146results about How to "Reduce correction" patented technology

The invention provides a liquid-crystal slit grating which comprises a first polaroid, a first substrate, a drive electrode structure, a first direction distribution layer, a liquid crystal layer, a second direction distribution layer, a public electrode structure, a second substrate and a second polaroid in sequential stacked arrangement. The drive electrode structure comprises a drive electrode layer and a lead layer, the drive electrode layer comprises multiple bar-shaped drive electrodes which are arranged in parallel. The multiple bar-shaped drive electrodes are grouped into a plurality of drive electrode groups. Each drive electrode group includes N drive electrodes. The lead layer comprises N lead electrodes. The N drive electrodes in each drive electrode group are electrically connected with the N lead electrodes in one to one correspondence. The invention further provides a stereoscopic display device with the liquid-crystal slit grating and a calibration method of the stereoscopic display device. The liquid-crystal slit grating and the stereoscopic display device have the advantages of automatic adjustment and simplicity in the calibration method.

The invention discloses a gray fringe projection light intensity nonlinear correction method and a phase correction method based on the method. Gray scale lookup tables are established for measurement datum planes and surfaces of objects to be measured respectively in fringe projection measurement, and light intensity correction is performed on actual measurement fringes of the measurement datum planes and the surfaces of objects to be measured, so that the light nonlinearity of the measurement fringes is reduced, and the phase measurement accuracy is improved. Not only is inherent light intensity gamma distortion of a projector and a camera considered, but also differences of light intensity reflection characteristics of the surfaces of the measurement datum planes and different areas of the objects to be measured in the setup procedure of the lookup tables are considered, so that the light intensity correction method is more effective than existing methods, corrected fringes are closer to ideal values, and the measurement accuracy can be improved substantially through the method.

The invention provides an image processing method and an associated device for correcting a fisheye image and reducing perspective distortion. The fisheye image processing method is used for correcting the fisheye image correction and reducing the perspective distortion and can save the space of a temporary storage and reduce the perspective distortion. The method comprises the following steps: sequentially calculating a fisheye original coordinate value corresponding to each pixel of a plurality of pixels in the fisheye image according to a coordinate conversion function between a fisheye corrected image and a fisheye original image; and calculating the pixel value of each pixel of the fisheye corrected image by interpolation in the fisheye original image according to the fisheye original coordinate value, wherein the coordinate conversion function contains a fisheye coordinate correcting conversion function, a perspective distortion reducing coordinate conversion function and an image cutting zoom-in and zoom-out coordinate conversion function.

A high-precision imaging attitude pointing control method based on the agile satellite is provided. The method comprises that: an attitude maneuvering process is planned according to the specified ground imaging target and time; in the maneuvering process, the attitude prediction information and the corresponding time are calculated and transmitted to a star sensor according to a fixed frequency;the star sensor predicts the sky window according to the time and attitude data, and keeps trying to extract the attitude in a window tracking mode; the satellite moves to the vicinity of the target and gradually decelerates, the star sensor captures the satellite attitude information quickly in a high-precision manner, and the system uses double star-sensitive optical axis vectors to compete attitude reset; and finally, closed-loop correction is performed by using the difference between the reset attitude and the target attitude to complete the high-precision attitude pointing control. According to the method provided by the present invention, the star-sensitive high-precision attitude information available only at small angular velocities is fully used to compensate for the influence ofgyro measurement information with nonlinear and drift variation characteristics on imaging pointing precision.

The invention discloses a three-dimensional multi-curvature part bending method. According to the method, the structural form of a section bending die and a suitable section bending machine are determined by analyzing the shape, size, bending angle and forming force of a part to be bent; a digital model of the part to be bent is subjected to curvature analysis through a computer; bending springback theoretical analysis is adopted, and a die profile springback compensation design is performed by a numerical simulation method through computer aided design. The bending of a section requires matched bending chuck, bending die and machine tool. The die is processed by a numerical control technology. A measuring machine is used for detecting and correcting the die to be processed, and an error between theory and actuality is reduced, so that the part manufacturing accuracy is improved. Dies for three-dimensional complicated parts can be formed on a two-dimensional bending machine, the bending pass is reduced, and repair or scraping of the bending dies is reduced, so that the part manufacturing cost is effectively reduced; moreover, the production efficiency is high.

The invention provides compositions and methods for achieving storage stable aceclidine. The compositions preferably comprise aceclidine, a cycloplegic agent, a surfactant, a tonicity adjustor and optionally a viscosity enhancer and an antioxidant. The invention further provides methods for treating refractive errors of the eye with a storage stable aceclidine composition.

The invention relates to a flexible positioning device for assembling and welding pipelines and belongs to the technical field of assembling and welding of pipe parts. A flange positioning assembly, a pipe joint positioning assembly, a pipe fitting bearing assembly and a telescope-feed positioning assembly are all arranged on a foundation plate, and adapting seats of the assemblies are connected with one another through positioning keys and T-shaped grooves. According to the flexible positioning device for assembling and welding the pipelines, by means of matching of the assemblies, accuracy of assembling and welding pipe fittings is improved, shape correction of the pipe fittings during assembling is reduced, manufacture qualities and reliabilities of the pipe fittings are improved, the assemblies can be adjusted, and requirements for assembling and welding the pipe fittings with different sizes and shapes can be well met by adjusting positions and heights of the assemblies when users are confronted with the pipe fittings with different sizes and shapes, so that the applicability of the flexible positioning device is greatly enhanced; the flexible positioning device can be continuously recycled and greatly reduce usage and manufacture costs; defects that medium frames in prior art are low in positioning accuracy, poor in rigidity, easy to deform during using, and long in manufacture period can be overcome.

The invention belongs to the field of underwater sound communication and navigation, and relates to a single transponder slope distance underwater sound positioning method and system based on adaptiveKalman filtering (AKF). The single transponder slope distance underwater sound positioning system based on the adaptive Kalman filtering (AKF) is composed of a transponder, a sound signal transceiver, a Doppler flow meter, an attitude sensor, a depth sensor and an adaptive Kalman filtering controller. Based on an adaptive Kalman filtering (AKF) algorithm, single slope distance information, depthmeasuring information, speed information and attitude information of an underwater vehicle are integrated, noise of the sensors, time delay, update frequency and multi-sensor information fusion are considered, a single slope distance positioning method of the single transponder of the underwater vehicle is established, three-dimensional coordinates of the underwater vehicle are calculated, and accurate positioning is realized. According to the single transponder slope distance underwater sound positioning method based on the AKF, arrangement and positioning work costs of the transponder can besaved, the problems that speed and distance are limited by a sound signal propagating underwater, the slope distance measuring updating frequency is slower, and updating with depth and attitude measuring information is not synchronous can be solved, and thus, the positioning efficiency and precision are improved.

A dielectric antenna having an at least partially dielectric body and an electrically conductive sheath, wherein the dielectric body can be struck on a supply section with electromagnetic radiation and the electromagnetic radiation can be at least partially emitted from the dielectric body via a lens-shaped radiation section. The conductive sheath essentially surrounds the dielectric body from the supply section to the radiation section and a supply opening is provided in the conductive sheath in the area of the supply section and a radiation opening is provided in the conductive sheath in the area of the radiation section. The dielectric antenna that is easy to produce, has little interference and has a short construction as a result of the dielectric antenna having essentially the cross section of the radiation opening in the area of the supply section.

The invention provides a multi-input large-area writing method, a multi-input large-area writing device and a multi-input large-area writing system. The method comprises the following steps that: a handwriting input detection device detects whether a light spot enters a plane monitoring area or not; when detecting that the light spot enters the plane monitoring area, the handwriting input detection device sends trigger information to a receiver; the receiver acquires the motion trace data sensed by a luminous device generating the light spot, wherein the motion trace data is sensed by a motion sensing unit in the luminous device; and a control host computer acquires or erases the handwriting information according to the motion trace data acquired by the receiver. The method, the device and the system can remarkably reduce the requirements on the precision and speed of an independently used CMOS sensor, lower the cost, improve the accuracy and speed of inputting and erasing the handwriting information and greatly improve the cost and performance while ensuring that multiple electronic light pens and electronic erasers are used at the same time.

Provided is a means of precisely reducing irregular concentrations in an array-like printhead for a long picture. An information of concentrations from each image N to be corrected (step S2) is obtained in the first place before a concentration information data DNi that corresponds to each recording pixel i is specified(step S4), and a deviation -DNi=DNi-DNave, iscalculated treating an average of the concentration data DNi as DNave (step S5). Then, a difference of the quantity of light, -ENi, corresponding to the deviation -ENi is obtained using a straight line of conversion (step S6), followed by (step S7) that calculates a correction quantity CNi=10(--ENi) that meets each recording pixel and (step S9) that composes the quantity of correction CNi calculated from each correction image N.

The invention belongs to the field of pipeline welding, provides a pipeline hot melting butt joint technology and hot melting butt joint equipment and solves the problem that the correction is large when a pipeline is welded. According to the scheme, the pipeline hot melting butt joint technology comprises the following steps: 1, preparing: levelling and cleaning to-be-butt joint end faces of a pipeline, arranging two heating plates at an interval, and enabling the two heating plates to be propped against the adjacent pipeline; 2, heating: heating the heating plates to melt the part, propped against the heating plates, of the pipeline; 3, crimping: pushing the pipeline towards the heating plates to deform the part, propped against the heating plates, of the pipeline into curled edges; 4, switching: rotating the heating plates in the circumferential direction, moving the heating plates in the axial direction simultaneously to enable the heating plates to deviate from the pipeline propped against the heating plates, and removing the heating plates when the pipeline and the heating plates are separated from each other; and 5, carrying out pressure welding: pushing the melted part of the pipeline to be propped against the heating plates. As the heating plates rotate in the circumferential direction and can form shearing force during rotating, the pipeline and the heating plates canbe easily separated, and the axial or radial moving degree of the melted part of the pipeline is lowered to reduce correction.

The invention relates to the field of a robot, and discloses an error compensation method of a robot off-line teaching device. According to the error compensation teaching method disclosed by the invention, by testing and analyzing the influence of system error of various teaching articulated arms on motion parameters and correcting the parameters of the various teaching articulated arms, accurate program control on a target robot is realized, and great flexibility is expressed; meanwhile, the teaching operation difficulty is reduced, the operation experience of a skilled worker can be deposited through a mechanical arm, and the skill level of the skilled worker can be achieved by the whole production line, so that tedious robot track correcting processes can be reduced, and the programming efficiency is greatly increased.

The disclosure provides an inspection device including a light source module, an image receiving module and a processing unit. The light source module emits a first incident light and a second incident light to a device under test (DUT). The image receiving module receives a first image corresponding to the DUT irradiated by the first incident light, and receives a second image corresponding to the DUT irradiated by the second incident light. The processing unit calculates the contrast ratio of the first image and the second image to obtain a high-contrast image for inspection.

A YOLOv3-based synchronous end-to-end license plate positioning and recognition method comprises the steps of obtaining a license plate external rectangular frame and license plate number information,making a data set label file, and dividing a data set into a training data set, a verification data set and a test data set; training a license plate detection and recognition integrated network LPYolo based on YOLOv3, so that the LPYolo recognizes various types of license plates while detecting the license plates, augments a training data set, and puts the augmented training data set into the network for training; and positioning and identifying the license plate on the test data set by using the trained network model, and then evaluating the model. According to the invention, license platesof various types and different lengths can be classified and identified. Meanwhile, the network LPYolo proposed by the method can quickly and accurately detect and identify the license plate in various unconstrained environments.

Pixel rows of an image sensor are respectively provided with a vertical transfer path. A charge detecting amplifier is disposed at one end of each of the vertical transfer paths. When a through image is outputted and when a moving-image mode is executed, a horizontal scanning circuit sequentially selects the charge detecting amplifiers on the basis of a thinning pattern, which is determined every imaging sensitivity, to output an image signal in which the pixel rows are thinned. As the thinning pattern, there are an odd-row pattern for selecting only the odd pixel rows from among the whole pixel rows, and an even-row pattern for selecting only the even pixel rows therefrom. The thinning pattern is selected in accordance with the imaging sensitivity so as to minimize line-shaped defects caused by the vertical transfer paths.

The invention discloses a remote sensing image change detection method under low registration precision based on a graph matching model. The method comprises the following steps: firstly, carrying outbuilding segmentation by using a multi-scale adaptive convolution remote sensing image semantic segmentation model to obtain segmentation results of different time phases in the same region, namely binary images; in order to solve the problem of region connection after segmentation of a plurality of different buildings, adopting plaque extraction based on an instance segmentation method, so thatdifferent plaques are distinguished, and regions corresponding to the plurality of buildings in different time phases are obtained; carrying out patch matching based on a graph matching model according to the position information corresponding to the patch and other characteristics; and finally, extracting corresponding change information according to the matched information. According to the invention, high-precision change detection of remote sensing images with different time phases under low registration precision is realized. The change detection scheme provided by the invention does notdepend on accurate image preprocessing, especially image correction and registration, and the requirements for image correction and registration are reduced.

The invention discloses a method for improving the product qualification rate based on a multi-dimensional decision tree group. The method is characterized by taking a production factor as a classification standard to construct different decision trees so as to form a multi-dimensional decision tree group; providing multi-dimensional optimization suggestions, including highest qualification rate configuration suggestions; determining the influence factors which have great influence on the product qualification rate can according to the influence degree of the influence factors; performing auxiliary decision making, so as to improve the product qualification rate, reduce correction, and improving work efficiency improvement; or directly giving the suggestions that the configuration is changed at least under the condition of higher qualification rate. The method supports real-time data updating and dynamic growth of the decision trees; basic working conditions can be finely adjusted through machine learning, so that the productivity change after fine adjustment can be learned; and at the same time, due to the fact that the number of equipment parameters is large, the decision tree group needs large space while the decision tree group and the machine learning module can be switched in different memory spaces, and needed data can be flexibly dispatched or learned in the corresponding memory space.