408 results about "Local space" patented technology

The invention belongs to the technical fields of mode indentifying and image processing, and relates to an image matching method based on characteristic points. The image matching method based on characteristic points defines an annular corner code model according to the K neighboring structure of the characteristic points and applies the model to point mode matching. Similar annular corner codes not only describe the similarity of the space structures of two characteristic points, but also can be used for local mapping estimation. The similarity of the two characteristic points is confirmed by the similar length of the largest similar annular corner codes associated with the two characteristic points. The image matching method based on characteristic points carries out structure matching according to the similarity of the local space structures of the characteristic points, and uses the local mapping and clustering to carry out optimization matching. The optimization matching result is used for optimizing the mapping estimation. The method can be used for the image matching based on the characteristic points, image jointing and embedding, motion tracing, information fusion of medical single-mode and multi-mode images, as well as image searching based on contents.

The invention discloses a fault tolerant control method for open circuit faults of a three-phase bridge PWM rectifier switching tube and belongs to the field of three-phase AC-DC (Alternating Current-Direct Current) conversion. The invention solves problems that multi-tube fault tolerant control is imperfect, the hardware cost is high and the like of the conventional fault tolerant scheme. According to the method, the characteristic that a local space voltage vector changes when a fault occurs to any switching tube of a three-phase bridge PWM rectifier is utilized; a PWM switching mode is corrected, i.e., a fundamental voltage vector is selected again and the action sequence and time is adjusted; and composition is conducted again to recover or approach an original rotary reference voltage vector, so that the normal operation of a system is maintained, and the fault tolerant control operation is achieved. According to the method disclosed by the invention, the effective fault tolerant control operation can be achieved when an open circuit fault occurs to a single switching tube and multiple switching tubes in the three-phase bridge PWM rectifier, and only a software algorithm is required. The method is simple and easy to achieve while no additional hardware cost is required.

The invention provides a cloud-computing-based HTML5 application packaging, installation, unloading and operation method and system. The system comprises a server and a plurality of clients, wherein interaction can be conducted between the clients and the server. Each client comprises a transceiver, a procedure file uploading module, an application configuration module, an application calling module, a procedure file loading module, a control creation module, an API interface module and an API and operation space set module. The server comprises a communication interface, a data center, a client code storage module, an application package analysis module, an application package creation module and a user identification module. The system has the functions of creating application packages, installing applications to a remote server, uploading the applications on the remote server, loading and applying the applications on the clients and closing the applications running on the clients. By means of the cloud-computing-based HTML5 application packaging, installation, unloading and operation method and system, users can install the applications to the remote server, the applications can be loaded to the local server to be run when in use, the local space is saved, and the applications are available at any place after being installed once.

The invention designs a blind channel balancing method based on an improved PSO (Particle Swarm Optimization) BP (Back Propagation) neural network. In the process of solving the blind balancing problem on the basis of a BP neural network, determination of an initial weight and a threshold of the BP neural network is lack of the theoretical basis and has the defects of low convergence speed, easiness for falling into a local minimal value and the like so as to cause a poor channel blind balancing effect. In order to overcome the defects of the BP neural network and improving the channel blind balancing effect, the invention discloses a blink balancing method based on the improved PSO-BP neural network. According to the method, firstly, defects of a basic particle swarm algorithm are overcome, parameters of the basic particle swarm are improved, and an inertia weight and a learning factor are adaptively regulated; secondly, the initial weight and the threshold of the neural network are optimized by utilizing the advantage of high global searching capacity of the improved particle swarm, and then more accurate searching is carried out in such local space by utilizing a BP algorithm soas to obtain an optimal connection weight and threshold of the neural network; and finally, blind balancing based on the the improved PSO-BP neural network is implemented.

The invention discloses a human body behavior identification method based on a depth sequence. The human body behavior identification method comprises the following steps: adopting a LBP (Local Binary Patterns) operator based on normal information; adopting a combined LBP operator of a spatial pyramid form; carrying out the sparse representation of the combined LBP operator; and carrying out the segmentation and the alignment of a behavior sequence. In order to obtain surface features which reflect different human body behavior surfaces in a depth map and further improve the robustness of human body behavior identification, a LBP description operator of normal information in the depth map is defined according to the similarity and the associated information of a human body structure in the depth map, wherein the operator keeps the geometric characteristics of a human body behavior surface on an aspect of details, extracts the local features of the surface on a local space, and expresses the local features as the human body behavior local features in the depth map. On the whole, detail information is integrated by a coding method based on dictionary learning, the local spatial structure relationship of a human body surface is kept by the pooling processing of an adaptive space-time pyramid and a thinning coefficient, and the detail and integral feature description of three-dimensional human body behaviors is realized.

The invention provides a tracking method of a relative movement angle and an angular velocity of a slowly rotating space fragment. The method comprises the following steps: after image information of a slowly rotating space fragment object is acquired, performing extraction, matching, tracking and pose solving of features points on current frames; performing object local space point three-dimensional reconstruction; performing object global closed-loop detection to eliminate accumulative errors; and obtaining an angular velocity and a total angle of object movement, and eliminating interframe revolution errors. According to the invention, from the perspective of image features, by use of a graph optimization method, tracking measurement of the slowly rotating space fragment is realized through a feature relay between key frames and key points, at the same time, the accumulative errors are eliminated by use of a closed-loop detection means, and thus the measurement precision is improved.

The invention relates to a method for file downloading on demand and automatic synchronizing on the basis of cloud storage and a device thereof. The device comprises a scheduling synchronization device, a cache management device and a monitoring device. The scheduling synchronization device is used for synchronizing the file change with a server locally and synchronizing the local modified files with the server automatically; the cache management device is used for managing cache file states and removing the cache; the monitoring device is used for capturing system behaviors. The server comprises a file service device and a storing device; the file service device is used for writing files uploaded by a client into the storing device and reading data of downloading files requested by the client and returning to the client; the storing device is used for storing and reading files. Compared with the prior art, the device has the advantages that the problems that a user hopes that used files can be accessed in an unimpeded manner and the local space is free of being occupied can be solved, and the operation can be completed on the conditions of changing the system using habits rarely and avoiding any interference.

The invention belongs to the technical field of remote sensing image processing and particularly relates to a dimension reducing and sorting method of a hyperspectral imagery based on blocking low rank tensor analysis. According to the dimension reducing and sorting method of the hyperspectral imagery based on the blocking low rank tensor analysis, a blocking idea is introduced into a dimension reducing method of the hyperspectral imagery based on the low rank tensor analysis according to three-dimensional data structures, spectral characteristics and correlated characteristics of local spaces of the hyperspectral imagery, the problems that integral spatial correlation of the imagery is weaker and negative effects of configuration of dimension-reduced subspace dimension on a dimension reducing effect are overcome, finally a novel dimension reducing method capable of greatly improving total sorting precision of the imagery is realized, and the novel dimension reducing method is a blocking low rank tensor analysis method. An algorithm presents well applicability on various different hyperspectral data (including simulation data and actual data sets). The dimension reducing and sorting method of the hyperspectral imagery based on the blocking low rank tensor analysis has important application values in high-precision terrain classification aspects based on the hyperspectral imagery.

The invention discloses a method for uploading, browsing and deleting pictures. The method for uploading the pictures comprises the steps that an indication through which the pictures stored locally are uploaded to a cloud end is received; information of the pictures uploaded to the cloud end is obtained; the obtained information of the pictures and the pictures are uploaded to the cloud end together. When a user browses the pictures stored at the cloud end through a cloud photo album of a terminal, whether pictures which are the same as the pictures at the cloud end exist locally is searched according to the picture information, if the pictures exist, the pictures do not need to be downloaded from the cloud end, and accordingly terminal local space is saved. The invention further discloses a device for achieving the method.

The invention relates to an image segmentation method based on improved intuitionistic fuzzy C-means clustering, belonging to the image segmentation field. Firstly, an improved non-membership functionis proposed to generate intuitionistic fuzzy sets, and a method based on gray feature is proposed to determine the initial clustering center, which highlights the role of uncertainty in intuitionistic fuzzy sets and improves the robustness to noise. Secondly, the improved nonlinear function is used to map the data to the kernel space in order to measure the distance between the data points and the clustering centers more precisely. Then we introduce the local space-Gray level information, considering membership degree, gray level features and spatial position information. Finally, the intuitionistic fuzzy entropy in the objective function is improved, and the fuzziness and intuitionism of the intuitionistic fuzzy set are considered. The invention can effectively overcome the influence ofnoise and blur in the image on the algorithm, improves the segmentation performance of the algorithm, the pixel clustering performance and the robustness, is suitable for various different types of gray-scale images, and can obtain more accurate segmentation results.

The invention provides a method and a system for measuring parameters of round holes in a plate. The method comprises the following steps: obtaining inner and outer parameters of a multi-view camera; performing image acquisition on the plate by utilizing the multi-view camera to obtain a local plane grey image of the plate; obtaining pixel values of characteristic parameters of linear segment end points and the round holes in the plate according to the local plane grey image; determining local space parameters of the round holes in each pair of binocular camera coordinate systems according to the pixel values of the characteristic parameters of the round holes in the local plane grey image and the inner and outer parameters of the multi-view camera; switching and concatenating the local space parameters of the round holes in each pair of binocular camera coordinate systems through the inner and outer parameters of the multi-view camera, and determining overall space information of the round holes in the plate. According to the scheme, the measurement precision and the completeness of the holes in the plate are improved, the measurement speed of the holes in the plate is increased, and the measurement cost of the holes in the plate is reduced.

The invention provides a device and method for calibrating a non-contact type electrostatic voltmeter. The method includes the steps that a direct-current high-voltage power supply and metal plate electrodes are utilized to acquire a standard static filed at local space near the centers of the electrodes, the voltage of the non-contact type electrostatic voltmeter can be traced to the direct-current voltage standard, and an experimental method that the edge effect is increased is adopted to test the edge effect of the metal plate electrodes. The risks that people get an electric shock are reduced through a glass flat plate support which bears the metal plate electrodes and through a protective resistor, and a perpendicular moving mechanism can adjust the test distance of a calibrated meter. A detachable vernier caliper is adopted, and therefore tracing is convenient to conduct.

The invention provides a video driver behavior identification method based on a multi-task space-time convolutional neural network. A multi-task learning strategy is introduced into a training processof a space-time convolutional neural network and is applied to driver behavior identification in a monitoring video. Auxiliary driver positioning and optical flow estimation tasks are implicitly embedded into the video classification tasks, so that the convolutional neural network model is promoted to learn richer driver local space and motion time characteristics, and the driver behavior recognition accuracy is improved. Compared with an existing driver identification method, the multitask space-time convolutional neural network architecture designed by the invention combines interframe information, is high in generalization and identification accuracy, can be used for real-time driver behavior identification under a monitoring video, and has an important application value in the field of traffic safety.

The invention discloses a method for completing continuous missing data of dam deformation monitoring, which solves the problem of completing continuous missing data of dam deformation monitoring. Firstly, the continuity loss of deformation monitoring data is pretreated. Secondly, from the angles of global space, global time, local space, local time and semantics, the spatio-temporal similarity and functional similarity of deformation monitoring points are calculated respectively, and the missing data in deformation monitoring data are interpolated globally and locally in space, time and semantics. Finally, a depth neural network model is constructed to complete the continuous missing data in dam deformation monitoring. The preliminary results of the missing data are used as input, and thenonlinear fusion is realized by using the depth neural network representation ability.

The invention discloses a robust FCM image segmentation method comprising the following steps: setting a fuzzy index, an iteration stop threshold and a maximum number of iterations, and initializing the clustering number and a neighboring window; initializing the membership degree, and calculating the clustering center of a 0th iteration; calculating a fuzzy factor according to the gray value difference between a sample point and an initial clustering center; using the fuzzy factor to calculate the membership degree of iteration; recalculating the clustering center of iteration; if judging that the difference between the membership degree before iteration and the membership degree after iteration is smaller than the iteration stop threshold or the number of iterations exceeds the maximum number of iterations, completing image segmentation and outputting an image after segmentation; or repeating next iteration until the condition is satisfied. According to the invention, a new fuzzy factor is built, the local space information and gray value information are fully utilized and directly act on an original image, the details of the original image are preserved as many as possible, and the robustness to noise is enhanced. The advantage of the robust FCM image segmentation method is more prominent especially in the case of high noise.

Disclosed is an optimization design method for an air vehicle lifting surface structure. The method includes utilizing a Latin hypercube design (LHD) method to extract experiment design samples and obtain a high-accuracy analysis model response value of an air vehicle lifting surface; constructing a radial basis function (RBF) broker model; generating a certain amount of grid points randomly by using the LHD method in design space, and calculating a corresponding response value of the RBF broker model; setting a global space reduction rate M1% and a local space reduction rate M2%, and performing two times of pseudo reduction; optimizing the RBF broker model by using a genetic algorithm in each clustering space; obtaining the space distance between an optimal point and a last-time space global approximate optimal point, and deleting the current space to obtain novel space if reduction criteria are met; and obtaining an actual response value of a current global approximate optimal solution, determining whether the value is convergent, and stopping optimizing if the value is convergent, otherwise, performing iteration repeatedly till the optimal solution is found. According to the optimization design method for the air vehicle lifting surface structure, the optimization efficiency is improved, and the optimization design cost of the air vehicle lifting surface structures saved.

The invention discloses a method for tracking and identifying a headlight and a taillight of an automobile under visual perception. The method comprises the steps of video image acquisition, local space automotive light detection, global space automobile detection, automotive light track real-time tracking, and matrix type LED particle control. On the basis of intelligent image recognition and tracking technologies, a camera is used for collecting and storing an automotive head target and an automotive tail target in front of an automobile; with a target detection and tracking technology and a feedback correction model, orientation, speed, and position motion parameter detection of the automobile by a vehicle in front of the automobile can be realized; and with an image view and an LED illumination area mapping model, adaptive controlling of single or grouped light brightness in a matrix type LED headlamp is realized by using a visual image sensing parameter as a control input. Therefore, stable multi-target vehicle tracking can be realized on the condition of complex light and the headlight of the automobile can be adjusted in real time based on a tracking result, so that the method can be applied to a headlight intelligent control service in auxiliary automobile driving.

The invention discloses a system and method for positioning an indoor three-dimensional space based on movable anchor nodes. The method includes the following steps: a wireless sensor network is established in the space, the anchor nodes in the network are divided into two sets, and the two sets are named planeXY and planeYZ respectively; According to local space environment, an appropriate XOY plane and an appropriate YOZ plane are chosen, all the anchor nodes in the planeXY set are deployed in the XOY plane, and all the anchor nodes in the planeYZ are deployed in the YOZ plane; the anchor nodes in the planeXY set can respectively move in the vertical direction in a periodic mode; the anchor nodes in the planeYZ set can respectively move in the horizontal direction in a periodic mode; the anchor nodes in the planeXY set are used for determining an X coordinate and a Y coordinate, wherein the X coordinate and the Y coordinate are provided with positioning blind nodes; the anchor nodes in the planeYZ are used for determining a Y coordinate and a Z coordinate, wherein the Y coordinate and the Z coordinate are provided with positioning blind nodes; according to coordinate values obtained through the two sets of anchor nodes, three-dimensional space coordinates of the blind nodes are determined, the movable anchor nodes are used for positioning, therefore, the positioning problem of the three-dimensional space of the wireless sensor network is solved, and the system and method for positioning the indoor three-dimensional space based on the movable anchor nodes have the advantage of being wide in positioning space.

The invention discloses and provides an effective multi-focus image fusion method based on the transformation between Pulse Coupled Neural Vetwork (PCNN) and Laplacian Pyramid (LP). Firstly, the LP is utilized to perform multi-scale decomposition to images with a tower structure and the PCNN is utilized to process the decomposed images of each of the scales so as to acquire a neurons firing frequency diagram of a describing feature cluster; the fusion to the images at each of LP decomposition scales is realized, based on the local space frequency (LSF) of the firing frequency diagram; and lastly, the fusion to the multi-focus images are realized through restricting algorithm of LP decomposition. The experimental results show that the multi-focus image fusion result acquired by the method of the invention has advantages over various traditional fusion algorithms at the aspects of subjective visual effect and objective evaluation index; and the method shows good performance.

The invention relates to an anchor-node-free three-dimensional wireless sensor network physical positioning method. In the method, the concept of partially concentrating and global distributing is adopted to solve the common problem of reversal differences in distance measuring methods, and achieving precise physical positioning is achieved through low calculated amount and communication traffic. The method comprises the steps of: 1) initializing a WSN (Wireless Sensor Network) system, searching an adjacent node of each sensor node through each sensor node, and measuring the distance and relative angle information between each sensor mode and the adjacent node; 2) establishing each local space rectangular coordinate system of each node, wherein all rectangular coordinate systems meet a right-hand rule, and the axis z always points to the same half space; 3) calculating the local coordinate of each adjacent node; 4) solving a three-dimensional coordinate system transformational matrix based on homogeneous coordinates through the adjacent node, and transforming the transformational matrix to calculate a coordinate relative to another local coordinate system; and 5) selecting the origin of global coordinate system, and through the recursion transmission of the coordinate system transformational matrix, calculating the global physical coordinate of each node.

The invention discloses an interactive natural scene text detection method. The interactive natural scene text detection method comprises the following steps of firstly labeling an area where an interested text is located in a scene picture, i.e. drawing a line on a text; selecting an expansion area from an original picture according to the line, carrying out edge detection in the expansion area, carrying out connected area labeling on an edge detection result, then carrying out projection on a boundary which is labeled with the result, and obtaining an edge text area; then carrying out operations of partitioning binaryzation, connected area labeling and projection on the edge text area, and obtaining a rough text area; and finally carrying out the operations of partitioning binaryzation and inverse color binaryzation on the rough text area, respectively carrying out connected area labeling, removing of various noises, boundary projection and connected area number determination on two binaryzation results, determining the polarity of the picture according to connected area number of two binary images, integrating connected areas, and obtaining an accurate text area. The interactive natural scene text detection method disclosed by the invention has adaptability on local space and has better detection effects for complicated-background text blocks in a natural scene picture.