87results about How to "Improve fit accuracy" patented technology

A photovoltaic unit includes a matrix having a plurality of frameless rectangular solar modules, and at least one module rail disposed on an underside of each solar module. Each of the at least one module rail is coupled to a substrate rail, which is releasably connected to a substrate. Each of the at least one module rail and the at least one substrate rail has a guide rail running parallel to an edge of the respective solar module on at least a longitudinal side of the respective rail facing towards an edge region of the solar module. The guide rails on the module and substrate rails are releasably connected to each other by at least one connecting element. A spacer gap of sufficient width for operating the connecting element is provided between adjacent solar modules of the plurality of frameless rectangular solar modules.

A portable communication device is disclosed comprising a front face (3) having a plurality of openings therein, and an array (6) of keys. Each key has a switch actuator, a switching element, and coupling means. The switch actuator has an upper surface bounded by side walls, and the switch actuator of each key of the array (6) protrudes from one of the plurality of openings to provide its upper surface for tactile actuation by a user and is physically discrete from the others. The coupling means couples the tactile actuation of the upper surface by a user to the switching element.

The invention discloses a nonlinear SAR image geometric correction method based on thin-plate spline interpolation, and mainly aims to solve the problem that the nonlinear distortion of an image cannot be corrected precisely in the prior art. The method is implemented according to the following steps: (1) using an SIFT-Like algorithm to extract the feature points of a reference image and a real-time image; (2) using a similarity measure method to select matching points from the feature points of the reference image and the real-time image; (3) using an RANSAC algorithm to purify the matching points, calculating an affine transformation matrix, and getting an accurate affine transformation model, namely, the linear mapping relation of the images; and (4) using a thin-plate spline interpolation function to locally and nonlinearly fit the affine transformation matrix and the actual error on the affine transformation model to get the nonlinear mapping relation of the images, and thus, realizing precise geometric correction of the real-time image. The influence of nonlinear geometric distortion can be eliminated, and the correction accuracy is improved. The method can be used in precise geometric correction of nonlinear distortion of image signals received by radar.

A stereo display panel and a display panel. The stereo display panel comprises a display panel and a 3-D retarder film. The display panel comprises a plurality of pixel units. At least one of the pixel units is a para pixel unit. The display panel comprises a first substrate, a second substrate, display medium and at least a para marker. The second substrate is oppositely arranged to the first substrate, the display medium is positioned between the first substrate and the second substrate, and the para marker is positioned in the para pixel unit. The 3-D retarder film is arranged on the second substrate of the display panel, and the image sampler of the 3-D retarder film is aligned to the para pixel unit.

The invention discloses a display method and system for an early warning function of an intelligent vehicle. The system comprises an obtaining module, a processing module, a first conversion module, a second conversion module and a display module. The method comprises the steps that based on a first coordinate system conversion module, vehicle body space coordinate expressions of the lane line and the vehicle are converted to camera coordinate expressions; based on a second coordinate system conversion module, the camera coordinate expressions of the lane line and the vehicle are converted to image coordinate expressions; and based on the image coordinate expressions of the lane line and the vehicle, the lane line and the vehicle under an image coordinate system are displayed. Through the display method and system, the lane line and the vehicle with the three-dimensional perspective effects can be quickly drawn on a liquid crystal instrument panel, the increasing displaying requirement of an advanced driver assistance system for early warning information is met, and the display method and system for the early warning function of the intelligent vehicle can be widely applied to the field of displaying of the vehicle-mounted liquid crystal instrument panels.

The aim of the invention is to provide grinding rolls in particular for high-pressure roller mills or rolling presses for pressure grinding a granular material not only with a wear-resistant coating but also with a face armour plating in the area of the roller annular edge, which has a long service life due to the high wear resistance thereof and whose production and repair are, in general, relatively simple and inexpensive. For this purpose, the inventive face armour plating consists of a plurality of prefabricated hard bodies which are circularly arranged side by side in such a way that the circular face edge of the roller is formed, said hard bodies are disposed on said circular shoulder of a roll shell such that said hard bodies are placed axially and radially against said circular shoulder of the roll shell and are movably assembled with the roll shell.

A joining method and to a joining device (1) for joining roof skins (11) to parts (4) of vehicle bodies (3) is provided. The joining device (1) includes one or more joining grippers (2), which are comprised of at least one frame (24), of a number of gripping elements (29, 30), of at least one pressing strip (27), which acts upon the roof skin (11), and of an associated adjusting device (32). The pressing strip (27) is adapted to the joining contour (6) of the part (4) and has one or more dimensionally stable strip segments (33, 34, 35, 36, 37) that, independent of one another, are mounted on a strip support (26) with a limited flexibility in the direction of adjustment (41). The roof skin (11) with the joining part edge (13) thereof, is pressed against and fixed to the contact area (7) of the part (4) by means of the strip segments (33, 34, 35, 36, 37) and, optionally, is permanently deformed.

The invention discloses a multi-source remote sensing image radiometric normalization method. The method is characterized in that relative radiometric normalization of a multi-source remote sensing image is divided into a sensor radiation correction process and a radiometric normalization process specific to external factors such as illumination. The method comprises the following steps: S1, acquiring a sensor radiation correction coefficient in a categorical regression way based on a clear sky image; S2, implementing semi-automatic classification and sensor radiation deviation correction of the multi-source image by a sample transmitting and reclassifying method; and S3, implementing relative radiometric normalization of the image through a PIFs (Pseudo Invariant Features) automatic selection method based on an NDVI (Normalized Difference Vegetation Index) histogram of differences and a category constraint. Through adoption of the multi-source remote sensing image radiometric normalization method, radiation deviations among sensors are effectively corrected, and higher radiometric normalization accuracy is achieved as a whole than a conventional method. Meanwhile, radiation feature fluctuation among time-sequence images can be eliminated effectively through the method, so that aspect change information of land such as vegetative can be expressed more accurately, and a reference method is provided for cooperative utilization of the multi-source time-sequence images.

A relay, in particular for a starter for an internal combustion engine for use in a motor vehicle. The relay has a self-resilient contact bridge, which for a reversible change in shape effected by the contact pressure, in particular widening, is embodied in curved form.

A semiconductor laser device includes a semiconductor laser element, a lead having an element disposing section where elements are disposed and a resin bonded to the lead. The lead has a thick portion and a thin portion with the thick portion being formed to extend at least in the direction of the width of the resin to cover a region having a length equal to or greater then the width of the resin. A semiconductor laser device has a semiconductor laser element, a lead having an element mount portion on which the semiconductor laser element is mounted and a resin maintained in intimate contact with the lead. The lead has a thicker portion and a thinner portion. The thicker portion is formed to extend in the direction of the width of the resin over a width equal to or greater than the width of the resin.

The invention discloses an LOCA jointing machine for CCD alignment. The LOCA jointing machine comprises a slit coating mechanism, a glue supply mechanism, a pre-curing mechanism, a turnover jointing mechanism, a carrying mechanism, a CCD alignment mechanism and a curing mechanism. The slit coating mechanism is used for linearly coating the glued face of a first substrate with LOCA glue. The glue supply mechanism is used for supplying LOCA glue to the slit coating mechanism. The pre-curing mechanism is used for curing the LOCA glue sprayed on the glued face of the first substrate on the first substrate. The turnover jointing mechanism is used for turning over the first substrate with the pre-cured glued face and jointing the first substrate with the glued face downwards after turnover and a second substrate. The carrying mechanism is used for carrying the first coated substrate to the position below the pre-curing mechanism and carrying the first substrate with the pre-cured glued face to the turnover jointing mechanism. The CCD alignment mechanism is used for adjusting the first substrate with the pre-cured glued face to be aligned to the position to be jointed. The curing mechanism is used for curing the first substrate and the second substrate which are jointed. The gluing and jointing process are automatic, no manual operation is needed, and jointing accuracy and jointing efficiency are high.

The embodiment of the invention discloses a character detection method and device, electronic equipment and a storage medium. The method comprises the steps of extracting multi-level features of a to-be-detected image through a convolutional neural network; carrying out regression on the multi-level features, and determining initial text box features in the multi-level features are determined; generating pyramid features according to the multi-level features; constructing a new text box feature according to the initial text box feature and the pyramid feature; performing quadratic regression on the new textbox feature according to the initial textbox feature to obtain a textbox feature; and combining the textbox features to obtain a text detection result in the to-be-detected image. According to the embodiment of the invention, on the basis of performing regression on the multi-level features to obtain the initial textbox features, secondary regression is performed on the new textbox features, so that the fitting accuracy of the textbox boundary is improved, and the accuracy of a text detection result is improved.

An orthodontic apparatus (1; 10) is designed with at least one orthodontically effective element (4; 12) and at least one holding element (5; 13) for holding the apparatus (1; 10) on the jaw, wherein the apparatus (1;10) is designed in one piece. In a method for manufacturing an orthodontic apparatus, an electronic data set is produced for a three-dimensional model (2) of a jaw to which the apparatus is to be attached. Further, a virtual positioning of one or more orthodontic elements (4; 12) of the apparatus on jaw model (2) takes place. An electronic data set of a three-dimensional model (3; 11) of the apparatus is subsequently produced. The apparatus (1; 10) is then produced by an additive and / or removal production method according to the data set of the apparatus model.

The invention relates to a modular shower cabin for aircrafts comprising an integratable faucet unit (6), and integrated technical facilities, like for example a water drain (15), illumination, a ventilation device, oxygen masks or the like, wherein the shower cabin comprises a center portion (2), in which an entrance area formed by a recess (5) and a mounting for the faucet unit (6) are provided, and wherein the shower cabin comprises a ceiling portion (4) adjoining the center portion (2) at the top, and a bottom portion (3) adjoining the center portion (2) at the bottom, and wherein the technical facilities are allocated to the ceiling portion (4) and / or to the bottom portion (3).

The invention discloses a scene image de-noising method based on a Type-2 fuzzy logic system and relates to the technical field of an ordinary image data processing or producing method. Distribution densities of input and output samples are respectively estimated by adopting a kernel density method and inputted into the Type-2 fuzzy logic system, and fuzzy weights of corresponding samples are obtained according to system reasoning to establish a region Type-2 fuzzy density weight support vector regression model which is applied to the scene image de-noising method to obtain better de-noising results. The fuzzy weights are designed by a novel fuzzy subordination degree design method based on the sample distribution density and have the obvious characteristics to support the regression precision of a vector machine.

The invention relates to a structure residual bearing capacity measuring method and system based on tunnel lining deformation features. The method comprises the following steps: 1) according to an actual tunnel structure, manufacturing a lining model according to a set similarity ratio; 2) carrying out loading test on the lining model and recording structural deformation and corresponding load ofthe lining model; 3) obtaining the structural deformation and the corresponding load of the corresponding actual tunnel structure according to the similarity ratio, and obtaining a deformation load function relation of the actual tunnel structure by adopting a data fitting mode; and 4) measuring to obtain current convergence deformation of the actual tunnel structure, and measuring the residual bearing capacity of the actual tunnel structure on the basis of the deformation load function relation. Compared with the prior art, the residual bearing capacity of the structure can be quantitativelyjudged according to engineering monitored data and is used as a structural safety evaluation basis; and meanwhile, the stress damage characteristic and the damage mode of the structure can be furtherrevealed through model test.

The invention discloses a tooth profile deviation measurement method based on DP-PSO-SVR, and belongs to the technical field of precision testing. The invention helps resolve problems of high measurement cost, low precision and short service life. The method comprises the following steps: measuring tooth profile data of a to-be-measured gear as to-be-measured data, measuring tooth profile data ofa test gear as sample data, and carrying out coordinate transformation processing on theoretical tooth profile data of the to-be-measured gear; establishing a tooth profile deviation mathematical model; establishing an SVR model; optimizing and training parameters in the SVR model by using a DP-PSO algorithm and the sample data to obtain a DP-PSO-SVR model; fitting the to-be-measured data by usingthe DP-PSO-SVR model to obtain an actual involute of the to-be-measured gear, fitting theoretical tooth profile data by using a least square method to obtain a theoretical involute, and calculating to obtain a tooth profile deviation between the actual involute and the theoretical involute through the tooth profile deviation mathematical model. According to the invention, the measurement cost canbe reduced, the measurement precision can be improved, and the anti-interference performance is good.

The invention relates to a method and a device for machining workpieces comprising an annular-shaped section that is to be provided with a split bore. Said device comprises: a fracture-splitting device for fracture-splitting the annular-shaped workpiece section and for creating fracture-split workpiece parts corresponding with one another with a split bore; an applicator device for a plain-bearing material for applying at least one closed annular plain-bearing layer to the inner periphery of the split bore of the fracture-split workpiece; and fracture-splitting device for the plain-bearing layer for fracture-splitting the closed annular plain-bearing layer of the previously fracture-split workpiece.

Based on differential AM-FM signal instantaneous frequency calculation method, the steps are: (1) input AM-FM signal x(t)=a(t)cos[θ(t)], and carry out denoising and discrete sampling processing; ( 2) Obtain the local maximum, minimum and occurrence time of the signal x(t)=a(t)cos[θ(t)], denoted as (t j , a(t j )); (3) According to the obtained a(t j ) value, using spline interpolation function to fit a(t j ) to obtain the envelope function a(t) of the input signal; (4) obtain the first order derivative x'(t) of the input signal x(t)=a(t)cos[θ(t)]; (5) Calculate the local maximum and minimum values ​​of x′(t) and the occurrence time, denoted as (t i , x′(t i )); (6) Calculate t according to the formula i The instantaneous frequency of the input signal f(t i ); (7) carry out phase-shift processing to x(t)=a(t)cos[θ(t)], obtain x respectively 1 (t),x 2 (t),...x n-1 (t), where n≥1; (8) for the obtained x 1 (t),x 2 (t),...x n-1 (t), repeat steps (2) ~ (6), get (t i , f(t i )) point sequence; (9) according to the obtained (t i , f(t i )) point sequence, using spline interpolation function to fit f(t i ), get the instantaneous frequency function f(t) of x(t)=a(t)cos[θ(t)].

A gas bag module (10) for a vehicle occupant restraint system comprises a housing with a gas generator support (17) and a covering (26). The housing consists at least partly of a thermoplastic elastomer. The thermoplastic elastomer is foamed using a physical or chemical foaming agent.

The invention discloses a low-code-rate audio quality objective evaluation method based on nonlinear parameter fitting, which belongs to the field of communication. The method is characterized by including steps: firstly, selecting some audio frequencies with typical characteristics, compressing the audio frequencies at different code rates, and subjectively evaluating and marking the audio frequencies decompressed after being compressed; secondly, extracting code rate and scale factor of each audio frequency at each code rate, subjectively evaluating and marking independently at each code rate and performing nonlinear parameter fitting of the scale factors to obtain respective fitting formula at each code rate; and thirdly, extracting the code rate and the scale factor from audio stream needed to be objectively evaluated, selecting the corresponding formula according to the code rate and substituting the scale factor in the formula to obtain the objective evaluation mark of the audio frequency by calculating. The independent parameter fitting at each code rate can eliminate influences of different code rates and more accurately measure quality damage caused by quantization. The conditions that the mark is higher when the scale factor is smaller or larger and the mark is lower when the scale factor is medium can be simulated by means of nonlinear parameter fitting of the subjective evaluation mark and the scale factor, fitting degree is improved, and subjective evaluation can be better simulated by objective evaluation.

There is provided a wind power installation pylon comprising a plurality of pylon segments which respectively have an upper and a lower horizontal flange. At least one of the plurality of the pylon segments has at least two longitudinal flanges. Each longitudinal flange has a first side for bearing against a first side of a further longitudinal flange and a second side to which the peripheral surface is welded, wherein the second side is opposite to the first side.

The invention discloses a human body model driving method, and the method comprises the steps: obtaining the posture and body shape parameters of a target human body model, and enabling the three-dimensional human body posture and body shape parameters to correspond to the skeleton and a plurality of basic parameters of a three-dimensional standard human body model; establishing a standard human body model in an initial posture; inputting the obtained multiple groups of basis and skeleton parameters into a standard three-dimensional human body model for fitting; obtaining a three-dimensional human body model with the same body shape as the target human body; driving the skeleton to move from the initial posture to the target posture; and obtaining a three-dimensional target human body model grid with the same posture as the target human body two-dimensional image. According to the method, fitting of the standard human body model and the target human body model is optimized, an animation series from the initial posture to the target posture is formed, and the process of driving the skeleton to move is optimized by using the variable-speed frame interpolation method, so that the whole fitting process achieves a good balance in time and effect.

The embodiment of the invention provides an image processing method, a movement control method, and a movement control system. The movement control system comprises a camera device. The method comprises the following steps: obtaining a to-be-processed image and area parameters, wherein the to-be-processed image is obtained by shooting a shooting area by a camera device, and the area parameters are used for representing to-be-identified area parameters in the shooting area; based on the area parameters, extracting a first area from the to-be-processed image; performing feature extraction on the first area to obtain a plurality of target points; obtaining a moving route based on the plurality of target points.

The invention discloses a greenhouse temperature and humidity control system and a method thereof. The system comprises sensors and controlled equipment, wherein the sensors and the controlled equipment are all connected with a microcontroller; the microcontroller is connected with a server; monitoring terminal equipment and the server perform information interaction through a wired or wireless mode; the server is frequently arranged in the server chassis; and the server chassis comprises a hollow cuboid-shaped shell and a cover plate pivoted on the hollow cuboid-shaped shell. In combination with the method thereof, the defects in the prior art that replacement is not easy; the overdue state of a dehumidizer can not be determined efficiently; the replacement speed is quick; the dehumidizeris much wasted and is replaced when not overdue completely; a fixed connection-type assembly structure increases the difficulty of maintenance and debugging, maintenance or debugging needs to be executed in the case of use in a not large area in the box body, and the difficulty of maintenance and debugging is thus added; subsequent exponential fitting accuracy becomes low; and the errors are large after fitting happens when time delay changes are not small can be effectively avoided.

The present invention discloses a nonlinear fitting method for a cooling process of a polymeric phase change material. The method comprises the following steps of: (1) taking the polymeric phase change material as a to-be-tested sample, heating the to-be-tested sample to an initial temperature T1, holding the temperature until the to-be-tested sample is completely melted, then rapidly transferring the to-be-tested sample into a cooling medium of which the temperature is T2, inserting a T-shaped thermocouple into the melted to-be-tested sample at the same time, and acquiring temperature variation values of the to-be-tested sample in the whole cooling process; (2) according to temperature and time data acquired in the step (1), drawing a cooling curve; (3) performing dimensionless processing on the cooling curve, and setting a dimensionless temperature theta to be equal to the value of (T-T2) / (T1-T2); and (4) according to the theta value obtained in the step (3), performing nonlinear fitting by means of the following algorithm: y=(A+B) / (C+ex-D)-B, and making x equal to ln t and y equal to theta, so that a nonlinear fitting parameter in the cooling process of the polymeric phase change material can be obtained. A four-parameter model formula fitting method proposed by the present invention is convenient, has very good fitting effects for different polymeric phase change materials, and is higher in fitting accuracy.

The invention relates to an arrival rate curve fitting method and device, and the method comprises the steps: obtaining a plurality of groups of monitored Internet gross evaluation points and information arrival rates with a corresponding relationship from a database as original data; screening the original data according to each group of information exposure parameters corresponding to the internet gross evaluation points and the information arrival rates with the corresponding relationship to obtain target original data; and fitting the target original data through the spline interpolation fitting model to obtain a target arrival rate curve. According to the invention, the technical problem of low fitting accuracy of the arrival rate curve in the prior art is solved.