42results about How to "The mapping relationship is accurate" patented technology

The invention discloses an image super-resolution enhancement method based on a bidirectional recursion convolution nerve network. The method comprises two parts, i.e., model training and resolution enhancement. The model training comprises: obtaining multiple groups of high and low-resolution images comprising the completely same content, extracting high and low-resolution image block sequences through a window sliding method, and accordingly, training a bidirectional recursion convolution nerve network model. The resolution enhancement comprises: segmenting low-resolution images to be processed into a group of low-resolution image blocks to be processed through the window sliding method, generating corresponding high-resolution image blocks through the well trained bidirectional recursion convolution nerve network model, and generating corresponding high-resolution images through fusion. According to the method, space position relations of adjacent image blocks are learnt by full use of the bidirectional recursion convolution nerve network, and accordingly, the high-resolution images comprising detail information such as more edge textures and the like are obtained.

The invention relates to a geographic position-based video search method and system. The method comprises the following steps: receiving a video search request sent by a client; analyzing a target keyword from the video search request; obtaining a geographic position corresponding to the target keyword; searching a video matched with the geographic position; and generating video display data according to the search video and returning the video display data to the client. According to the method, the target keywords input by the users are mapped into one or more geographic positions when video search is carried out, and then matched videos are searched on the basis of the geographic positions; and an accurate video search method is provided, so as to satisfy the accurate video search requirement.

The invention discloses a hybrid deep regression network-based head pose estimation method and belongs to the computer vision and machine learning field. According the method of the invention, on thebasis of a traditional regression algorithm and a deep network framework, image information of different modalities is applied to a head pose estimation problem for the first time; and sub-networks ofa plurality of modalities, which are obtained through training, are fused, so that a final hybrid deep regression network can be obtained. Compared with other head pose estimation methods, the algorithm of the invention can achieve higher estimation accuracy and have high robustness; and in addition, the method of the present invention has a certain universality, and is not only applicable to typical head pose estimation problems such as human-computer interaction, safe driving, and face recognition, but also can expand problem conditions to other deep regression problems.

The invention discloses a SAR image change detection method based on a complex neural network, which inputs two original SAR gray-scale images and uses a preliminary difference map obtained by a traditional method as a preliminary label. According to the preliminary tags, some pixel blocks in the two original images are selected as candidate training samples by confidence detection. A complex network is constructed, which consists of three layers of all connection layers and two layers of multiple batches of normalized layers. The complex training samples are selected from the candidate training samples according to the stochastic proportional method to train the complex network. The training complex network is used to test the test complex samples constructed directly from two original SAR images, and the final change detection results are obtained. The invention not only makes full use of the advantages of the traditional change detection result, but also displays the characteristicsof the original data, so that the neural network can better learn the relationship between the two images, thereby obtaining a better change detection result.

The invention provides a coordinate conversion method and device, relating to the field of a wafer scriber and solving the problems of inexact mapping relationship, influence to quality and efficiency of an image processing algorithm and influence to precision of a machine. The coordinate conversion method comprises the steps of: regarding one point as an original point of a coordinate system, and obtaining a first workbench coordinate difference vector of a workbench from the original point to a first target position and a first image coordinate difference vector of an image of a processing object on the workbench from the original point to the first target position; obtaining a coordinate conversion matrix based on the first workbench coordinate difference vector and the first image coordinate difference vector; performing conversion between the coordinate of the workbench and the coordinate of the image of the processing object based on the coordinate conversion matrix. Pixels are determined through non-single-direction movement of the workbench of the wafer scriber, so that the problem of mapping between the image pixel coordinate and the workbench coordinate is solved, and the mapping precision is improved.

The invention discloses a spatial double-arm system constraint motion planning method based on deep intensive learning. The method comprises the following steps of 1) establishing a kinematic model ofa free floating spatial double-mechanical-arm system by utilizing a generalized jacobian matrix; 2) designing a spatial double-arm system motion planning algorithm based on the kinematic model of thespatial double-arm system in combination with a DDPG algorithm; and 3) designing a reward function in the DDPG algorithm to realize the satisfaction of each constraint condition in the spatial double-arm system motion planning algorithm, including speed constraint of a mechanical arm end effector and self-collision constraint of double-mechanical-arm cooperative motion. The universality and safety of the motion planning algorithm is improved.

The present invention discloses an image-based texture generation method. The method comprises the following steps that: S1, the three-dimensional model and color coding index table of a target to be generated are established, and the index table are mapped onto the surface of the three-dimensional model as textures; S2, at least two images of the target to be generated at different angles are obtained; S3, outer contour matching is performed on the images and the three-dimensional model; S4, the index table is utilized to index the images which have been subjected to the outer contour matching, and the texture features of each of the images of the target to be generated are extracted; S5, different texture features are fused; and S6, the fused textures are mapped onto to the surface of the three-dimensional model as surface textures. According to the texture generation performed according to the steps of the method of the present invention, the textures can be generated with manual texture drawing not required; and accurate mapping relations can be established with manual adjustment of the locations of geometric model vertexes and texture details not required. The method has the advantages of high efficiency, simplicity and high automation degree.

The invention provides a pin mapping method and system to solve the problems that, in the prior art, the mapping method is large in working amount and low in efficiency and obtained pin mapping relationships are not accurate. The pin mapping method comprises calling pre-allocated script files under appointed catalogs and analyzing the script files which comprise a number of a main socket into which an auxiliary panel is inserted, a number of an auxiliary socket pin of the auxiliary panel which has a mapping relation and net names of pins of functional modules on the auxiliary panel; searching corresponding main panel number functional module pin numbers in a preset database according to the main socket number of the main panel and the auxiliary socket pin number of the auxiliary panel; integrating into an output file according to the pin net names of the functional modules on the auxiliary panel and the pin numbers of the functional modules on the main panel; outputting the output file. The pin mapping method and system has the advantages of being high in efficiency and accurate in obtained pin mapping relation.

The invention belongs to the field of artificial intelligence, and particularly relates to a false public opinion identification system based on information propagation characteristics and a processing method thereof. The system comprises a data acquisition and data management module, a false public opinion influence propagation calculation module, a user grouping module and a false public opinionidentification module. The system can judge whether a certain hot event is a false public opinion or not by analyzing the propagation rule of social information. The system is mainly applied to environments such as a mobile phone, wearable equipment and other intelligent terminal equipment, can calculate influence weights among users in a social network, analyzes a strong and weak relationship ofuser interaction in the social network, and constructs a more accurate mapping relationship between a graph model and the social network. According to the system, the user with the maximum influencein each group is mined on the basis of the aggregated group, the false public opinion tendency of the user group is estimated by analyzing the social data of the user, the false public opinion of thegroup interactive stir-frying is recognized, the accuracy is high, and the recognition speed is high.

The invention provides a three-dimensional image reconstruction method, comprising steps of extracting a two-dimensional image feature point of an input image; estimating the two-dimensional image feature point by using a first mapping relationship to obtain a three-dimensional image feature point corresponding to the two-dimensional image feature point, wherein the first mapping relationship is obtained by learning a mapping relationship between an original two-dimensional image feature point and an original three-dimensional image feature point in a training sample; expanding the training sample based on a second mapping relationship, wherein the second mapping relationship is obtained by learning a mapping relationship between the two-dimensional image feature point and the three-dimensional image feature point; and acquiring a full-pixel image based on the extended training sample and the three-dimensional image feature point. The three-dimensional image reconstruction method provided by the invention can expand the training sample, and effectively solves the problem of insufficient training samples in the full-pixel image reconstruction process.

The invention provides a controller area network (CAN) signal transmission verifying method for an automobile controller. The CAN signal transmission verifying method for the automobile controller comprises the following steps: according to the CAN signal, obtaining the first value range of the transmission of the CAN signal between a CAN bus and the bottom procedure of the controller, the second value range of the transmission between the bottom procedure and the intermediate port of the controller and the third value range of the transmission between the intermediate port and the application procedure of the controller; according to each value range, generating a test valve sequence; sending the corresponding the test value from the test valve sequence to the controller; according to the test value, generating the corresponding feedback value by the controller; and according to the test value and the feedback value, judging whether the CAN signal transmission of the controller is normal. The embodiment is capable of roundly and systemically detecting the CAN signal transmission of the controller, and avoiding the phenomenon of missed sanction. The invention further provides a CAN signal transmission verifying system for the automobile controller.

The invention relates to a photovoltaic output power ultra-short-term local emotion reconstruction neural network prediction method, and the method employs a singular spectrum analysis method to reduce the noise of original photovoltaic power, and can reduce the complexity of a photovoltaic power time sequence. Implicit fluctuation information mining of a photovoltaic power time sequence through achaotic nonlinear dynamics method is facilitated, meteorological data does not need to be acquired in advance in actual engineering, and the problem of model complexity caused by improper selection of input parameters and the problem of error accumulation caused by inaccurate prediction of the meteorological data can be avoided; the built local emotion reconstruction neural network does not needto determine the number of hidden layer nodes, modeling is simple, and the prediction cost is reduced; the established local emotion reconstruction neural network determines expansion signals and emotion parameters according to photovoltaic power chaotic attractors, so that the prediction model more pays attention to tracking each input mode and selects the most useful input mode information, themapping relationship is more accurate, the prediction accuracy is higher under different weather conditions, and the adaptability is higher.

The invention provides a steel support axial force intelligent adjustment method based on a bp neural network. The method comprises the following steps that an initial axial force is set according toa design value; data of loading days of axial force of a steel support, an axial force mean value of the steel support every day and accumulated deformation of a foundation pit every day are acquired,training is conducted by taking the three data as samples through a bp neural network model, and parameters of the bp neural network model are stored after the training result meets requirements; theaxial force mean value of the same day and the loading days of the next day in the sample data are input into the bp neural network model for calculation to obtain the predicted deformation of the foundation pit of the next day; if the predicted deformation of the foundation pit is smaller than a threshold value, the initial axial force is kept; and if the predicted deformation amount is larger than the threshold value, the axial force is adjusted and continuously input into the bp neural network model for calculation until the predicted deformation amount of the foundation pit meets the requirement, and the finally adjusted axial force is taken as a set axial force. The purpose of actively adjusting the axial force in advance to control the deformation of the foundation pit before the deformation of the foundation pit occurs is achieved.

Embodiments of the invention disclose an upgrading processing method and a related device. The upgrading processing method comprises the steps that the upgrading processing device sends an interface conversion description file Dv2-v1 to a device in which a business software module A is located, wherein the interface conversion description file Dv2-v1 is used for describing an interface conversion algorithm for conversion from a business software module B with a version v2 to the business software module A with a version v1; and upgrading data UB_v1-v2 and an interface conversion description file Dv1-v2 are sent to a device in which the business software module B is located, wherein the upgrading data UB_v1-v2 is used for upgrading the version of the business software module B from the version v1 to the version v2, and the interface conversion description file Dv1-v2 is used for describing an interface conversion algorithm for conversion from the business software module A with the version v1 to the business software module B with the version v2. According to the technical scheme provided by the embodiments of the invention, the business continuity in a gray-level upgrading process can be ensured as far as possible.

The invention provides a pin mapping method and system to solve the problems that, in the prior art, the mapping method is large in working amount and low in efficiency and obtained pin mapping relationships are not accurate. The pin mapping method comprises calling pre-allocated script files under appointed catalogs and analyzing the script files which comprise a number of a main socket into which an auxiliary panel is inserted, a number of an auxiliary socket pin of the auxiliary panel which has a mapping relation and net names of pins of functional modules on the auxiliary panel; searching corresponding main panel number functional module pin numbers in a preset database according to the main socket number of the main panel and the auxiliary socket pin number of the auxiliary panel; integrating into an output file according to the pin net names of the functional modules on the auxiliary panel and the pin numbers of the functional modules on the main panel; outputting the output file. The pin mapping method and system has the advantages of being high in efficiency and accurate in obtained pin mapping relation.

The invention discloses a stained image segmentation method integrating a deep neural network and a CV model, which belongs to the technical field of image processing and solves the problems of using the existing neural network training samples for image segmentation and the unsupervised nature of the CV model. The technical solution is: constructing a random contamination model, loading the image data of the contamination image, constructing a contamination image fusion segmentation model, extracting the characteristics of the contamination image, and using the energy function of the CV model to complete the error back propagation to update the network parameters; After multiple rounds of iterative learning, the defiled image fusion segmentation model is used to complete the segmentation of the defiled images; the present invention integrates the unsupervised characteristics of the CV model into the neural network, so that the neural network can obtain accurate images with a small number of training samples. Mapping relationship, applied to the segmentation of defiled images.

The invention provides an ANN-based optical characteristic modeling database generation method, which comprises the following steps of generating training data of an ANN, the training data at least comprising an optical characteristic parameter and a grating size parameter; training the grating size parameters based on an ANN training model so as to learn and obtain a mapping relationship between the optical characteristic parameters and the training labels of the grating size parameters; evaluating the quality of grating size parameter training in the step S2; and performing experimental verification on grating size parameters generated by the model obtained by training in the step S2 based on real optical characteristic parameters obtained by an optical measurement means; the accurate mapping relation between the optical characteristic parameters and the grating size parameters is obtained through repeated learning optimization in the step S3 and experimental verification in the step S4, so that an optical characteristic modeling database is established, and the defect that a relatively accurate database is difficult to establish efficiently in a time-saving mode through an existing optical characteristic modeling database generation method is overcome.

The invention, which relates to the field of tilted imaging relative positioning method, discloses a tilted shooting relative positioning method based on depth learning. The method comprise: step one,after preprocessing collected data, classifying the collected data into training data and test data; step two, establishing a basic matrix by the training data and carrying out reduced rank constraint on the basic matrix to calculate an initial value of a relative orientation element; step three, with the basic matrix as input data of the established deep learning neural network and the initial value of the relative orientation element as the output data, acquiring a final value of the relative orientation element based on a set iteration condition; and step four, inputting the test data intothe trained deep learning neural network to obtain a relative orientation element test value. Therefore, a defect that the positioning accuracy is greatly affected by the conversion accuracy in the method of obtaining the orientation element based on conversion with the essential matrix is overcome; the relative orientation element can be obtained precisely; and the relative positioning accuracyis improved.

The invention discloses an article space dose distribution detection method and device. The article space dose distribution detection method comprises the steps that a dose distribution standard modelis constructed, wherein the dose distribution standard model comprises a standard mapping relation between dose distribution data and density values of an article under irradiation of energy of a preset size; perspective scanning is performed on the article to be detected, and a three-dimensional perspective image of the article to be detected and detection density values of all space points of the article to be detected are obtained; the dose distribution data of all the space points of the article to be detected are obtained according to the detection density values of all the space pointsof the article to be detected and the standard mapping relation of the dose distribution standard model; and the dose distribution data of all the space points of the article to be detected are matched with the three-dimensional perspective image and displayed. According to the article space dose distribution detection method and device, the dose distribution data of all the three-dimensional space points of the article to be detected are more visual and accurate, and reasonable configuration of various parameters in the irradiation process is facilitated.

The invention discloses a bone traction needle force sensing system based on FBG optical fibers, the bone traction needle force sensing system comprises optical fibers and a bone traction needle, and the optical fibers comprise FBG optical fibers 1-4; the FBG optical fibers 1-3 are arranged on the outer surface of the bone traction needle along the length direction of the bone traction needle; the FBG optical fiber 4 is arranged in a hollow shaft of the bone traction needle in the length direction of the bone traction needle, coincides with the central axis of the bone traction needle and is used for sensing the axial force borne by the bone traction needle. The outer surface of the bone traction needle is radially provided with three grooves at equal intervals, the included angle between every two grooves is 120 degrees, and FBG optical fibers 1-3 are arranged in the grooves and used for sensing the radial force borne by the bone traction needle. According to the method, the mapping relation between the surgical instrument deformation and the reset force is constructed, and the instrument-bone-soft tissue interaction force sensing model is established, so that unnecessary injury to a patient due to overlarge reset force in the reset operation process is avoided.

The invention provides a coordinate conversion method and device, relating to the field of a wafer scriber and solving the problems of inexact mapping relationship, influence to quality and efficiency of an image processing algorithm and influence to precision of a machine. The coordinate conversion method comprises the steps of: regarding one point as an original point of a coordinate system, and obtaining a first workbench coordinate difference vector of a workbench from the original point to a first target position and a first image coordinate difference vector of an image of a processing object on the workbench from the original point to the first target position; obtaining a coordinate conversion matrix based on the first workbench coordinate difference vector and the first image coordinate difference vector; performing conversion between the coordinate of the workbench and the coordinate of the image of the processing object based on the coordinate conversion matrix. Pixels are determined through non-single-direction movement of the workbench of the wafer scriber, so that the problem of mapping between the image pixel coordinate and the workbench coordinate is solved, and the mapping precision is improved.

A structured-light measuring method, includes: matching process, in which the number and the low-precision depth of a laser point are achieved by using the imaging position of the laser point on a first camera (21) according to a first corresponding relationship in a calibration database, and the imaging position of the laser point on a second camera (22) is searched according to the number and the low-precision depth of the laser point so as to acquire the candidate matching points, then the matching process is completed according to the imaging position of the first camera (21) and the candidate matching points of the imaging position of the first camera (21) on the second camera (22) so that a matching result is achieved; and computing process, in which the imaging position of the second camera (22) matching with the imaging position of the first camera (21) is achieved according to the matching result, and then the precision position of the laser point is determined by a second corresponding relationship in the calibration database. A structured-light measuring system utilizes the above measuring method.

The invention belongs to the field of artificial intelligence, and in particular relates to a false public opinion identification system based on information dissemination characteristics and a processing method thereof. The system includes: a data collection and data management module, a false public opinion influence propagation calculation module, a user grouping module and a false public opinion identification module, which can determine whether a hot event is a false public opinion by analyzing the propagation rules of social information. The system is mainly used in environments such as mobile phones, wearable devices, and other smart terminal devices. It can calculate the influence weight of users in social networks, analyze the strength and weakness of user interactions in social networks, and build a more accurate graph model. Mapping relationship with social network. The system mines the most influential users within each group on the basis of aggregated groups, estimates the tendency of false public opinion of the user group by analyzing the social data of the user, and realizes the identification of group interactive hype and false public opinion with high accuracy , the recognition speed is fast.

The invention discloses an electronic whiteboard positioning method, which comprises the following steps of: 1, selecting four point coordinates in an electronic whiteboard coordinate system as whiteboard reference point coordinates, selecting four point coordinates in a computer screen coordinate system as computer screen reference point coordinates, and building a mapping relationship between two reference point coordinates; 2, substituting the coordinates into a whiteboard coordinate and computer screen mapping relationship to solve parameter values; 3, building a mapping relationship between electronic whiteboard measuring coordinates and computer screen points according to the parameter values obtained through calculation; and 4, performing any position positioning in a located region of an electronic whiteboard according to the mapping relationship. The electronic whiteboard positioning method has the advantages that an accurate mapping relationship between the electronic whiteboard measuring coordinates and the computer screen point coordinates is built; the errors of an electronic whiteboard plane due to ideal easel plane rotation are eliminated; the positioning method is simple; and the positioning effect is precise.