47results about How to "Efficient processing speed" patented technology

The invention provides a satellite-borne multichannel aerial SAR data channel magnitude-phase error correction platform, wherein, ratio results of magnitude-phase characteristics of other channels anda magnitude-phase characteristic of a selected standard channel can be obtained after selection of the magnitude-phase characteristic of the channel as criterion and complex division operation of themagnitude-phase characteristics of the other channels and the magnitude-phase characteristic. When actual data is processed, that the magnitude-phase characteristics of the other channels except thestandard channel is divided by the corresponding ratio results is only needed, thereby errors which are led in by a signal generator are avoided; in addition, though the magnitude-phase characteristicof the selected standard channel is not completely ideal, occurrence of false targets in the azimuth direction can be effectively prevented and an ideal azimuth direction image performance index canbe obtained due to the fact that data of each channel is provided with the same magnitude-phase characteristics after magnitude-phase error correction. Simultaneously, error compensation effects are evaluated and adjusted by self adaptive means from the imaging angle, thereby satisfactory compensation effects are obtained.

The invention relates to a method and device for deep demethanolization of C4. The demethanolization treatment of C4 extracted from the top of the azeotropic distillation tower is divided into three steps: firstly, the pre-distribution extraction of C4 is carried out by using microphase distribution technology; The C4 and the extract after the final extraction fully enter the modular deep separation device for deep separation of C4 and the extract; the C4 after demethanolization enters the subsequent production device, and the extract returns to the methanol recovery tower for further extraction. Recycle. The present invention also provides a set of devices for realizing the method, including an extraction device and a separation device; the extraction device is divided into a microphase disperser and a micro-swirl flow extraction device; the separation device is a coalescence separation module and a corrugated plate enhanced sedimentation, water bag, Boundary meter and several other parts. The method of the invention has high processing speed and high efficiency, and the methanol content in the treated C4 is not more than 5 ppm, thereby improving the performance of the C4 follow-up deep processing device.

The invention provides a missile attitude measuring method based on satellite navigation. According to the missile attitude measuring method, spatial coordinates of a missile and a satellite are utilized for defining a ground deviating direction as a reference direction of missile rotation firstly; then the frequency and the phase of a modulating signal are estimated by means of a signal processing method; and finally rotating speed and rotating angle of the missile are estimated. The method comprises the following steps of 1, acquiring a signal; 2, processing a geometrical position signal; 3, estimating the frequency change range of the modulating signal by means of a difference method; 4, estimating the frequency of the modulating signal by means of discrete Fourier transform; 5, estimating the phase of the modulating signal; 6, calculating a parameter which describes rotating attitude of the missile; and 7, describing the attitude of the missile. The missile attitude measuring method overcomes defects of incapability of performing attitude measuring on a remote missile, requirement for prestoring missile track and storage space consumption in prior art. Application range of missile attitude measuring is expanded. Furthermore storage space consumption and processing time consumption are reduced.

The invention discloses a method for classifying massive network text and non-text images. First, a multi-scale space division network is constructed, and then the multi-scale image block label information of the image is obtained for the images in the training image set, and the multi-scale space division is performed according to the constructed multi-scale space. Network, use the marked training data set to train the network parameters of the multi-scale space division network, and then use the constructed multi-scale space division network and the network parameters obtained from training to classify the large-scale network images to be tested, and finally obtain the classification of the images As a result, a decision is made as to whether the image is a text image, and the approximate location of the text region in the image is obtained. The method of the invention has high classification accuracy rate of text and non-text images, and high classification efficiency.