77results about How to "Does not affect running speed" patented technology

The invention provides a method for live broadcast of a mobile terminal, and a streaming media server which supports multiple protocols and a players used for playing network video streaming on a mobile device need to be provided, wherein the player is developed based on an open source playing VLC multimedia player. The method includes: Step 1. collecting video and audio data through the mobile terminal, compressing the collected audio data and video data through soft coding mode or a hard coding mode, and after packaging the compressed audio and video data through a protocol that the mobile terminal supports, pushing to the streaming media server; Step 2. according to the type of the protocol used when a player is accessed to the streaming media server, converting audio and video streaming to the protocol used by the player to be distributed; and Step 3. decoding and playing audio and video code steam data through the player. The invention also provides a system for live broadcast of the mobile terminal, and the system supports a dual-coding mode and multi-protocol transmission of data and realizes video collection and live broadcast on the mobile terminal.

The invention discloses a method for regulating operating frequency of a CPU (Central Processing Unit). The method comprises the following steps: determining tasks in a foreground running mode; determining a task type of tasks in the foreground running mode by utilizing a mapping relation between the prestored tasks and task type; determining the operating frequency of the CPU which corresponds to the task type by utilizing a mapping relation between the prestored task type and the operating frequency of the CPU; and configuring the operating frequency of the CPU by utilizing the operating frequency of the CPU which corresponds to the task type. According to the method for regulating operating frequency of the CPU, disclosed by the invention, the operating frequency of the CPU is configured according to the current task interacting with a user. Therefore, the influence of a background running task on the operating frequency of the CPU can be reduced, so that the operating frequency of the CPU is further reduced, the power consumption of the CPU is reduced, and the operating speed of the task interacting with the user is not influenced. Therefore, the user experience can still be guaranteed. The invention also discloses a device and a mobile terminal for regulating the operating frequency of the CPU.

The invention provides a software transfer installation method and a software transfer installation system for a mobile terminal. The method comprises the following steps of: detecting available capacity of a storage partition under each letter in the mobile terminal and installed software in each partition; according to the detection result, generating a transfer prompt corresponding to transferable installation software, wherein the transfer prompt comprises the prompt of a transferable target partition and a catalogue to a user; receiving the transferable installation software, the target partition thereof and the catalogue thereof which are selected by the user according to the transfer prompt; searching transferable information of the software; and transferring and installing a software file to the target partition and the catalogue according to the transferable information. Due to the application of the method and the system, the storage space of the installed software can be managed conveniently; the installed software in a mobile phone can be reserved; and the operating speed of the mobile phone is not influenced.

The invention relates to the field of integrated circuits and aims to provide an IP hard core intellectual property protection method and device based on path delay, wherein the IP hard core intellectual property protection method and device can be used for intellectual property protection of a timing sequence logic which can be applied to input and output and needs multiple delay periods. The IP hard core intellectual property protection method and device can effectively prevent various IP core theft problems, such as a brute force attack, a cleaning attack and reverse engineering. According to the technical scheme, the IP hard core intellectual property protection method based on the path delay comprises the following steps that 1, signals need to pass through a timing sequence logic path circuit of multiple time periods from input to output; 2, a protective path is designed; 3, a designed protective circuit comprises a shifting register, a finite state machine circuit and an interconnecting line, wherein the shifting register is used for storing an initialization secret key, the finite state machine circuit is used for secrete key checking, and the signal delay of the interconnecting line is adjustable; 4, the purpose of protecting a circuit IP is achieved by adjusting and controlling the input and output delay of the protective circuit and the protected circuit. The IP hard core intellectual property protection method and device based on the path delay are mainly applied to integrated circuits.

The invention discloses a plug-in calling method and a plug-in manager for a browser. The plug-in manager comprises an input/output device which comprises a first input/output operation module which is suitable for judgment on whether a plug-in unit gets the permission to execute an input/output operation according to an operation permission list for maintenance of the input/output device; if the plug-in unit is judged to get the permission to execute the input/output operation, the input/output operation is executed; if the plug-in unit is judged to not get the permission to execute the input/output operation, the input/output operation is prohibited; and the plug-in manager intercepts behaviors related to the input/output operation in the called plug-in unit, and conducts the behaviors to the input/output device. The plug-in calling method and the plug-in manager can prevent potential safety hazards caused by the plug-in unit, and cannot affect the running speed of other functions of the plug-in unit, thereby improving the executing efficiency of the plug-in unit.

The invention discloses a control system of an integrated circuit tendon-cutting device. The control system comprises a main controller and a vision detection device, wherein the main controller is provided with a control output end connected with a feeding device, a control output end connected with a driving mechanism of the tendon-cutting device and a control output end connected with a material receiving device. The invention has the advantages of simple control program and safe and reliable control, carries out vision detection on a tendon tablet to be cut so as to ensure that the direction and the model of the tendon tablet to be cut, which is fed into the tendon-cutting device, are correct and protect a tendon-cutting die, and carries out detection on the tablet subjected to tendon cutting so as to reflect the condition of the tablet subjected to the tendon cutting in time and reduce the emergence of an unfavorable tablet subjected to the tendon cutting.

The invention relates to a method for operating a Web page in a flash memory of an embedding system application device. In the method, the non-compressed Web page is directly stored in the flash memory; the application device mounts the Web page to the embedding operation system; the operation system operates a Web server, and appoints the Web page catalogue as the Web page mounting catalogue; the users access the Web page by a browser. By adopting the method for operating the Web page in the flash memory of the embedding system application device, the method is different from the prior art that the Web page file is firstly compressed and stored in the flash memory, and then decompressed in the memory, and then the Web page is operated in the memory; the method directly stores the non-compressed Web page file in the left space of the flash memory so as to fully use the left flash memory space while the flash memory is operated and the operation speed of the embedding system is not affected; the method saves the space of the memory.

The invention relates to the technical field of optical character recognition in computer vision, and provides a character recognition method for enhancing an attention mechanism by fusing multilayer features. The method comprises the following steps: selecting a training picture; extracting picture features; constructing a feature fusion matrix and fusing multilayer features; carrying out feature fusion by using the associated features to enhance the feature expression ability; carrying out sequence modeling on the fused features; performing probability prediction on the features after sequence modeling; in the training stage, adopting back propagation to update the parameter weight of a network model, and obtaining a standard network model capable of being used for character recognition; in a test stage, inputting a to-be-recognized picture into the trained network model, recognizing the model, and outputting characters in the picture. According to the method, the features extracted from each level of the neural network are mutually mapped, so that the expression ability of the features is improved, and the accuracy of character recognition is improved.

The invention discloses a method of for calling a plug-in board by a browser, a plug-in board manager and a browser. The method for calling the plug-in board by the browser comprises the steps of leading actions of I (Input)/O (Output) operation in the plug-in board into an I/O device when the plug-in board is called by the browser, and maintaining an operating authority list by the I/O device; judging whether the plug-in board has the authority of executing the I/O operation or not according to the operating authority list by the I/O device, executing the I/O operation by the I/O device if the plug-in board has the authority of executing the I/O operation, and inhibiting the implementation of the I/O operation if the plug-in board does not have the authority of executing the I/O operation. According to the method for calling the plug-in board by the browser, the I/O device is utilized for controlling the plug-in board to conduct the I/O operation on the local resources, and preventing potential safety hazards brought by the plug-in board; and moreover, the low-speed I/O operation can be separated from the other threads to be executed by the I/O device, so that the operating speed of the other functions of the plug-in board can not be influenced, and the execution efficiency of the plug-in board is enhanced.

The invention discloses a method for managing terminal recording information, which comprises the steps of: memorizing the recording information with preset quantity to be a plurality of files and recording files number by a terminal, marking the time range of the memorized files according to memory time, and sequentially arranging the files according to time sequence; recording files memorizing status information; and building the array of nonius of the recording information in the files for marking initial recording numbers which are memorized in the files at each time and recording numbers to be memorized at the next time. By building file blocks, each set file block is used for memorizing a certain quantity of the recording information, and the recording information among each file and in each file are sequenced by time, therefore, the method not only can make full use of the limited memory space of a terminal module and has small internal memory, but also is convenient for users to fast research and count the recording information which is memorized by the terminal, and can not influence the running speed of the terminal.

The present invention relates to a method for analog-digital conversion and an analog-digital converter. The method is applied to an SAR ADC comprising a capacitor array and a comparator, and the capacitor array comprises n capacitors which are connected in parallel. The method comprises the steps of: employing a successive approximation method to determine an error array corresponding to an offset voltage of the SAR ADC; according to the error array, determining a first reference voltage V'j corresponding to a jth capacitor in the n capacitors; and after a sampling phase of analog signals tobe converted being subjected to sampling, and employing the successive approximation method to convert the analog signals to be converted to target digital signals, wherein in the sampling phase, thefirst end of each capacitor of the n capacitor is connected with the analog signals to be converted, and the second end of the jth capacitor is connected with the first reference voltage V'j. The method for analog-digital conversion and the analog-digital converter achieve self-elimination of the offset voltage without adding extra circuits.

The invention provides a convolutional neural network processing method and device and an electronic system. The method comprises the following steps: training a first convolutional neural network, and initializing parameters of a second convolutional neural network based on the trained parameters of the first convolutional neural network; training the second convolutional neural network; and performing structure conversion on the trained second convolutional neural network to enable the structure of the converted second convolutional neural network to be the same as the structure of the firstconvolutional neural network. In the mode, the convolution kernel of the first convolution neural network is expanded to obtain the second convolution neural network, so that the expression capability of the network in the network training process can be improved; the parameters of the second convolutional neural network are initialized based on the trained parameters of the first convolutional neural network, so that the training time of the network can be reduced; and the structure of the second convolutional neural network is converted into the same structure of the first convolutional neural network, so that the network operation speed is not influenced while the network precision is improved.

The invention provides a novel electric car which can be charged while running, and the electric car adopts a novel electric machine which combines a high-speed middle motor with a transmission device with gears. In the novel electric car, four generators (12V) are arranged on four wheel hubs so as to generate voltages for charging four storage batteries (12V 100A) while the novel electric car runs.

The invention provides an application freezing method and apparatus, and a terminal. The terminal is provided with at least two screens. The application freezing method comprises the following steps: receiving an application freezing instruction on a first screen; hiding an icon of an application to be frozen at the first screen according to the application freezing instruction, and displaying the icon of the application to be frozen on a second screen, wherein the first screen and the second screen are two different screens in the at least two screens. By adopting the scheme, the situation of mistakenly unfreezing the application can be effectively prevented, the occupation of a terminal memory is avoided, the operation speed of the terminal is not affected, the user requirement can be better met, and the user experience can be improved.

The invention discloses a method for storing history, including the following steps: the current operation is set as latest history and the latest history is stored in a first storage medium; and the latest history is stored in a second storage medium. By the invention, every operation of a user on software can be stored as history without influencing the operating speed of system memory, the user can call the history of every operation, and the history can be protected and restored when the system is suddenly halted.

The invention discloses an auto-calling dialing method of a communication terminal, comprising the steps of: according to prompts of situations which possibly appear during dialing, collecting data of a counter regularly, judging the types of dialing signals, and outputting required pulse number through a decoder; comparing the frequency number with a preset value; calling after getting through, wherein the pulse signal is stabilized, and automatic dialing logic is stopped; if a signal is not connected, changing the pulse signal before and after collection, starting an auto-calling dialing program, and entering into a user acknowledgement state. The auto-calling dialing is circulated until the connection is made. According to the auto-calling dialing method of the communication terminal, the trouble of repeating dialing is reduced, the inconvenience caused by forgetting of calling is effectively avoided, and unnecessary trouble is prevented from being brought to the called party.

The invention belongs to the technical field of software modularization and particularly relates to a map module software organization method of a radar terminal. The method comprises: a radar terminal and a map loading module are run on the same machine; data can be transmitted between the radar terminal and the map loading module through a local loopback address; the map loading module is in a message waiting state all the time until a map drawing request of the radar terminal is received; when the radar terminal is started or the map state of the radar terminal is changed, map parameters are transmitted through the local loopback address to the map loading module by the radar terminal; and after receiving the map parameters, the map loading module processes the map parameters and generates a corresponding map as a feedback to the radar terminal. According to the invention, a map loading function integrated in the radar terminal can be separated with hardly any effect on the overall operating speed posed of the radar terminal.