31results about How to "Accurate identification and positioning" patented technology

The invention provides a distributed big data log link tracking method and system under a micro-service architecture. The method comprises the steps of: after receiving a calling request, enabling a micro-service module to judge whether the current micro-service module is called for the first time or not, and obtaining a trace ID and a span ID; judging whether the current service method is a service method called for the first time or not, and updating the span ID relationship in the call chain; caching the IP address of the server where the micro-service module is located, the name of the micro-service module, the name of the service method, the traceID, the spanID, the parentSpanID and the starting execution time information of the service method; executing a log tracking record according to the link log tracking mark, binding the log information with the calling link information and the micro-service module operation performance information respectively, and sorting the log information according to the calling link information; and binding the performance data operated by the micro-service module with the calling link information, compressing the sorted log records, sending the compressed log records to a log storage server, deleting cache data corresponding to the calling link, and releasing the memory space.

The invention discloses an AGV automatic charging device, system and method. The automatic charging method includes: a central processing module starts a visual module after receiving a charging instruction; the visual module shoots an image of a to-be-charged AGV vehicle; the central processing module processes the image of the AGV vehicle and positions a charging electrode of the AGV vehicle; amechanical motion module controls butt joint of a charging contact and the charging electrode of the AGV vehicle according to position information of the charging electrode of the AGV vehicle under the control of the central processing module; a charging module detects a charging state of the AGV vehicle and correspondingly charges the AGV vehicle; after the charging module detects that charging of the AGV vehicle is finished, a charging circuit is closed; the mechanical motion module controls the charging contact to separate from the charging electrode of the AGV vehicle and controls the charging contact to return to an initial position. The AGV automatic charging device, system and method have advantages that self adaptability of the AGV automatic charging device is improved, and the requirement on AGV vehicle parking accuracy is lowered.

The invention discloses an adaptive target detection method based on a lightweight residual network and deconvolution cascading. The method comprises the following steps: obtaining an image training data set and a test data set; extracting the deep features of a to-be-detected image through a lightweight residual network in combination with deep separable convolution and residual learning to obtain the deep expression of a target; using 1 * 1 convolution to fixedly output feature map dimensions for the extracted feature maps of different levels; increasing the resolution of the deep-level feature map by using a deconvolution cascading structure so as to realize that the spatial size of the deep-level feature map is consistent with the spatial size of a previous-level feature map; guiding the candidate region generation network to adaptively generating a target candidate box which is more matched with the real target on the multi-scale feature map by utilizing the semantic features; andfinally, correcting the generated target candidate box Anchor. According to the method, the accuracy of target detection is effectively improved, the target can be rapidly and accurately detected under complex conditions, and the real-time performance of target detection is effectively improved.

The invention discloses a chess robot. The chess robot comprises a chessboard seat, a mechanical arm and a visual assembly; the chessboard seat comprises a chessboard seat bracket, and a chessboard area and a chess piece storage area are arranged on the upper surface of the chessboard seat bracket; the visual assembly comprises a camera and a camera bracket, the lower end of the camera bracket isfixedly installed on the chessboard seat bracket, and the camera is fixedly installed at the upper end of the camera bracket; and the mechanical arm comprises a small arm skeleton, a large arm skeleton, a tool end assembly and a mechanical arm bracket assembly, the mechanical arm bracket assembly is fixedly installed on the chessboard seat bracket, a first power mechanism for driving the large armskeleton to rotate is arranged on the mechanical arm bracket assembly, a second power mechanism for driving the small arm skeleton to rotate is arranged on the large arm skeleton, and the tool end assembly is fixedly installed at the free end of the small arm skeleton. The chess robot can better improve the experience degree and reduce the cost.

The invention discloses a large and small two-dimensional code mixed image landmark design method and a UAV landmark identification method. The large and small two-dimensional code mixed image landmark design method comprises the following steps that large and small two-dimensional code units which form a landmark are created, and each of the two-dimensional code unit includes only one corner witha white subunit; the size of the landmark is set, and 4 large two-dimensional code units and 4 small two-dimensional code units are selected to form a landmark image; the large and small two-dimensional code units are arranged in a 2*2 rectangular array manner; and the white subunits of the large two-dimensional code units are arranged in the right lower corner. The large and small two-dimensional codes of different sizes are used cooperatively, so that a UAV can be identified and positioned more accurately in a smaller landmark area within the identification range of 0.5m-5m, and the problems of visual information loss and too large landmark during positioning in the whole UAV landing process can be solved directly.

The invention relates to a converter fault diagnosis method based on kernel density estimation. The method comprises the steps of: performing pre-processing of collected data through cubic B-spline wavelet analysis based on a mallat algorithm to obtain samples with fault features; employing a KDE fault classifier to perform offline training to select better parameters of the fault classifier and accurately dividing the normal conditions and each type of fault condition included in the training samples, and using the better parameters into a classifier network to obtain the optimal parameters;implanting the classifier network with the optimal parameters into online simulation to perform real-time online monitoring fault diagnosis of an actual circuit; and allowing the classifier network with completion of optimal parameters to distinguish known fault type samples and normal samples, complete the location of the known fault types of faults and identify the unknown faults for achievementof circuit protection in a condition of generation of unknown types of faults. The converter fault diagnosis method based on kernel density estimation can determine the health condition of the converter more accurately and more reliably, and also can improve the efficiency of the fault diagnosis of the converter.

The invention discloses an intelligent visual garbage sorting robot for building garbage. The intelligent visual garbage sorting robot for the building garbage comprises a robot body, a visual area located at the front end of the robot body, a grabbing assembly located at the tail end of the robot body and a shell surrounding the robot body, and the grabbing assembly used for grabbing garbage is installed at the tail end of the robot body. Through a sensor in the visual area, the robot body in multi-axis connection and the grabbing assembly, automatic recognition is achieved, and meanwhile automatic following and garbage grabbing are achieved; and the intelligent visual garbage sorting robot for the building garbage is accurate in recognition and positioning and high in operation precision.

The invention discloses a library book classification management device based on big data, which comprises a moving mechanism, a lifting mechanism, a telescopic mechanism, a control mechanism, a support frame and a classification mechanism, and is characterized in that the moving mechanism is positioned on the upper side of the lifting mechanism, and the telescopic mechanism is positioned on the lifting mechanism; the control mechanism comprises a PLC, a color detection controller, a code scanning module, a color sensor and a touch screen, the PLC is installed at the front end of the supporting frame, and the color detection controller is installed at the front end of the supporting frame. According to the invention, the color detection controller and the color sensor are used for detecting different types of books according to the plurality of color bars attached to the two sides of the bookshelf, so that the positions of the required books can be quickly detected, and the PLC controller and the code scanning module are used for identifying bar codes attached to the side edges of the books, therefore, the books can be quickly and accurately identified and positioned.

The invention discloses a graph defect recognizing and positioning method based on a latent Dirichlet allocation model. The method comprises a training phase and a testing phase. The training phase comprises a step of using the chaos theory to calculate pixel time series of trained images, extracting multiple characteristics, and forming chaotic feature vectors, wherein one trained image is represented by a chaotic feature vector matrix, a step of clustering all trained chaotic feature vector matrixes through a k-means clustering method, forming a code book, and then forming a primary histogram through the code book, and a step of learning the primary histogram through the latent Dirichlet allocation model to obtain advanced histograms, and representing the trained images by the advanced histogram. The testing phase comprises a step of representing a tested image by an advanced histogram 1, a step of calculating a level of similarity between the above advanced histogram 1 and the advanced histograms to judge the defect category and location of the tested image.

A dynamic monitoring system with radio-frequency identification includes a radio-frequency identification (RFID) tag, at least two radio-frequency identification (RFID) detectors and a processing device. The RFID tag is installed on a moving user. When the user enters a preset detection range, at least two RFID detectors detect the RFID tag and generate an identification signal, respectively. The processing device is coupled to at least two RFID detectors. The processing device generates a distance signal between the RFID tag and the at least two RFID detectors according to the identification signals and positions of the at least two RFID detectors. The processing device generates a positioning signal relevant to a position of the RFID tag within the preset detection range according to the distance signal.

The invention provides a device for mode recognition and positioning. The device comprises a male device and a female device which are in mutual precise matching, wherein one of the male device and the female device is a device with a recognized mark, and the other one is a device with a mark recognition sensor; the male device comprises a hollow cavity and a part for accommodating the male device; the bottom plane of the hollow cavity is a recognized mark placing surface; the part for accommodating the male device is provided with an opening in which the male device can be conveniently inserted; a positioning device is arranged on the opening; the male device comprises an image recognition sensor, a positioning post and a functional electronic device; the positioning post is matched with the positioning device of the female device; the image recognition sensor is used for recognizing the recognized mark; the functional electronic device is an electronic device used for driving a camera or other main equipment to perform connection and communication or an electronic device for display operation. The device provided by the invention has the advantages that the fast recognition is realized; the user experience is improved. The novel method and path for realizing interaction and data mutual transmission can be added between equipment.

The invention provides an recognition and positioning method and system based on a backlight module and multi-template matching. The method comprises the following steps: carrying out image collection on a target area based on an image collection module to obtain a to-be-recognized sole image; performing image smoothing processing on the to-be-recognized sole image through an image processing module to obtain template contour information of the to-be-recognized sole image; and matching in a preset sole template library based on the outline information of the to-be-recognized sole image template to obtain sole information corresponding to the outline information of the to-be-recognized sole image template. In the automatic production of the shoemaking industry, after the products are fed into the backlight module, the products of different models can be quickly and accurately recognized and positioned by utilizing a machine vision algorithm, cooperating with the backlight module and adopting a multi-template matching algorithm, the operation is simple, and the technical cost is low.

The invention relates to the field of welding and leakage-detecting, in particular to an intelligent welding and leakage-detecting device for a refrigerator press cabin pipeline. The intelligent welding and leakage-detecting device is characterized by comprising a welding system, a leakage-detecting system and a code, wherein the coder is arranged on a production line conveyor belt; the welding system comprises a first industrial controller, a first PLC and a first mechanical arm controller, which are connected to the first industrial controller, a first image acquisition trigger, a welding position image acquirer, a first mechanical arm trigger, a first mechanical arm and welding equipment; and the leakage-detecting equipment comprises a second industrial controller, a second PLC and a second mechanical arm controller, which are connected to the second industrial controller, a second image acquisition trigger, a welding point image acquirer, a second mechanical arm trigger, a second mechanical arm and leakage-detecting equipment. The intelligent welding and leakage-detecting device can improve welding quality, improves production efficiency and reduces labor cost.

The invention relates to the field of data monitoring, and discloses a risk identification method and device for public opinion data, computer equipment and a storage medium, and the method comprises the steps: obtaining public opinion data collected from a platform; identifying the public opinion data, and extracting feature information of the public opinion data; obtaining grade division of each piece of feature information from a feature database based on big data pre-collection, and matching a score value corresponding to the feature information according to the grade division of each piece of feature information; acquiring influence weights of the feature information, wherein the influence weights of the feature information are obtained based on historical public opinion data and risk marks of the historical public opinion data and through training of a machine learning algorithm; calculating a negative risk level of the public opinion data according to the influence weight of the feature information and the corresponding score value; and outputting the negative risk level of the public opinion data. The public opinion data can be accurately positioned, and the efficiency and accuracy of public opinion data risk identification are improved.

The invention relates to a tracking shooting system which comprises a management platform, a panorama camera and a detail camera, the panorama camera and the detail camera are connected with the management platform, the panorama camera is connected with the detail camera, and the management platform is used for calling the panorama camera and the detail camera to shoot a tracking target when receiving a shooting trigger signal; the panoramic camera is used for shooting the tracking target when the detail camera does not recognize the tracking target, acquiring position information of the tracking target and sending the position information to the detail camera; and the detail camera is used for carrying out tracking shooting on the tracking target and identifying the tracking target again based on the position information when the tracking target is not identified. By adopting the method, the tracking target can be identified and positioned more accurately, so that the tracking video is smooth and not blocked, and the target tracking effect is ensured.

The invention discloses a power distribution terminal operation data management system. The power distribution terminal operation data management system comprises a data acquisition module, a mobile terminal, an intelligent terminal, a control module and a data management module, the data acquisition module is in communication connection with the mobile terminal, the intelligent terminal is in communication connection with the mobile terminal, the control module is connected with the control module, and the data management module is connected with the control module. In the power distribution terminal operation data management system, the arranged mobile terminal is small in size and convenient to carry during cross-regional operation, various power distribution devices can be conveniently and automatically identified, and the current power distribution equipment is quickly and accurately identified and positioned, so that the data of the power distribution equipment can be classified and sorted according to different types of information, the sorting efficiency is improved, and the dynamic characteristics of various partial differential equation description systems are adapted through the distribution parameter model.