265results about How to "Run fast" patented technology

The invention discloses a city regional railway signal system and a control method thereof. The city regional railway signal system comprises a control center device, a plurality of station devices, a line side unit and a vehicle device; the control center device is used for obtaining the running information of every train and outputting the movement information of the trains to the station devices after processing; the station devices are used for outputting the train movement information output from the control center device and the route information established through the line side unit to the vehicle device after processing; the line side unit is used for detecting the train position information, establishing train routes according to the interlocking information, enabling the route information of the trains to be generated into the message information and outputting the message information to the vehicle device; the vehicle device is used for adjusting the running state of the trains according to the train route control information and the message information and feeding back the adjusted running state of the trains to the station devices. According to the city regional railway signal system and the control method thereof, the functions such as the interconnection and intercommunication of the line network interior, accurate stopping, shielding gate linkage and interval automatic running and automatic turn-back can be achieved based on an ETCS-1 (Europe Train Control System-1) platform.

The invention discloses a mobile GIS (Geographic Information System) system based on an intelligent mobile phone. The mobile GIS system comprises a wireless communication network, a GPS positioning device and a mobile terminal as well an internet and a geographic application server, wherein the mobile terminal is the intelligent mobile phone used by a mobile subscriber; the mobile GIS system is characterized in that the mobile terminal is provided with the following software functional modules: an electronic map module, a map display module, a layer management module, a graphic editing module, an information inquiry module, a database management module, a style setting module, a GPS positioning and sampling module, a communication module as well as a map file module, a mobile database module, a GIS desktop integration platform module, a spatial database module and a shapefiles/Mif module. The mobile GIS system based on the intelligent mobile phone, disclosed by the invention, is mainly applied to digital city grid management and has the characteristics of simplicity and convenience in use and high storage efficiency.

The invention relates to a method for identifying information of an image tag in the technical field of image processing. The method comprises the following steps of: building a template image library; performing Hessian characteristic point detection on a template image and an image to be identified respectively; performing SURF characteristic extraction and LBP characteristic extraction on each characteristic point in the template image and the image to be identified respectively; performing SURF characteristic matching on the image to be identified and the template image to acquire the template image successfully prematched; performing LBP characteristic matching on the image to be identified and the template image successfully prematched to acquire the template image successfully prematched for the second time; and weighting the template image successfully prematched for the second time, wherein the template image with the maximum weight value is the template image most matched with the image to be identified. The method can effectively identify the image tag with complex background, scale change, direction change and visual angle change, avoids large-scale machine learning and complex image preprocessing at the same time, and has the advantages of high calculation speed, high identification accuracy rate and wide application scenarios.

The invention relates to a defect detection method of an irregular shape metal working surface based on depth learning, By preprocessing the collected surface images of metal machining, image enhancement, YOLOv3-based network architecture, Deep learning network suitable for defect detection is built. The samples are labeled manually and the deep learning network is trained with labeled samples toobtain defect detection model. Finally, the defect detection model is used to detect the surface image of metal machining, and the defect detection results are obtained. Compared with the prior art, the network structure in the invention has good adaptability to the detection of small objects and small targets, and unifies the four basic steps of the candidate region generation, the feature extraction, the classification and the position refinement of the target detection into the same depth network framework, thereby improving the running speed and making the detection more accurate.

The invention discloses a sine stripe projection device and a three-dimensional profile measuring method realized by using the sine stripe projection device. The method comprises the following steps of: (S1) correcting the intensity of each wavelength light emitted from the sine stripe projection device; (S2) acquiring the actual phase shift between every two wavelength stripes; (S3) marking the camera, and carrying out phase-height marking on the projection of each wavelength stripe in sequence; (S4) projecting each wavelength stripe to a measured object in sequence through the projection device, shooting each wavelength stripe projection image of the surface of the measured object through the camera, calculating out the initial phase of each measuring point through each wavelength stripe projection image, and unwrapping the initial phases; and (S5) respectively acquiring the height coordinate of each measuring point through the unwrapped initial phase of the measuring point according to the phase-height marking, and acquiring the three-dimensional coordinate of the measuring point through the camera marking. The measuring scheme has the advantages of high operating speed, high measuring precision and low cost.

The invention provides a robot positioning and mapping method, a computer device and a computer readable storage medium. The method comprises the following steps of pre-integrating information acquired by an inertial measurement unit by utilizing information acquired by a camera device and information acquired by the inertial measurement unit; using the front end module to carry out motion tracking by utilizing pre-integrated relative pose information, and judge whether a visual tracking loss condition occurs or not, if the visual tracking loss condition does not occur, carrying out the visualand inertial measurement fusion operation, and updating the state information of an inertial measurement unit; and when a new key frame is generated, using the rear end module to acquire the pose information of the key frame acquired by the front end module and the corresponding relation information of the feature points, create and optimize a factor graph, and update a rear end map by using theacquired pose information and the depth information of the 3D points. The invention also provides a computer device for realizing the method and a computer readable storage medium. According to the invention, the positioning and mapping of the robot can be realized efficiently and accurately.

The invention discloses a multi-layer anomaly detection method based on network traffic. The invention is capable of detecting the small traffic attack behavior well with high detection accuracy, andmay adapt to different data sets. The invention comprises: firstly adopting a binary representation of symbol attributes in the data preprocessing stage to eliminate the negative influence of the traditional numerical size on the classification, and raising the attribute set of the data set to a relatively high dimension, so that the subsequent data classification effect is more accurate; then using the dimension reduction method to extract features and reduce the amount of data, so that the running speed is faster and the memory consumption is lower during the subsequent steps; subsequently,using the KNN outlier detection method and genetic algorithm combination method for data selection, so that different types of data are more balanced, each type of data is separated as far as possible, and the classification result is fairer; finally, using the constructed multi-layer classifier, thereby enabling more accurate identification of large-flow attacks and small-flow attacks.

The invention relates to a digital television audio and video technology and provides a method for determining the displaying content of widget by single television program content. The method based on television widget for realizing bidirectional interaction advertisement comprises the following steps: a. a widget engine and a local application program run automatically after a set-top box is started; b. the local application program obtains the current program information; c. the local application program transmits the current program information to a front end application server; d. the front end application server calls the widget exhibition advertisement data file corresponding to the classification according to a classification of the current program information and sends the widget exhibition advertisement data file to the set-top box; and e. the set-top box displays the widget advertisement exhibition on a display terminal by the widget engine. The method has the beneficial effects that the advertisement display made by widget is released in real-time and pertinently according to the program content watched by users.

The invention discloses a battery sorting system. The battery sorting system comprises a code scanning mechanism, a voltage internal resistance measuring mechanism and a sorting mechanism, wherein the code scanning mechanism, the voltage internal resistance measuring mechanism and the sorting mechanism are respectively connected through a material channel. The invention simultaneously also discloses an automatic sorting method. The battery sorting system disclosed by the invention has the advantages that the full automation progress is adopted, the operation speed of each mechanism is rapid, the precision rate is high, the voltage internal resistance after measuring and capacity data which is read from a server according to a bar code are calculated to obtain battery grade information which is then written into a server database, the batteries are controlled to be sent to a set material groove through a material loading platform are automatically classified, no manual intervention is needed when the batteries enters the sorting system and are sorted, and the personal errors are greatly reduced.

The present invention provides a system of supporting multi-language user interface, the system includes an update processing subsystem and a switch controlling subsystem; the update processing subsystem substitutes a global string pointer for an original language field in the user interface, determining an object language field and its loading address of each language corresponding to the original language field, transmitting the global string pointer, the object language field of each language and its loading address to the switch controlling subsystem; the switch controlling subsystem receives an object language of user determination, valuating the global string pointer received from the update processing subsystem according to the object language field and its loading address corresponding to the object language, refreshing the user interface. The invention improves the speed of switching language, reducing the development difficulty. The invention also provides a method of supporting multi-language user interface, capable of improving the speed when switching the language, reducing the development difficulty.

The invention discloses an authority limit check system, device and method for accessing and saving operation under the sectionalized protection mechanism in the complicate instruction set computer sectionalized protection mechanism technique field. The method comprises the following steps: A, judging accessing and saving instruction type of receiving self accessing and saving instruction launching device by the accessing and saving operation authority limit check device, proceeding the procedure B if the accessing and saving instruction is the special accessing and saving instruction, proceeding the procedure C if not; B, checking if the field selector type, the segment descriptor word type and the field franchise of the special accessing and saving instruction is legal by the accessing and saving operation authority limit check device, ending the check if all are legal, proceeding the procedure D if any one isn' t legal; D, transferring the exceptional signal to the exceptional dealing device by the accessing and saving operation limit check device. The invention achieves the authority limit check for accessing and saving operation.

The invention discloses a page turning method for E-books. The page turning method for E-books includes the steps of firstly, automatically acquiring screen size and display data of a window of an E-book, and initializing the acquired data; secondly, starting to monitor on-click events of a mobile phone; and thirdly, when page turning is monitored, calculating a click position for page turning, refreshing a current interface, controlling implementation of animation simulation, and refreshing the interface. True page turning can be simulated by the page turning method for E-books, users can be attracted, and reading E-books is amused.

The invention discloses a multi-motor torque output and distribution controlling method. The multi-motor torque output and distribution controlling method comprises the following steps of: a, calculating torques, needed by a vehicle during running, by using vehicle controller modules; b, comparing motor working efficiency of different motor numbers during working, selecting the motor number with the high efficiency, calculating the total motor working efficiency at different torque distributing modes according to the selected motor number, and selecting a torque scheme when the total motor working efficiency is highest as the optimal scheme; c, distributing the motor torques according to an optimal scheme; d; distributing the torques, needed by the vehicle during running, to the motor controller modules with a set number; e; exerting the motor controller modules to control the adaptive motors to work according to the distributed torques; and f, exerting the motors to output the torques to the same driving shaft. According to the scheme, the motor working efficiency can reach the highest level through small calculated amount so that the energy is saved and the driving mileage is increased. The scheme is applied to electric automobiles, which are driven by two and more non-independent motors.

The invention relates to a hyperspectral real-time detection method based on recursive analysis. The method is characterized by including 1, establishing a spectral vector related matrix R(n) of hyperspectral image pixels; 2, establishing a state equation for the related matrix R(n), and updating an estimated value of current state according to an observed value rn of current state and an estimated value R(n-1) of previous state; 3, updating an inverse matrix R(n)-1 of the related matrix R(n) by a Woodbury identical equation; 4, detecting a hyperspectral image in real time by anomaly detection operators. In the state equation, R(n-1) refers an estimated value of the previous state, and rn refers to an observed value of the current state.

The invention discloses an image brightness compensation method for image coding, belongs to the field of image processing, and solves the problems of complexity, low speed and low compression efficiency of the conventional algorithm for performing brightness compensation for multi-viewpoint video coded images. The image brightness compensation method comprises the following steps: determining brightness difference between a current block P0 and a reference block PS by comparing the brightness difference between a current block template X0 and a reference block template XS, thereby realizing the brightness compensation for the current block; and utilizing a series of solved candidate weight matrixes B alpha, solving the optimal Bj and assigning the optimal Bj to weight matrixes B of the reference block PS, thereby solving the brightness difference between the current block template X0 and the reference block template XS according to B(X0T-XST), which serves as the brightness difference between the current block P0 and the reference block PS for brightness compensation.

The invention provides a light gas pipeline engineering structure, which can implement pipeline superspeed transportation under normal pressure state, and also can implement higher speed transportation under low vacuum or super low vacuum. The technology not only can remarkably reduce the gas-flow resistance of transporting vehicles in the pipeline so as to improve the running speed, but also can save energy consumption under the same technical condition, reduce project construction and operation management cost, and is easy for constructing because the materials of light gas have rich resources.

The invention relates to a road mark line damage detection method based on unmanned aerial vehicle shot road images. The method comprises the steps of: for a preprocessed image, detecting white mark lines on two sides of a road by adopting a Hough transformation method for detecting straight lines, then positioning the road, calculating a threshold to binarize the image, removing a repeat area ofadjacent images using an image matching method, finally, extracting the mark lines of the road, and performing damage detection on the mark lines. The road mark line damage detection method based on unmanned aerial vehicle shot road images effectively solves the problems of low efficiency, resource waste and the like of the traditional manual view method, solves the influence of road color changesunder different light on mark line extraction and the problems that the damaged mark lines are difficult to analyze and the like, is free from the limitations of time and space, and has good applicability and reliability.

A method for a multi-mode terminal to select target networks with guaranteed quality of service on the basis of ant colony optimization is used for selecting optimized target networks for each multi-mode terminal by specifics that a base station acquires value of quality of service of each multi-mode terminal in each target network according to feedback of the multi-mode terminal and then simulates the ant colony optimization for searching advancing paths during foraging of ants, and includes steps of firstly, sequencing and numbering steps of ant advancing paths and selectable paths with natural numbers respectively and setting the total number of iterative algorithm; secondly, simulating the multi-mode terminals to select the target networks according to value of quality of service of different networks by the method selecting paths on the basis of different pheromone concentration and visibility in the ant advancing paths; calculating fitness of each path in the iterative algorithm after each section of advancing paths, updating the pheromone concentration of the paths according to the fitness, judging whether reaching the set times of the iterative algorithm or not so as to stop the ant colony iterative algorithm to output the optimal target networks selected by different multi-mode terminals.

The invention discloses a twist drill abrasion monitoring method. Firstly, four parameters such as drilling force, torque, electric current of a spindle motor and electric current of a feed motor in the different abrasion states are collected under the same cutting parameter; the four parameters are subjected to wavelet packet decomposition to obtain energy spectrums of eight frequency bands; and the energy spectrums of the second frequency band, the third frequency band, the fourth frequency band and the fifth frequency band are used as characteristic values serving as conditional attributes, the three abrasion states of a drill bit of a twist drill are used as decision attributes, a decision table is established, and the characteristic values are used as input neurons of a BP neural network for training and learning. The abrasion state recognition of the twist drill is performed through the established BP neural network, and the recognition rate is high.

The invention discloses a catenary geometric parameter detection method and system, and relates to the technical field of rail transit detection. The method is based on a hand trolley on a rail, and comprises the steps of based on a light source arranged on the trolley and first and second cameras located respectively at two sides of the light source, enabling optical axes of the first and secondcameras to be coplanar with the light source to constitute a light curtain target covering an extreme position area of a contact line; randomly shooting pictures of the contact line in the light curtain target by using the first and second cameras; obtaining the position of the contact line based on the obtained pictures of the contact line and calculating position information of a catenary according to a trigonometric function formula, the position information including the height of the contact line and horizontal distances to the first and second cameras; and obtaining the height of conductor and a pull-out value of the contact line according to the calculated position information of the contact line. The comprehensiveness and efficiency of geometric parameter detection of the contact line are improved.

The invention provides a multi-target detection method based on a cascaded hourglass neural network, and aims to solve the technical problems that an existing detection method is too slow in speed anda small target is difficult to identify. The method comprises the following steps: step 1, collecting a training sample; step 2, building a deep learning framework and a backbone network cascade hourglass network of target detection; step 3, designing a label of the training sample as a confidence coefficient heat map; step 4, designing a loss function of the cascade hourglass network to optimizethe loss function; step 5, training the cascaded hourglass network to obtain a detection model; And step 6, multi-target detection. The method has the beneficial effects that various types of targetscan be quickly and accurately identified, and the small target identification capability is improved.

The invention provides a method and device for operating an Android application in a Windows system. The method comprises the steps of responding to an instruction for starting an application, and loading a dynamic link library related file of an Android virtual machine by loading a Linux executable file of a preset format, thereby starting an Android virtual machine; starting loading the dynamic link library related file of the application; loading a preset engine file and a related executable file in the application by starting the Android virtual machine; calling an Activity resource corresponding to the application by calling a preset starting process; and finishing loading the dynamic link library related file of the application in the life cycle of the Activity resource, thereby operating the application. Through application of the method and the device, the Android application can be normally operated in the Windows system according to a native application form, moreover, the Android application can be normally and rapidly operated in the Windows system without a complicated operation mode of a user, and further the use experience of the user is improved.

The invention provides an acceleration method and device for a convolutional neural network, and relates to the technical field of image processing. The method comprises the steps that a semi-stop module is arranged in the convolutional neural network; in the prediction process of the convolutional neural network, a current prediction result of the prediction process is calculated when the semi-stop module is executed; whether the current prediction result meets preset prediction conditions or not is judged; when the current prediction result meets the preset prediction conditions, the prediction process stops, the current prediction result is taken as a final prediction result of the convolutional neural network, and otherwise, the prediction process continues to be executed. According to the acceleration method and device for the convolutional neural network, the technical problems that in the prior art, the operation speed of the CNN is low, and tasks which have a high requirement on the real-time performance are difficult to execute can be solved.

An embodiment of the invention discloses an application recommendation method. The application recommendation method includes: starting test items to acquire test results of the test items according to test instructions inputted by a user; determining whether a terminal is in a normal operation state or not according to the test results of the test items; when the terminal is in an abnormal running state, finding target applications corresponding to the abnormal running state of the terminal in a preset application library; prompting found application information of the target application to the user for installing the target applications so as to relieve the abnormal running state of the terminal. The embodiment of the invention further discloses an application recommendation device and the terminal. By the embodiment, the running state of the terminal can be improved, and running speed of the terminal can be increased.

The invention provides a dual-mode masked man event automatic detection method based on video characteristic statistics. The dual-mode masked man event automatic detection method comprises the steps of: zooming a video frame and converting the video frame into a grayscale image after acquiring a source video image, acquiring a maximum object motion foreground rectangular frame, positioning a head region coarsely, acquiring face position information, then estimating a mouth region within the obtained face range, cutting the mouth region to narrow a range, calculating a gradient map, primarily judging whether a person is a masked man through calculating number of connected domains on the basis of the obtained gradient map, and calculating proportion of the areas of the top two connected domains for further judging whether the person is a masked man. The dual-mode masked man event automatic detection method based on video characteristic statistics does not have specific requirements for a video scene, is simple and efficient in the calculation method, can perform real-time processing, can be applied to real-time investigation and case solving, saves suspicious masked man information, and can provide first-hand information for evidence collection after a crime is committed.

The invention relates to a skin image affected part segmentation method comprising the steps that an input skin image is preprocessed, RGB component values are extracted, and the image is converted to an HSV color space; skin color areas and non-skin color areas are detected, and internal filling is performed in the skin color areas; pixel value histogram statistics is performed on the skin color area H component diagram obtained in 3), and a new skin color area H component diagram is obtained by using an Otsu threshold method; and binarization is performed on the skin color area H component diagram and morphological processing is performed. The skin image affected part segmentation method is suitable for segmentation of the affected part in multiple skin disease images and has the characteristics of high operation speed and high accuracy.

The invention discloses a useless item clearing method and a useless item clearing device, and belongs to the technical field of data processing. The method comprises the following steps: starting scanning program, automatically scanning useless items including running programs, useless files and at least one type of power-on start-up items; displaying scanned useless items; clearing the useless items after receiving a clearing instruction. The useless item clearing device comprises a scanning module, a displaying module and a clearing module. According to the invention, the useless items are automatically scanned by starting the scanning program, and the scanned items are displayed, as a result, a user can timely determine whether the scanned useless items are cleared or not, the scanned useless items are cleared, further the running speed of a system is accelerated, and resources consumed by the running useless items are saved.

The present invention discloses a ladder diagram matrix compiling and interpreting method based on a PLC. According to the method, the matrix operation of a ladder diagram is carried out directly without IL language conversion and interpretation. The method comprises a process compiling process and an interpreting process. According to the compiling process, since an IL language conversion process is omitted, only the simple scanning of the connection relationship of a matrix is needed, the ladder diagram compiling process is greatly simplified, fault points are reduced, and the compiling efficiency is improved. Since only the connection relation between elements in the matrix is processed in the compiling process, the complex AND OR and AND OR nesting logic are not processed, the logic operation process is simplified, the fault points are reduced, and the compiling efficiency is improved. In the interpreting process, the process of interpreting an IL language into a binary system index table is omitted, the scanning combination operation of the ladder diagram matrix is directly carried out, a lot of bottom layer CPU resources are saved, the program operating speed is improved, and the fault points are reduced.

The invention discloses a midnight-noon ebb-flow low frequency symptom. The midnight-noon ebb-flow low frequency symptom and sign differentiation therapeutic apparatus comprises a display terminal module, a control module and an output module. The invention is characterized in that an acupuncture point data database and a disease data database are stored in a storing device in the control module, a computer model is established between the two databases through time acupuncture rules of midnight-noon ebb-flow and eight methods of intelligent turtle, a user is clued to find out a therapeutic program corresponding to symptom through a software interface, and when or how to treat or keep health is clued according to the choice of the user. The invention can essentially provide an intelligent acupuncture system of treatment planning and best treatment time for the users.

The invention discloses an accelerating core virtual scratch pad memory method based on a heterogeneous multi-core platform. The accelerating core virtual scratch pad memory method includes the following steps: (1) dividing shared L2Cache into two portions of common L2Cache and a virtual scratch pad memory (SPM) logically; (2) setting a virtual SPM access interface; (3) setting a replacing strategy of the common L2Cache and the virtual SPM again; (4) addressing the virtual SPM and a memory uniformly; and (5) defining a virtual SPM space request and a released MIPS assembling command. By optimizing a storage cache sub-system of the heterogeneous multi-core platform partially, data interaction between a general processing core and an accelerating core is no longer finished through the memory and is finished by achieving data sharing in the virtual SPM. By means of the method, shared data interaction speed between the general processing core and the accelerating core is increased effectively, and the integral performance of the heterogeneous multi-core platform is obviously improved. Meanwhile, the virtual SPM can replace an SPM of the accelerating core partially, the capacity of the SPM of the accelerating core itself can be saved, and the cost of hardware is reduced.