34results about How to "Low processing efficiency" patented technology

A transmission apparatus for use in fixed bandwidth communications and variable bandwidth communications. The transmission apparatus has a memory unit, which stores information contained in a frame and indicating the amount of data to be transmitted from a terminal device, a computing unit, which calculates a bandwidth amount to be assigned to the terminal device based on the data amount, and a transmission unit, which sends the calculated assigned bandwidth amount to the terminal device. The transmission apparatus is characterized in that the computing unit calculates bandwidth amounts to be assigned to different types of flow of which different transmission qualities are requested according to a priority level that is set to each flow type. The transmission apparatus is also characterized in that bandwidth is controlled by designating, instead of a necessary bandwidth amount, transmission starting timing and transmission ending timing, particularly for fixed bandwidth communication data.

To provide a process for coding the moving picture with higher speed and enhanced efficiency. The present invention provides a moving picture coding method for coding a moving picture of processing object through a predetermined process including a motion estimating process, determining whether or not the predetermined process (e.g., DCT process and quantization process) for a block of processing object can be omitted on the basis of the information regarding a difference between the block and a reference block of a frame to be referenced in coding (i.e., a frame to be referenced in making the motion vector detection), and a quantization parameter for use in a quantization process of the moving picture, performing the predetermined process if it is determined that the predetermined process can not be omitted, and omitting the predetermined process if it is determined that the predetermined process can be omitted and making a set defined value the result of the predetermined process for the block.

A data updating request and a data processing request may be synchronously received. The data updating request may replace to-be-deleted data with to-be-written data, while the data processing request may operate using the to-be-deleted data. The solutions described throughout the present document facilitate execution of the two conflicting requests in parallel, substantially simultaneously, and synchronously, for example on respective threads. To facilitate the execution of the two requests, the to-be-written data may be stored to a first storage space, the to-be-deleted data may be stored to a second storage space, and the to-be-deleted data in the second storage space may be released if execution of the data processing request that meets a releasing condition is completed. The respective threads may synchronize with each other by postponing the release of memory. Thus, the two requests may execute synchronously and substantially simultaneously and consequently improving processing efficiency without occupying significant memory space.

This application discloses a pooling processing method, applied to a pooling processing system of a convolutional neural network. The pooling processing system includes a first storage device, a data region, a pooling computation kernel, and a pooling controller. The method includes: reading, by the pooling controller, k pieces of feature data from the first storage device in each reading cycle, the k pieces of feature data being components in a feature map generated by a convolution operation of the convolutional neural network, and k being an integer greater than 1; writing, by the pooling controller, the k pieces of feature data read from the first storage device into the data region, wherein the k pieces of feature data form one group among n groups of k pieces of data with each group arranged in a first dimension and the n groups arranged in a second dimension, wherein the n groups of k pieces of data are written into the data region in an updating cycle, wherein a duration of the updating cycle is n times a duration of the reading cycle, and wherein n cis an integer greater than 1; and transmitting, after the updating cycle is ended, data in the data region to the pooling computation kernel to perform a pooling operation, wherein the data in the data region comprises the n groups of k pieces of data and last m groups of data from a previous updating cycle with each group along the second dimension, wherein the last m groups of data are temporarily stored in the data region for use in pooling calculation by the pooling computation kernel in a next updating cycle. The technical solution in this application reduces the number of storage, numbers of reading and writing due to data reuses, and improves the efficiency of pooling processing.

An antecedent determining method is performed by at least one processor, and includes obtaining statement information, and determining whether a pronoun exists in the statement information. The method further includes extracting candidate antecedents and lexical features of the candidate antecedents, from the statement information, in response to the pronoun being found in the statement information, and determining, based on the lexical features of the candidate antecedents, a target antecedent that is referred to by the pronoun, among the candidate antecedents.

The present invention discloses a reverse ultrasonic assisted gas-liquid-solid three-phase abrasive particle flow polishing processing method and device, an ultrasonic generator is mounted in a position which is reverse in direction to abrasive particle flow, an included angle is formed between the abrasive particle flow direction and the position, a reverse ultrasonic is produced, the direction of the reverse ultrasonic is reverse to that of the abrasive particle flow, an included angle is formed between the abrasive particle flow direction and the reverse ultrasonic, the reverse ultrasonic can compress micro-sized bubbles near workpiece machining surface in an abrasive particle flow medium containing the micro-sized bubbles to collapse and perish the micro-sized bubbles, when the micro-sized bubbles are collapsed and perished, a reverse micro shock wave is produced, the direction of the reverse micro shock wave is reverse to that of the abrasive particle flow, the reverse micro shock wave allows the inner local area of the abrasive particle flow to produce a great pressure so as to produce microscopic explosion effect; two component forces are generated by the microscopic explosion effect, a vertical component force is used for improving the contact probability and acting force of abrasive particles and workpiece machining surface convex peaks, a reverse horizontal component force is used for forming relative block effect on the abrasive particles, the abrasive particle and workpiece action time is increased, and the abrasive particle flow polishing efficiency is greatly improved.

An injection mold (10) comprises a female mold (20), a male mold (40) matching with the female mold (20), and a cutting portion (48) is positioned on either the female mold (20) or the male mold (40) positioning, a receiving slot (28) is defined in another said mold for accommodating the cutting portion (48).

The present disclosure discloses an image processing method of a mobile terminal, and a mobile terminal. To process an image, an audio signal is collected using a sensor of the mobile terminal. Air-blowing signal intensity is acquired by performing a windowed fast Fourier transform (FFT) on the audio signal to obtain a frequency domain signal of the audio signal. The frequency domain signal of the audio signal is divided into at least one time-slice frequency domain signal. It is then identified whether the audio signal is an air-blowing signal according to the at least one time-slice frequency domain signal and a preset property of the air-blowing signal. The air-blowing signal intensity is acquired according to the at least one time-slice frequency domain signal when the audio signal is identified as an air-blowing signal. An image is acquired according to the air-blowing signal intensity. In this manner, the image processing method using the air-blowing signal intensity may provide improved processing efficiency and simplified operation.

A coil feeding device and a graphic machine thereof are provided. The coil feeding device includes a roller mechanism for carrying a coiled material, a base mechanism, and a compression rod mechanism. The roller mechanism includes a shaft penetrating through the shaft hole and a tension structure disposed on the shaft. The tension structure includes a first sleeve, a second sleeve and a tapered fixing member. When the first sleeve and the second sleeve are fixed to each other, the tapered fixing member holds the shaft to achieve mutual fixation of the coiled material and the shaft. By providing the second sleeve, the first sleeve and the tapered fixing member that cooperate with each other, the coiled material and the shaft are fastened to each other quickly, thereby improving production efficiency. Through the base mechanism, the coil feeding device is directly used on existing graphic machine for material processing.

The present document provides a method and apparatus for login processing of a Machine-to-Machine/Man (M2M) terminal device, herein, the method includes: transmitting a login request to an M2M platform; receiving an operation corresponding to a detection result obtained after the M2M platform detects a parameter configuration status of an M2M terminal device; and executing the operation corresponding to the detection result. The present document solves the problem of low login processing efficiency of an M2M terminal device in the related art, and then realizes processing the parameter configuration status of the M2M terminal during the login processing of the M2M terminal device.