55results about How to "Save running memory" patented technology

The invention discloses a method for intelligently optimizing internal storage of a mobile terminal and the mobile terminal, and belongs to the field of mobile terminals. The method for intelligently optimizing the internal storage of the mobile terminal comprises the steps that when a preset application program is started, the frame rate of a frame currently displayed by the mobile terminal is detected, progresses of application programs without tags set in the background are cleared away according to the frame rate and a preset frame rate threshold value to save the running internal storage of the system, the use rate of the internal storage of the system is detected, progresses of application programs with set tags are cleared away or controlled to be restarted according to the use rate and a preset internal storage use rate threshold value, the started preset application programs have the enough running internal storage, and under the conditions of not changing hardware, the running speed of the system is improved, and user experience is improved.

The invention belongs to the technical field of unmanned ship navigation control, and specifically relates to anunmanned ship track planning method based on deep optimization. The unmanned ship trackplanning method based on deep optimization comprises the following steps that rasterized map information is read, starting and ending points are set, the obtained starting point and reachable neighbornodes in the eight directions of the starting pointare added to theopenlist; the point with the smallest F value in the openlistis searched for and set as the current node; whether the current node is the ending point or not is determined; if yes, then path searching is over; if not, then the current node is utilized to search for a next jump-point based on a jump-point search strategy; whether the jump-point is in the openlistor not is determined, and the point with the smallest F value is continuously searched for; and an optimal path is outputted. According to the unmanned ship track planning method based on deep optimization,the execution efficiency of grid method track planning algorithm is greatly improved based on the jump-point search optimization preprocessing algorithm,an path turning point is optimized, and finally a smooth navigation track is outputted.

An embodiment of the invention provides a memory optimization method and device. The method comprises the steps of determining an image object with color data to be released by an application if it is switched to a background application; storing the color data of the image object; releasing the memory occupied by the color data of the image object. By means of the memory optimization method and device, the operation memory of a terminal device can be optimized at a lower hardware cost.

Embodiments of the invention disclose an application startup control method and apparatus, a storage medium and a mobile terminal. The method comprises the steps of determining a to-be-processed target application, and obtaining startup feature information of the target application; obtaining a preset machine learning model, wherein the preset machine learning model is obtained by training application samples of multiple known startup types and is used for classifying applications based on the startup feature information; inputting the startup feature information to the preset machine learningmodel, and obtaining an output result of preset machine learning; and according to the output result, determining a startup permission of the target application. According to the technical scheme provided by the application startup control method and apparatus, the startup permissions of the applications are reasonably controlled according to startup features of the applications based on the machine learning model, so that the resource and flow consumption of the mobile terminal caused by self-startup or associated startup of a few abnormally started applications can be effectively prevented.

The utility model relates to a method for displaying characters, which belongs to the technical field of computers, and comprises the steps as follows: lattice information of characters to be displayed is obtained according to a typeface file used for keeping lattice information of characters in a program and a whole character table keeping the position information of the characters in the typeface file; and the characters are displayed according to the lattice information of the characters to be displayed. Besides, the invention also relates to a system for displaying characters according tothe method. In the method and the system, a program developer can use self-defined typefaces through using the built-in typeface file, and the flexibility on the aspect of using different styles of typefaces is improved. Moreover, compared with the characters displayed through pictures, displaying the characters by using the lattice information of the characters can greatly reduce a running memory, and can display the unknown typefaces in advance in real time according to the typeface file.

The invention provides a data classification method and system based on a convolutional neural network, a medium and equipment. The data classification method comprises the steps: carrying out the preprocessing of obtained classification data, and constructing a data set; constructing a convolutional neural network, wherein the convolutional neural network at least comprises a convolutional layerused for extracting local features, and the convolutional layer compresses a feature matrix, and sparse matrix vector multiplication operation is performed on the generated sparse matrix on a graphicprocessing unit, and data in a data set is utilized to train the convolutional neural network; and preprocessing the to-be-classified data, inputting the preprocessed to-be-classified data into the trained convolutional neural network model, and outputting a data classification result. According to the data classification method, the feature matrix of the convolutional layer is compressed, and parallel computation is performed on the GPU, so that the memory consumption and zero value computation in the computation process are reduced, and the training time and memory consumption of the neuralnetwork are reduced.

The invention discloses an installation method for a preinstalled application of a mobile terminal. The installation method comprises the steps as follows: detecting model information of the mobile terminal; determining type information of the preinstalled application corresponding to the model information; if the type of the preinstalled application is the type of installation, the corresponding full-version application is installed to the mobile terminal; if the type of the preinstalled application is the installable type, the corresponding preview application is installed to the mobile terminal. It can be seen that in this scheme, the type of the preinstalled application is determined according to the type of the mobile terminal. If the preinstalled application type is the installable type, the preview application is installed to the corresponding mobile terminal. The preview application can not only realize main functions of the full-version application, but also reduce the occupied running memory and storage capacity compared to the full-version application, accordingly, the operating speed of the mobile terminal is increased, and the system performance is improved; the invention further discloses an installation device of the preinstalled application of the mobile terminal, and the installation device can also realize the technical scheme.

The invention discloses a pipeline defect intelligent detection method based on image processing and deep learning, and application thereof. The intelligent detection method comprises the following steps: carrying out preprocessing including edge detection on pipeline image data, inputting the preprocessed image into a trained detection model, and carrying out defect detection, wherein the detection model is constructed based on a Darknet53 backbone network and a multi-scale feature extraction network. According to the detection method, different defects of the pipeline can be accurately recognized and detected, the model meeting the industrial application level is obtained, the detection efficiency is high, and the environment universality is high.

The invention discloses an information technology equipment safety detection system and a detection method thereof, which relate to the field of equipment safety detection. In order to formulate a more effective detection scheme according to the operation condition and the information source of the equipment, the detection system specifically comprises an information acquisition module, an information analysis module, an information processing module, a safety detection function module and a data storage module, wherein the information analysis module is respectively connected with the information acquisition module and the information processing module; the safety detection function module is connected with the information processing module and the data storage module. The detection method specifically comprises the following steps that the information acquisition module monitors the operation of the information technology equipment in real time, acquires operation information and sends the operation information to the information analysis module. Information technology equipment can be monitored in real time, and meanwhile, the risk information is analyzed, so that the detectionwork is carried out in a targeted manner, the system operation memory is saved, and the safety of the operation environment is guaranteed.

The invention discloses a method for automatically configuring and generating an OTT platform EPG interface. The method comprises the following steps: a client initiates an http data request, a serverresponds to the http data request, and a background configures data necessary for the EPG interface, wherein the data are configured according to layers, after receiving the data, the client parses the data layer by layer and fills the data layer by layer to generate a view, recovery is performed by a memory in a data parsing and view generation process to perform optimized operation, and the client comprises a background modeling module, a data parsing and filling module and a memory recycling module. According to the method, the most complex function is achieved in the simplest way, and better and more experience is brought to users by dynamic configuration. The immature fixed EPG homepage is discarded, and we hope to achieve a very high level of realization, simplicity and simplicity without losing the technical level.

Provided is a device for detecting the curved contour of a section of the planar bent sheet metal part. The device comprises a mechanical device and a control device. The control device comprises a control box, a PLC, a single-chip microcomputer and a computer. The mechanical device comprises a frame, an X-direction translation module, a Y-direction translation module and a linear array CCD camera. A detecting method adopting the detecting device mainly comprises the steps the contour position information collected by the linear array CCD camera and Y-module coder data of the mechanical deviceare combined and serve as a Y coordinate value of a metal plate contour two-dimensional diagram after data processing, position data fed back by a X-module servo motor coder in the mechanical deviceis adopted as an X coordinate value of the metal plate contour two-dimensional diagram, metal plate contour circulating scanning and contour point data collection are completed through control of a driving program, and a two-dimensional point cloud file related to the curved contour of the sheet metal part is obtained after upper computer data processing is carried out. The device is automatic, efficient and cheap and can extract contour data fast and conveniently.

The embodiment of the invention provides an interface implementation method and device and a set-top box. The method comprises the steps that a plugin calling request for calling a browser plugin is received; a target interface needing to be executed is determined according to the plugin calling request; the target interface is executed, and an execution result is returned; and the execution result is taken as a plugin calling result and displayed. According to the novel interface implementation mode, a JS interface is taken as the browser plugin, and a corresponding JS interface function is achieved by calling the browser plugin; and compared with the prior art, the method has the advantages that great progress is achieved, an obvious effect is achieved, the development workload of the JSexpansion interface and the memory occupied during browser operation are reduced, the working efficiency of developers is improved, the browser operation speed is increased, the problem that an existing browser JS expansion interface needs to recompile a browser code is solved, and the use experience of a user is enhanced.

The invention relates to a picture generation method based on picture editing of a mobile operating system. Editing and generation of a high-definition picture of a mobile terminal are achieved through foreground and background separated rendering. The method comprises the following steps: firstly, loading a preview obtained by reducing an original high-definition image, and submitting the previewto an editing view for rendering; the user edits the rendered preview to generate edit element information, and the edit elements are rendered according to the equal-ratio reduced size, so that the effect of saving the running memory during editing is achieved; when the picture is generated, the original picture of the high-definition picture is rendered at the background, and then rendering is performed according to the original size of the editing element so as to generate the high-definition picture.

The invention discloses an error correction method used for a NAND-FLASH-adopted electric power acquisition terminal, which comprises the following steps: determining to adopt a BCH (Bose, Ray-Chaudhuri and Hocquenghem) CODE algorithm to realize data error correction; in the BCH CODE algorithm, determining that a minimum error correction unit fixedly has 512 bytes and 4 error correction bits; according to a rule that the minimum error correction unit fixedly has 512 bytes and 4 error correction bits, generating an index table used for error correction bit flipping; storing the index table in a memory area ROM (Read Only Memory) of codes required by the electric power acquisition terminal; carrying out a data reading operation on Nand Flash by an application program on the electric power acquisition terminal, and utilizing the index table stored in the memory area in the step 4) by the CODE algorithm to realize data correction through a BCH algorithm if 1 or 2 or 3 or 4-bit bit flipping is in the presence. The error correction method has the advantages that a backup power supply can be called to realize a purpose that a lamp of a bathroom can normally illuminate after outage.

The invention discloses a method and a terminal for achieving displaying cache. The method for achieving displaying cache includes storing a to-be-displayed text inside an internal storage as a component object, and creating a caching picture object which is of the same side with the text component for the text component when the stored text component is displayed on a screen. As is seen from the method of screen caching, a caching picture just need to be created in an area with characters, and thus the internal storage occupied by a non-character area is saved. Therefore, the size and the occupied internal storage of the caching picture are decreased. Through the method for achieving displaying cache, on the premise that the repaid display effect is not affected, running cache is saved for a program, and cache requirements of product features are met.

The invention discloses a voiceprint-based 10 kV circuit breaker mechanical fault detection method. Comprising the following steps: S1, building a MobileNetYOLOv4 neural network model; s2, acquiring voiceprints through a high-sensitivity sensor; s3, performing convolution operation on a convolution kernel of edge side identification early warning; s4, circuit breaker fault identification; s5, carrying out improved S transformation on the signal; according to the invention, task allocation is dynamically calculated by a resource allocation module embedded in a cloud platform, a lightweight model is constructed, the model operation speed is accelerated, and a real-time target identification and classification system is constructed; the Moblienet neural network structure reduces the running memory of a neural network model and accelerates the running speed of the neural network model, and in order to improve the target recognition speed and solidify the model to an edge device, the MobileNet network structure applies a deep separable convolution technology, so that the target recognition efficiency is improved. The standard convolution of the convolutional neural network is divided into two steps of deep convolution and point-by-point convolution in the calculation process, so that the calculation amount of the network is greatly reduced.

The invention provides a grid data distributed storage service method, device and equipment and a medium, and the method comprises the steps: based on a parallel subdivision principle, storing grid data in a memory of each process in a dispersed manner, and constructing distributed storage of a grid in a group of processes; according to the source parallel subdivision in the existing distributed storage on the grid, redistributing the grid data, and constructing the distributed storage of the target parallel subdivision corresponding to the source parallel subdivision on the grid; constructing a distributed storage service model according to the distributed storage of the grid in a group of processes and the distributed storage of the target parallel subdivision; according to the distributed storage service model, executing a service function; the operation memory of the grid data in one process can be effectively reduced, the computing processing capacity of a computer is improved, and the performance bottleneck of storing large-scale grids in development and application of numerical programs such as a coupling mode is solved.

The invention discloses an image display method, device and equipment and a storage medium. The method comprises: obtaining operation information on a screen; obtaining a corresponding image accordingto the image display area on the screen after operation; and if the resolution of the image is greater than the resolution threshold corresponding to the image display area, displaying the reduced image in the image display area, the resolution of the reduced image being less than or equal to the resolution threshold corresponding to the image display area. According to the technical scheme provided by the embodiment of the invention, the resolution threshold corresponding to the image display area is determined, the image with the resolution greater than the resolution threshold is reduced,and then the reduced image is displayed in the image display area, so the resolution of the reduced image meets the requirement of the resolution threshold value, pixel information contained in the displayed image in the operation process is reduced, the operation memory occupied by image display is reduced, and the image display efficiency and the operation performance of the application programare improved.

The invention belongs to the technical field of charging, and particularly relates to a Bluetooth connection-based charging method and system and a storage medium. The method comprises the steps that a Bluetooth module in a charging pile is converted from a closed state to an open state when the charging pile detects the access of a to-be-charged device, and a broadcast notification with a unique identifier of the charging pile is transmitted; a mobile terminal obtains the unique identifier of the charging pile in the broadcast notification, and establishes a Bluetooth connection with the charging pile through the unique identifier of the charging pile; the mobile terminal uploads the obtained unique identifier of the charging pile to a server, so that the server sends a charging instruction to the charging pile according to the unique identifier of the charging pile; and a charging pile charges to-be-charged equipment according to the received charging instruction, and converts a Bluetooth module of the charging pile into a closed state from an open state. The mobile terminal identifies the charging pile in a Bluetooth communication mode, the occupied running memory of the mobile terminal is smaller, and operation is smooth; and the Bluetooth module is built-in and is not easy to damage.

The invention relates to the field of security fence equipment, in particular to a defense organizing system and method for the periphery of a living area, the system comprises a plurality of image monitoring modules, a processing module, an alarm module and a plurality of supplementary detection modules, the method comprises the following steps: S1, identifying abnormal information in image information of the living area and giving an alarm, and S2, acquiring multiple pieces of image information and comparing to judge whether an abnormal target of the abnormal information moves or not; S3, when the abnormal target moves, detecting and tracking the abnormal target, and when a tracking signal is not received, sending a defense withdrawing signal to stop the detection and tracking of the abnormal target. Detection and tracking are carried out only when the system is abnormal, the running memory of the system is saved, and the energy consumption of the system is reduced.

The invention relates to a game data management method and device, a computing device and a storage medium. The method includes the steps of obtaining the current display data and current memory resources of a current running game interface in a game, determining display data unrelated to the current running game interface according to the current display data, determining memory resources unrelated to the current running game interface according to the current memory resources, and closing the unrelated display data and unrelated memory resources in memory. Through the display data and memoryresources corresponding to the current game interface, the data related to the current running game interface and the data unrelated to the current running game interface are determined, the unrelated data is closed, the running memory is reduced, and the running speed is increased.

The utility model relates to a method for displaying characters, which belongs to the technical field of computers, and comprises the steps as follows: lattice information of characters to be displayed is obtained according to a typeface file used for keeping lattice information of characters in a program and a whole character table keeping the position information of the characters in the typeface file; and the characters are displayed according to the lattice information of the characters to be displayed. Besides, the invention also relates to a system for displaying characters according to the method. In the method and the system, a program developer can use self-defined typefaces through using the built-in typeface file, and the flexibility on the aspect of using different styles of typefaces is improved. Moreover, compared with the characters displayed through pictures, displaying the characters by using the lattice information of the characters can greatly reduce a running memory, and can display the unknown typefaces in advance in real time according to the typeface file.

The present invention discloses a method, device and system for quickly updating the correction coefficient of an LED display screen, which belongs to the field of LED display screen correction. The correction coefficient of the entire LED display screen is obtained, compressed and stored in a data frame according to the data format of the region; the regional offset data of the selected region in the LED display screen is obtained, the region where the coefficient needs to be replaced is calculated in the data frame and the corresponding compressed correction coefficient is obtained, which is decompressed and sent to the receiving card corresponding to the selected region to realize the rapid update of the correction coefficient in the region. The present invention divides the correction coefficient on the data frame into blocks for compressed storage. When the correction coefficient of the selected region needs to be updated, the region where the coefficient needs to be replaced in the data frame is calculated according to the regional offset data of the selected region, and the compressed correction coefficient in the region is obtained and decompressed. Compared with the traditional correction coefficient overall update method, the correction coefficient replacement rate is improved, thereby improving the display rate.

Provided is a device for detecting the curved contour of a section of the planar bent sheet metal part. The device comprises a mechanical device and a control device. The control device comprises a control box, a PLC, a single-chip microcomputer and a computer. The mechanical device comprises a frame, an X-direction translation module, a Y-direction translation module and a linear array CCD camera. A detecting method adopting the detecting device mainly comprises the steps the contour position information collected by the linear array CCD camera and Y-module coder data of the mechanical deviceare combined and serve as a Y coordinate value of a metal plate contour two-dimensional diagram after data processing, position data fed back by a X-module servo motor coder in the mechanical deviceis adopted as an X coordinate value of the metal plate contour two-dimensional diagram, metal plate contour circulating scanning and contour point data collection are completed through control of a driving program, and a two-dimensional point cloud file related to the curved contour of the sheet metal part is obtained after upper computer data processing is carried out. The device is automatic, efficient and cheap and can extract contour data fast and conveniently.

The invention discloses a method for controlling a default NFC (Near Field Communication) card, which comprises a mobile device, the mobile device is provided with a gravity sensor, and the mobile device has and starts an NFC function, and comprises the following steps: default setting: starting an APP (Application) supporting an NFC simulation card, and setting a default card in different gravity directions; the reading setting is as follows: the mobile equipment is close to equipment with an NFC (Near Field Communication) reader; state judgment: the APP receives the NFC reading request and obtains the current position and gravity state of the mobile equipment; and judging a default card: reading the set default card according to the current position and the gravity state of the mobile equipment. The invention provides a method for controlling a default NFC (Near Field Communication) card, which is characterized in that the NFC card is preset according to the use frequency and the size required by use of each NPC (Near Personal Computer) card, and the following specific operations are performed: a preset APP (Application) is entered, and four areas are sensed by a gravity sensor: the area is vertically downward and reverse when a mobile phone is in a use state, and the area is horizontally laterally placed and directional when a power supply side key is on the upper side.

The invention relates to the technical field of wind turbine generator set state monitoring, in particular to a wind turbine generator set state monitoring data acquisition method and query method andelectronic equipment. The acquisition method comprises the steps: receiving a query request of wind turbine generator set state monitoring data; calling a data interface of a data server; acquiring an acquisition path list of the data server through a data interface; obtaining a query range in the query request, and querying wind turbine generator set state monitoring data in the query range fromthe data server by utilizing the acquisition path list; and sending the wind turbine generator set state monitoring data to a master station server for a user to check. According to the acquisition method provided by the invention, the data interface of the data server is called, the wind turbine generator state set monitoring data in the query range is queried from the data server and sent to the master station server by utilizing the acquisition path list of the data server, and the master station server does not need to install software corresponding to each station server, so that the user can check the data conveniently.

The invention discloses a method for mining a frequent item set based on an effective weighted tree, comprising the following steps: S1, reading a database D from a network, the database D comprisingN transactions, each transaction comprising different items, the number of items and the weight w occupied by each item; S2, calculating a transaction weight tw of each transaction in the database D,and generating a transaction weight table; S3, calculating the weight W of the item set S, and presetting a threshold value minws, wherein if W is greater than or equal to minws, sthe item set S is afrequent item set; if W < minws, the item set S is a non-frequent item set; and S4, constructing an effective weighted tree model for mining the frequent item set. According to the method, weights aregiven to the item sets, weight calculation is carried out, so that an effective weighted tree model is constructed, the mining efficiency of the frequent item sets is improved, memory application isreduced, and the method is suitable for a database of big data.

The invention discloses a cloud-controlled automatic driving system based on SUMO and a unified timeline, including a driving client, an application server, a vehicle control server, an intelligent electric vehicle and a sensing device; the automatic driving system uses the driving client to issue travel instructions , use the application server to parse and encapsulate the travel instructions, use the SUMO module in the vehicle control server for path planning and driving task division, and use the GPS and correction module to generate vehicle correction instructions; the vehicle uses the driving task instructions and vehicle correction instructions of the vehicle control server travel. In the system of the present invention, the application server, vehicle control server, and smart electric vehicle use a unified time axis, and the vehicle executes at the specified time according to the time stamp in the instruction. The invention realizes cloud-controlled vehicle automatic driving while avoiding network delay and instability problems. It uses SUMO and unified timeline to further simplify the system and improve system reliability.