceptable batteries; the acceptable batteries are charged to a state with 50-90 percent of rated capacity and the rechargeable batteries with acceptable voltage drop are sorted; the obtained acceptable batteries are fully charged and are connected in parallel for at least 4 hours to enable the voltage difference between any two rechargeable batteries to be less than or equal to 1mV; the obtained rechargeable batteries are connected in series to discharge, the voltage of each rechargeable battery during discharge is acquired and recorded, and the discharge is stopped when the voltage of any battery reaches cutoff voltage; and a voltage difference standard is set according to the number of series batteries required for group matching and the battery voltage at the moment of discharge cutoff, and the batteries with voltage differences within the preset range are matched into a group. Since the parameters of the batteries are measured and sorted under the simulated battery use conditions and the single rechargeable batteries with close characteristics are sorted, the quality of the group matching is improved.

The invention discloses a multi-hearing-aid matching device which comprises a matching model and a base machine box. The multi-hearing-aid matching device is characterized in that a plurality of sets of artificial ear canals are symmetrically arranged on the two sides of the matching model, a hearing-aid is arranged in each artificial ear canal, the distance between two artificial ear canals in each set simulates the distance between the two ears of a human body, at least one earphone jack capable of being connected with the hearing-aids in the article ear canals is formed in the base machine box, and detecting earphones are inserted into the earphone jacks. According to the multi-hearing-aid matching device, the artificial ear canals are arranged on the matching model so that the hearing-aids can be put in the artificial ear canals at the same time, and a matching person can carry out fast switching and compare effects conveniently in the matching process. The human ears are simulated to hear sounds from the two sides of the head, and the matching effect is more real and accurate. The distance between every two artificial ear canals is arranged to simulate the distance between the human ears, and the matching quality of the hearing-aids is further improved. The earphone jacks are formed in the base machine box, and multiple persons can try to hear synchronously and conveniently.

A method for managing a cloud music account of a user is provided, including the following method operations: receiving metadata obtained from an audio file stored on a client device; receiving an audio sample obtained from audio content of the audio file; receiving image data processed from an image associated with the audio file; analyzing the metadata, audio sample, and image data to determine an identity of the audio content of the audio file; and associating to the cloud music account of the user a permission to access cloud-based audio content corresponding to the identity of the audio content of the audio file.