139results about How to "Improve positioning accuracy" patented technology

The present invention relates to a device for inserting a self-expanding stent into a body vessel. The device comprises a tube, a pushing element and a grip having a housing via which the pushing element is secured on the grip. Further, the device comprises a stent carrier and a moveable element, the latter of which is guided in the housing of the grip and is coupled to the pushing element. Moving the movable element in the proximal direction effects a movement of the pushing element in the distal direction and a movement of the tube in the proximal direction.

A recess 14 is provided in a lens accommodating portion 13 of a laser holder 10 for holding a collimator lens C, whereby positional accuracy and adhesion strength of the lens and a light source are improved.

First elasticity information regarding elasticity of a subject in a first image and second elasticity information regarding the elasticity of the subject in a second image are acquired, and the first image and the second image are positioned with respect to each other on the basis of at least one of the first elasticity information and the second elasticity information.

The invention provides an IP address geographic position determination method. The IP geographic position determination method comprises the following steps of acquiring an IP address to be analyzed, searching the IP address to be analyzed through a disclosed IP base to obtain the published geographic position of the IP address to be analyzed, acquiring the last address geographic position corresponding to the IP address to be analyzed through a network detecting tool, and confirming the geographic position of the IP address to be analyzed according to the published geographic position of the IP address to be analyzed and the last address geographic position. The geographic position of the published IP address to be analyzed and the last address geographic position of the IP address to be analyzed are analyzed, the geographic position of the IP address to be analyzed can be determined according to an evaluation value of the candidate geographic position, so that the method has the advantage of improving locating accuracy. Furthermore, user experience on targeted delivery is improved. The invention further discloses an IP address geographic position determination system.

The invention discloses a geographic position sharing method and system and terminals, and belongs to the field of communication. The method includes the steps that the first terminals acquire current geographic position information of the first terminals, and the current geographic position information is sent to the second terminals; the current geographic position information of the first terminals is refreshed every preset time, and the refreshed geographic position information is sent to the second terminals, wherein the number of the second terminals is at least one. After the current geographic position information of the first terminals is refreshed every preset time, the refreshed geographic position information is sent to the second terminals, dynamic geographic positions can be shared to the second terminals in real time, the shared geographic position information is more indicative, and the accuracy of positioning on first users corresponding to the first terminals is improved through second users corresponding to the second terminals according to the refreshed geographic position information of the first terminals.

The embodiment of the invention provides a human head detection method and device, and the method comprises the steps: inputting a to-be-detected image into a head detection model, and obtaining a boundary frame set, the classification confidence degree of each boundary frame in the boundary frame set, and the IOU positioning confidence degree of each boundary frame; based on classification confidence coefficient of each bounding box in bounding box set and IOU positioning confidence coefficient of each bounding box, adopting IOU-NMS algorithm to carry out back-end processing on bounding box set to obtain a human head detection result, wherein the head detection model is obtained by performing training according to an image sample and a head target true value frame corresponding to the image sample. According to the embodiment of the invention, a human head detection model composed of a convolutional neural network and a specific recurrent neural network model is utilized to carry outfeature extraction and decoding regression on an input image, and then the IOU-NMS algorithm is used for carrying out back-end processing on the bounding box, so that the accuracy of human head detection in a complex scene can be effectively improved..

The invention is applicable to the terminal application field and provides a method and a system for tanking desktop objects. The method comprises the steps of obtaining the desktop thumbnail of each desktop page in a desktop page set and showing all the obtained desktop thumbnails together on a desktop main page; monitoring a clicking operation sequence for the desktop thumbnails shown on the desktop main page and determining the clicking position of each clicking operation in the clicking operation sequence; and ranking and displaying the desktop objects on the various desktop pages on the multi-screen desktop according to the clicking position of each clicking operation in the clicking operation sequence. According to the method and the system, the desktop thumbnails of various desktop pages in the desktop page set are shown on the desktop main page together and clicking operations are carried out on the desktop main page so that the desktop objects on the various desktop pages on the multi-screen desktop are ranked and displayed; and therefore, the ranking operation is simple, the elapsed time is short and the positioning accuracy is high.

The invention discloses a detecting method and a detecting device for a printed image, wherein the detecting method comprises the following steps of acquiring the printed image; performing coarse extracting on the edge contour of a foreground area according to an automatic threshold segmentation method; editing the edge contour according to a gray scale difference threshold, thereby accurately extracting the edge contour line of the foreground area; positioning the edge contour line according to an area positioning method and a linear positioning method; and detecting whether a defect exists in the edge contour line according to a Blob analysis method. The detecting method provided by an embodiment of the invention improves extraction accuracy and positioning accuracy of the edge contour, thereby realizing accurate detection for edge defects.

The invention relates to a measuring method and device. The method comprises the following steps: a network administrator or network element acquires first indication information, wherein the first indication information is used for instructing the selection of a mobile terminal to execute a performance measurement task or an MDT task according to carrier aggregation (CA) information; the network administrator or network element selects the mobile terminal according to the first instruction information to conduct performance measurement or MDT measurement on a target community. According to the embodiment of the invention, a base station can select the target community as an auxiliary community and/or select the mobile terminal of a main community to conduct targeted performance measurement or MDT measurement, so as to improve the positioning accuracy of network problems.

The invention discloses a user positioning method and device. The method comprises the following steps: acquiring cell data and measurement report data; calculating a neighbor cell and a cell coveragerange of the cell according to the cell data; extracting an information fingerprint according to the measurement report data, the neighbor cells of the cell and the cell coverage area; and calculating the position of the user terminal according to the neighbor cell, the cell coverage area and the information fingerprint of the cell. According to the method, the user positioning accuracy is greatly improved, the resource use efficiency is improved, and the method is a basis for all position-related applications; The method can be applied to the FDD/TDD network, and the application range is wider.

The invention provides a stent conveying device, which comprises a connecting part, an outer tube, an inner tube and a balloon. The outer tube is connected with the balloon and the connecting part, the inner tube penetrates through the inside of the outer tube and the inside of the balloon, the connecting part is communicated with the balloon via a joint of the connecting part and through an inner cavity between the outer tube and the inner tube, the outer tube comprises a far outer tube portion and a near outer tube portion, the far outer tube portion is located at a far end of the outer tube, and the near outer tube portion is located at the near end of the outer tube. The stent conveying device is characterized in that at least one of the far outer tube portion and the near outer tube portion comprises at least two layers, and materials and structures of the at least two layers can strengthen positioning accuracy of the stent conveying device. According to the invention, an outer tube structure with the special design is provided, and accordingly, the positioning accuracy of the stent conveying device can be improved.

The present invention discloses a WiFi positioning method. The method comprises a 100th step of acquiring signal strength data of a signal sent by a to-be-detected client that is received by each wireless access point in a detection region; a 200th step of inputting the signal strength data acquired in the 100th step into an input data layer of a positioning model; and a 300th step of calculating the signal strength data according to a network layer of the trained positioning model, outputting a position prediction set in an output layer of the positioning model, and determining a position corresponding to the prediction position tag with the maximum probability value in the position prediction set as a position at where the to-be-detected client is, wherein the position prediction set comprises a prediction position tag and a corresponding probability value. The positioning model in the present invention uses a trained depth neural network, and the depth neural network is trained by training massive sampling data, so that the positioning accuracy and precision are improved.

The invention discloses a method for automatically positioning urine. The position of the urine is determined according to the result of temperature scanned by a non-contact temperature sensor, and position point coordinates of the urine in a closestool are calculated. The invention further discloses a device for automatically positioning the urine. From the perspective of a virtual module framework, the device comprises an automatic urine positioning module adopted in the method; from the perspective of a solid structure, the device comprises one or more sets of positioning assemblies, and the positioning assembly comprises the non-contact temperature sensor connected with a stepping motor spindle. The invention further discloses the closestool comprising the device for automatically positioning the urine. The point coordinates of the urine are determined according to the temperature (the urine temperature or both the urine temperature and the environment temperature) scanned by the non-contact temperature sensor, the urine is sucked in the air to serve as a sample for health analysis, the problem of cross contamination caused by suction of the urine sample on the inner surface of the closestool is avoided, accuracy of health data analysis is improved, and the method, the device and the closestool are suitable for application and popularization.

The invention discloses a color time varying characteristic-based traffic light relocating method, the method is used for relocating the traffic lights in images, and the traffic lights comprise a plurality of indicator lights which are mutually matched to display traffic signals. The traffic light relocating method comprises the following steps: respectively obtaining the relocating area of each indicator light; and by taking a plurality of frames as a locating period, calculating the color component strength of each pixel in the relocating area of the image of each frame in a single locating period; when each locating period is ended, deleting the pixels which do not satisfy the characteristics of the indicator lights, finding the center points of the indicator lights according to the retained pixels after the deletion, and combining the center points with the sizes of the indicator lights to determine the positions of the indicator lights when the current locating period is ended. The invention furthermore discloses a corresponding traffic light relocating device.

The method includes steps: selecting source terminal and target terminal, background testing control module in source network takes task of collecting data needed by testing and positioning fault of speech sound; background corresponding to source terminal initiates starting up test of source terminal, collects running states of resources at source network used by source terminal; entering into analysis for tested result, ending positioning flow. Features are: all procedures are in real time; accuracy of positioning fault is in key node; background is only to receive testing result from foreground testing module without need of receiving huge original data.

A method includes; a step of setting square mask cells in rows and columns on a transparent mask-substrate surface by demarcating by orthogonal lines of equal intervals, each of which has one side having a length smaller than a resolution limit of an optical system; a step of setting the resist thicknesses corresponding to the mask cells; a step of assigning normalized light-intensities respectively to the mask cells as transmission intensities, corresponding to the film thicknesses and having three or more different values; a step of setting each of the mask cells a light-transmission area and a shade-area and determining a transmission-light intensity by an transmission area ratio; a step of providing shade films on the shade areas of the mask substrate.

The invention discloses a method and a server for updating geographic location information, and belongs to the field of positioning. The method comprises the following steps of: receiving positioning information uploaded by a terminal; acquiring an address sequence set corresponding to the positioning information, and transmitting the address sequence set to the terminal; receiving an address selected by a user according to the address sequence set from the terminal; and updating an area coverage of the address according to the address and all positioning information positioned to the address, so that the address can be accurately positioned according to the positioning information during subsequent positioning. The area coverage of the address in a map is continuously optimized and updated according to the positioning information of the user and a specific address selected by the user according to the positioning information, so that the problem of lower user positioning accuracy caused by problems of the positioning accuracy and errors is gradually solved.

The invention relates to a method and a device for detecting installation problems of antenna feeders in a base station. The method includes: extracting and calculating N RSSI (received signal strength indicator) averages from the base station; rejecting invalid data in the RSSI averages; calculating an absolute value of a slope difference of the RSSI averages of any two feeders in any sector of the base station at any equal time intervals; determining count values of any two feeders in any vector according to the absolute value of the slope difference and symbols of the RSSI averages of any two feeders in any vector of the base station at any equal time intervals; determining relative uniformity of the RSSI averages of any two feeders in any vector according to N count values of any two feeders in any vector; comparing the relative uniformity of the RSSI averages of any two feeders with a set threshold to determine whether each feeder in the base station has connection errors or not. By the method and the device for detecting the installation problems of the antenna feeders in the base station, accuracy and efficiency in judgment of the installation problems are evidently improved.

The invention provides a user depth portrait method and device, and relates to the technical field of big data user portraits. The user depth portrait method comprises the following steps: acquiring original data of a user, and determining the data type of the original data; performing abstract conversion processing on the original data according to the data type to obtain abstract data; determining a representation learning neural network corresponding to the type of the abstract data, using the representation learning neural network to model the abstract data, and obtaining a data representation model; and designing and training an objective function of the representation learning model according to the data representation model, and performing depth portrait on the user by using the objective function to obtain a user depth portrait. The user depth portrait method provided by the invention maps user data to a multi-dimensional space for modeling and user depth portrait, more user data are reserved, and meanwhile, privacy data are shielded, so that the user depth portrait method is more suitable for complex data mining and multi-service scenes.

A controller (1) converts physical movement in eight directions of motion into electrical signals. The controller (1) has two opposed handles (2a, 2b) connected together using a gimbal (150) having a telescoping device (200) that is rotatable relative to the gimbal (150). The telescoping device (200) has an inner member (18) that is biased in either telescoping direction using one spring. The inner member (18) can be locked along either of the telescoping directions using a locking device (26). The gimbal (150) has four sensors sensing movement of one handle (2a) relative to the other handle (2b).

Provided in the invention is a license plate positioning method. The method comprises: a shooting device carries out triggered shooting on a license plate by a trigger; an image shot by the shooting device is transmitted to a license plate positioning system by a network, and the license plate positioning system carries out pretreatment on the image processed by the last step to obtain a license plate area image; the license plate positioning system carries out license plate positioning processing on the license plate area image; and inclination correction processing is carried out on the processed license plate area. According to the invention, the accuracy of license plate positioning and the operation speed of the license plate positioning system are improved.

The invention discloses a measurement device for accelerator ray bundles. The measurement device for the accelerator ray bundles comprises an attenuation body device, a detector and a position adjusting device. The attenuation body device is provided with a working part, the working part resists ray bundles from an accelerator target spot and reduces strength of the ray bundles, the detector is used for detecting strength or dosage of the ray bundles which penetrate through the working part of the attenuation body device and the position adjusting device is used for carrying the attenuation body device and the detector and adjusting spatial positions of the attenuation body device and the detector. The thickness of the working part of the attenuation body device is variable, and therefore different attenuation to the strength of the ray bundles is conducted. The attenuation body device with the working part of which the thickness is variable and the position adjusting device for carrying and positioning the attenuation body device and the detector are arranged on the measurement device, and therefore adjustment of the attenuation body device and positioning accuracy of the device are improved.

The invention provides a pulmonary nodule image feature detection method based on a CT image, and the method mainly comprises the steps of carrying out the translation scanning of a pulmonary region,and automatically positioning all suspected pulmonary nodule image regions; based on the suspected areas, measuring the long and short diameters and the volume of the pulmonary nodules automatically with the coordinates of the center point as the sphere center and multiple scales as the diameter ranges; and then, filtering the CT value of the input image, automatically measuring the average CT value of the pulmonary nodules, and finally obtaining the average CT value. According to the invention, the functions of automatic positioning of pulmonary nodules, automatic measurement of long and short diameters and volumes, automatic measurement of CT average values and the like can be realized. Various image features of the pulmonary nodules automatically obtained through the detection method can serve as judgment bases, and doctors are assisted in better diagnosing the benign and malignant degree and the invasion degree of the pulmonary nodules.

The invention discloses a vehicle positioning method and device, electronic equipment and a computer storage medium, and relates to the technical field of navigation positioning. According to the specific implementation scheme, the method comprises the following steps: acquiring road network information, periodically positioning a vehicle to obtain a positioning position, determining a target roadsection in which the vehicle travels in the current period from a plurality of road sections indicated by the road network information according to the positioning position obtained by positioning ineach historical period, and correcting the positioning position in the current period according to the target road section to obtain the positioning position in the target road section. According tothe method, the current driving road section is predicted through historical positioning, and the positioning position is corrected according to the predicted current driving road section, so that thecondition of large positioning information fluctuation is avoided, and the positioning accuracy is improved.

The invention discloses a machine vision indoor positioning method based on an improved convolutional neural network structure and mainly provides an improved convolutional neural network structure and a neural network model training method for the structure. And finally, the input video images are classified through the trained convolutional neural network to obtain indoor position of a mobile robot equipped with a RGB camera. The convolutional neural network function comprises the steps of extracting position features of the semantic segmentation image and the RGB image, and determining thereal-time indoor position of the mobile robot by utilizing the two types of position features. The improved convolutional neural network structure is a product of combination of U-Net, front 13 layersof two VGG16Net and rear 3 layers of one VGG16Net, and the convolutional neural network is composed of five parts of U1, VGG2, VGG3, VGG4 and an ArcFace classifier. According to the invention, real-time positioning of the indoor position of the mobile robot can be accurately realized.

An exposure method capable of performing accurate exposure without using a large photomask. The exposure method performs exposure while relatively moving a photomask above a substrate and includes a step of performing position correction of the photomask by performing, on a front side of the photomask relatively moved in a moving direction, image recognition of a pattern prearranged on the substrate such as a line and a black matrix and by correcting deviation of the photomask with respect to the pattern, and a step of checking the position correction of the photomask by performing image recognition of a reference mark arranged on the photomask and by determining whether or not the position correction of the photomask is accurately performed in the step of performing the position correction of the photomask.

The invention discloses an intelligent self-positioning system of a spraying robot and belongs to the technical field of machine vision detection. The intelligent self-positioning system is characterized by comprising an ARM controller (1), an image capture card (2), a display module (3), a keyboard input module (4), a wireless transmission module (5), a power module (6), a sound and light alarm module (7), an upper computer (8), a drive module (9), a horizontal extension unit (10), a spraying actuator (11), a vertical lifting unit (12), a background processing module (13), a CCD (charge coupled device) camera (14) and a light source module (15). Compared with the prior art, the intelligent self-positioning system has the advantages such as simple system structure, convenience in installation and debugging, high system reliability, high measuring precision, low power consumption, ease of popularization and convenience in maintenance.

The invention belongs to the field of magnetic particle imaging and particularly relates to a magnetic particle imaging three-dimensional reconstruction method, system and device based on a field freepoint (FFL). The magnetic particle imaging three-dimensional reconstruction method, system and device based on the FFL aim at solving the problem of low accuracy of a magnetic particle distribution three-dimensional image based on the FFL. The magnetic particle imaging three-dimensional reconstruction method comprises the steps that current signal data of an induction coil in the three-dimensional scanning process of a scanned target based on the FFL are obtained; based on the current signal data, deconvolution is conducted through a preset kernel function, and thus a two-dimensional image data set is obtained; the kernel function is a jump function for increasing L2 regularized constraint; based on the two-dimensional image data set, an initial three-dimensional image is obtained througha Wiener filtering deconvolution algorithm; and based on the initial three-dimensional image, deconvolution is conducted through a Langevin function, and a final three-dimensional image is obtained through Radon transformation. The magnetic particle positioning accuracy in the reconstructed three-dimensional image is improved.

A shield case of the invention includes a base of a generally U-shape in front view having widthwise end portions, the end portions each having a front surface and a back surface; a pair of folded-back portions, provided on the front the back surfaces of the end portions of the base and folded back backward, or alternatively provided on the back surfaces of the end portions of the base and folded back forward; a pair of side walls, extending from the folded-back portions along a length of the end portions of the base; a pair of first locking pieces, extending downward from the side walls; and a linking portion adapted to link between the side walls. The base, the side walls, and the linking portion define a body containing portion.

A valve element of a hydraulic multi-way valve comprises a valve element body, and the valve element body is provided with a first throttling groove set, a second throttling groove set, a third throttling groove set, a fourth throttling groove set and a fifth throttling groove set. The throttling groove sets of different shapes and different opening lengths are formed in the valve element body so that the flow of the valve element can be directly proportional to the whole process of valve element displacement; the impact phenomenon generated in the opening and closing process of the valve element can be effectively avoided, motion stationarity can be effectively achieved, and therefore microoperation performance and precise location of lifting or descending of a working device of a loader are improved.