44results about How to "Improve real-time processing" patented technology

The invention discloses a transformer station intelligent expert system based on editable formula scripts. The transformer station intelligent expert system comprises a real-time analysis display main module, an expert knowledge base offline configuration tool module, a knowledge database module, a processing unit in which an editable formula script preprocessing layer is combined with a rule mode matching layer, a discrete event dynamic comparison unit capable of configuring effective time domains, a rule condition delay effectiveness and delay return trigger mechanism unit and a configurable 'before happening' and 'after happening' correlation site information automatic collection unit. By the adoption of the scheme, information of a transformer station monitoring system, such as the electric quantity, signals, warning and events, can be processed in a unified and comprehensive mode, an expert knowledge base can be flexibly modified and extended, the expert system can accurately and timely assists transformer station operators on duty and maintenance staff in correctly judging and processing accidents and exceptions through intelligent warning and real-time online analysis, and therefore the new requirements oriented to digital and intelligent transformer stations are better met.

The invention belongs to the technical field of SAR (synthetic aperture radar) image power line detection and particularly relates to a local Hough transform and morphological optimization SAR image power line detection method. The method specifically includes: subjecting an SAR original image to edge detection by adopting an Edison operator to obtain a binary image including edge information; dividing multiple edge belts in the binary image including the edge information; subjecting each edge belt in the binary image with the edge belt divided to straight line detection by adopting a polar-coordinate transformation space based Hough transform method to obtain a Hough transform detection result; in each edge belt of the Hough transform detection result, selecting a line segment with the longest length, if the closest range between an optional rest line segment and the line segment with the longest length is smaller than a preset range threshold and an included angle between the corresponding line section and the line section with the longest length is smaller than a preset included angle threshold, deleting the corresponding line section, and if not, retaining the corresponding line section; finally obtaining an SAR image power line detection result.

The invention discloses a real-time low-light color image enhancement method. The method includes the steps: firstly, converting an RGB color image to a YCbCr color space; secondly, subjecting a Y-component of the YCbCr color space to histogram statistics and performing histogram equilibrium processing; thirdly, performing threshold determination on Cb and Cr color components respectively, and performing nonlinear transformation on the components out of a set threshold range; fourthly, subjecting the processed YCbCr color components to inverse transformation to an RGB color space, and finally acquiring the enhanced color image. The invention further provides a device of the real-time low-light color image enhancement method. All processing procedures are implemented in an FPGA (field programmable gate array). By the method and the device, the color video images low in light can be subjected to real-time enhancement, the dynamic range of the images is expanded, contrast ratio is increased, enhancement of high-speed high-resolution video monitoring images can be realized, and visual effect of the video images is improved.

The invention discloses an apparatus for short range 3D imaging apparatus based on a MIMO system and an imaging method thereof. The apparatus includes an N x N millimeter wave receiver area array and a synchronous transmitter. The N x N millimeter wave receiver area array is composed of N millimeter wave receivers. The synchronous transmitter is composed of M transmitting antennas, wherein the M transmitting antennas are distributed in the periphery of the N x N millimeter wave receiver area array and are mutually arranged in a same plane. The beneficial effects of the invention are: a novel type of millimeter wave human body detection method for short range 3D imaging is proposed based on the MIMO system and the imaging method has the characteristics of high resolution ratio and excellent instantaneity.

The invention belongs to the technical field of star extraction, and specifically relates to a star extraction method based on energy distribution. The star extraction method comprises following steps: images are scanned successively, and grey values of the imagines are compared with threshold value S; if the grey value of an image element is larger than S, then a nearby 3X3 image element area is searched; it is determined that in the 3X3 image element area whether the grey values of p image elements are larger than T; if it is determined that in the 3X3 image element area the grey values of p image elements are larger than T, then a star centroid is extracted from a nearby 5X5 area via centroid algorithm, otherwise scanning is carried out continuously. The star extraction method is capable of increasing accuracy of star centroid extraction; simultaneous performing of the two star map dividing stages and centroid algorithm can be realized; and real-time performance of star map processing is improved.

The invention provides a data frame transmitting method and a receiving method based on pipelining, relates to the processing field of data frames and solves problems that the serial processing method has longer time delay and worse practicability and influences communication efficiency during the processing of data frames. The technical scheme is as follows: when the data is transmitted, the to-be-transmitted data is extracted and divided into a frame header section and a data section, and then the frame header section and the data section are data-encrypted and encoded in segments in a pipelining manner; when the data is received, the frame header section and the data section are decoded and data-decrypted in segments in the pipelining manner. With the adoption of the pipelining processing manner, the rapid processing of the data frames is realized, the processing timeliness of the data frames is improved and the actual requirement is satisfied; compared with the traditional method, the time of processing the data frames with 6 information bits in the data transmission process is saved by 8.04 milliseconds.

The invention relates to a hardware receiver and method for achieving GNSS-R phase difference extraction. The hardware receiver comprises a code phase difference fast rough measurement module, a code phase difference accurate measurement module, a carrier phase difference estimation module, and a full-phase transformation and carrier phase difference correction module. The method is based on a FPGA and a DSP, achieves reflected signal effective coherent integration by using a FIFO to perform data time delay, uses a code phase difference rough measured value as an initial computation code phase different accurate measured value of a Gauss-Newton iteration algorithm, guides carrier phase difference solution by using a code phase difference accurate measured value, computes the initial phase values of a direct signal and a reflected signal by using full-phase FFT, and corrects a carrier phase difference final result by means of the comparison of the initial phase values. A hardware device may accurately process acquired direct navigation signals and reflected navigation signals in real time and may correct a carrier phase difference measurement error caused by a reflection plane.

The invention discloses an infrared imaging apparatus and a nonuniformity processing method thereof and belongs to the infrared imaging technology field. The apparatus comprises an infrared detector signal processing module (M1), an infrared detector switching module1 (M2), an infrared detector switching module2 (M3), an infrared imaging apparatus shell fixing structure module (M4) and an infrared imaging apparatus optical lens fixing structure module (M5). The nonuniformity processing method based on the infrared imaging apparatus comprises hardware circuit correction, and single point and two-point calibration adaption correction processing methods. In the invention, a single-chip FPGA is used to realize infrared focal plane array independence imaging and fusion display of two paths of different wave bands or a same wave band; a modular design architecture is used; according to different detection distances, an infrared optical lens is replaced; advantages that a system integration level is high, the architecture is simple and flexible, processing real-time performance is high, the size is small, system reliability is high and so on are possessed; a software and hardware cooperation nonuniformity correction capability is possessed; and a correction effect is good and an adaptive correction capability is possessed.

The invention discloses a frequency-sweeping nonlinear correction distance measurement method based on similar triangular interpolation sampling, and relates to the technical field of frequency-sweeping interference (FSI) measurement, FMCW laser radar and the like. In allusion to the problem that an existing method cannot provide an accurate resampling sequence, is not thorough in frequency-sweeping nonlinearity elimination and results in reduction of the precision of frequency-sweeping interference measurement, the method is used for eliminating the spectrum broadening effect generated by the frequency-sweeping nonlinearity of the sweep frequency laser, and the excellent property of measuring the full width at half maximum of the spectrum can be effectively improved. Especially, when the limit distance is measured, the error is smaller than that of taking a near zero point as a sampling point. Meanwhile, the signal processing speed of the similar interpolation algorithm is increased due to the fact that matrix multiplication is used. According to the invention, high-precision and high-real-time processing of a measurement result is ensured when the absolute distance measurement system carries out remote measurement.

The invention discloses a message processing method and device, and relates to the technical field of computers. One specific embodiment of the method comprises the following steps: acquiring positioninformation of an unread message in a partition according to a message theme and the partition subscribed by a message user, and storing the position information of the unread message in a cache; counting the number of the unread messages, and allocating threads corresponding to the number of the unread messages, the threads being used for acquiring the unread messages; and obtaining an unread message in the partition and corresponding to the position information in the cache through the thread, so as to carry out service processing on the obtained message. According to the method, the position information of the unread message is cached, the unread message corresponding to the position information in the cache is acquired through the threads, and then service processing is performed on the unread message, so that the unread message can be acquired according to the specified position information in the cache, and the message processing real-time performance is improved.

The invention discloses an image mapping conversion method based on a multiple lenses and multiple sensors and a device. The device comprises a lens group, a sensor module and an image processing module; the image processing module comprises a data interface, projection conversion modules and a global control module; the lens group comprises at least one lens; the sensor module is used for converting an optical signal reflected by an object to an electric signal; the data interface is used for transmitting the data of at least one lens to the projection conversion module or converting the data of all lenses to an unified clock domain; the projection conversion module is used for processing the data of at least one lens and the sensor; the global control module is used for performing global control processing on the projection conversion modules. The image mapping conversion disclosed by the invention comprises an image conversion and an image mapping, simplifies the complexity of the conversion and mapping, reduces the operation quantity, improves the timeliness and reduces the power loss. Besides, the invention simultaneously process mapping and conversion of multiple lenses and supports the data reuse processing of multiple imaging units.

The invention specifically relates to a data service system based on the Internet of Things, belonging to the field of data services. According to the invention, a data acquisition module is connected with a wireless network and a data processing module; a wireless network is connected with the data processing module via a remote data transmitting and receiving module; the wireless network and the data processing module are both connected with an emergency control module; the data processing module is further connected with a client and a server; the client is connected with the server; the server is connected with a memory management module; and the memory management module is connected with a memory device. The data service system employs the wireless network data transmission schemes of GPRS and GSM, so integrity of data transmission of the system is improved; a multi-thread scheduling strategy is employed to realize improvement of the real-time processing performance of data and reduction in waiting time for data; and a second-level hierarchical directory storage structure is employed for storage of classified data, so data storage is convenient, simple and fast.

The invention belongs to the technical field of SAR image power line detection, and particularly relates to a SAR image power line detection method based on local Hough transform and morphological optimization. Value image; divide a plurality of edge bands in the binary image containing edge information; in the binary image divided into multiple edge bands, use the Hough transform method based on the polar coordinate transformation space for each edge band. detection to obtain the Hough transform detection result; in each edge band of the Hough transform detection result, select the longest line segment, if the shortest distance between any other line segment and the longest line segment is less than the set distance threshold, And if the angle between the corresponding line segment and the longest line segment is less than the set angle threshold, the corresponding line segment will be discarded, otherwise, the corresponding line segment will be retained; finally the SAR image power line detection result is obtained.

The invention discloses a metadata cluster parallel distribution processing method and a metadata cluster parallel distribution processing system. The method comprises the following steps: receiving network packets in network connection with metadata clusters in parallel, and distributing the received network packets to priority processing queues corresponding to the metadata clusters according to the preset priority, wherein each priority processing queue includes processing sub queues different in priority; and up-sending the network packets according to the priority of the processing sub queues in the priority processing queues, and up-sending the network packets of the processing sub queues with the same priority in different priority processing queues in parallel. By adopting the method and the system of the invention, the real-time performance of metadata cluster control packet processing is improved, and stable operation of metadata clusters at high load is realized.

The invention discloses a counter electromotive force detection circuit of a brushless direct current motor and a range hood applying the counter electromotive force detection circuit. The counter electromotive force detection circuit of the brushless direct current motor comprises an input comparison module, a conversion module and a latch output module. The input comparison module compares a busvoltage with a back electromotive force and transmits a comparison result to the latch output module, and the latch output module latches the comparison result of the input comparison module after receiving a trigger signal of the conversion module and finally outputs the comparison result. According to the counter electromotive force detection circuit of the brushless direct current motor, a hardware circuit detection method is used for replacing a software algorithm, internal resources of the MCU processor are released, MCU space resources are provided for expansion of other functions except motor control, and the cost is reduced; meanwhile, the hardware processing speed is higher than that of software processing, a zero-crossing detection result is directly outputted by hardware without a software algorithm, the complexity of software design is greatly simplified and the real-time performance of processing is improved.

The invention discloses a side-scan sonar image processing method and device. The method comprises the following steps: 1) setting a sliding window M for an acquired side-scan sonar image to perform sliding caching; 2) carrying out denoising processing on the cached image; 3) image layering enhancement; taking the denoised image as an input signal of bilateral filtering layering, calculating high-frequency image data of single-line frame data based on cache block filtering data, and performing weighted fusion on the high-frequency image and an original data image to realize image enhancement; and 4) carrying out gray scale equalization processing on the enhanced image. According to the method, the details of the image are effectively enhanced through processing, and the real-time performance of processing is ensured.

The invention discloses a method for demodulating a quadrature phase shift keying modulation signal, which mainly solves the problems of long time required by carrier synchronization, poor receiving sensitivity and dynamic range, small phase margin and more occupied resources in the existing method. According to the implementation scheme, after being subjected to AD analog-to-digital conversion, multiplying a frequency band signal by a pair of initial orthogonal signals generated by a numerical control oscillator at the same time, and then carrying out low-pass filtering on the result to obtain two paths of intermediate signals I and Q; performing direct connection, inversion, summation and differencing operation on the two paths of intermediate signals in sequence to obtain four pairs of result orthogonal signals; and judging the quadrant and the blind phase point of the phase difference phi between the receiving end and the transmitting end according to the values of the four pairs of result orthogonal signals, and selecting the sign bit output of the four pairs of result orthogonal signals by using the result to obtain a demodulated baseband signal. According to the invention, the time delay of signal demodulation is reduced, the receiving sensitivity and the dynamic range are improved, the phase margin is improved, hardware resources are reduced, and the method can be used for electronic navigation and communication data transmission systems.

The invention discloses a financial stream data-oriented anomaly detection method, which comprises the following steps of: (1) acquiring financial data as a data source by kafka, ensuring stream semantics of the data by using the kafka, and transmitting the data to an Flink end; (2) the Flink platform preprocesses the data; and (3) the outlier detection model performs outlier detection on the stream data by adopting a kernel density estimation (KDE)-based method. And (4) dividing the data space into cells with fixed area sizes by an algorithm, assigning weights to the cells according to the distribution condition of the data objects, and processing the influence of the newly arrived data objects and the expired data objects in the sliding window on the KDE value. And (5) cells with small weight net variation are filtered by adopting a weight-based mode, so that the calculation cost is reduced, and the detection speed is improved. The invention provides a set of flow processing system and an anomaly detection method based on kernel density for anomaly detection of financial flow data, and provides a solution for anomaly detection of flow data.

The invention discloses an image mapping transformation device based on multi-lens and multi-sensor, which includes a lens group, a sensor module and an image processing module. The image processing module further includes a data interface, a projection transformation module and a global control module; the lens group includes at least A lens; the sensor module is used to convert the optical signal reflected by the target object into an electrical signal; the data interface is used to transmit the data of at least one lens to the projection transformation module or convert the data of each lens into a unified clock domain; the projection transformation module is used to process data from at least one lens and sensor; the global control module is used to perform global control processing on each projection transformation module. The image mapping transformation of the present invention includes image transformation and image mapping, which simplifies the complexity of transformation and mapping, reduces the amount of calculation, improves timeliness, and reduces power consumption; multiple lens mapping transformations are processed simultaneously, and multiple imaging unit data multiplexing processing is supported. .

The invention provides a database access method and device, and relates to the technical field of databases. The method comprises steps of acquiring an access request; determining an initial server responding to the access request according to the operation type of the access request; judging the online state of the initial server according to the detection operation, wherein the detection operation is an operation of detecting the online state of each server in the database cluster according to a preset time interval; and if the initial server is online, sending the access request to the initial server for execution. According to the method and the device, the processing real-time performance of the online state of the server can be improved, network waiting caused by connection failure is avoided, and the complexity of database access can also be reduced.

The invention discloses a bit synchronization method of a digital communication data transmission system, which mainly solves the problems of long synchronization establishment time and more occupied resources in the existing method. According to the scheme, a pseudo-random sequence is selected to carry out white noise processing on data information sent to a sending end; taking the sign bit of the baseband signal of the receiving end, and detecting the edge jump of the sign bit; two modulo 10 counters are designed as a main counter and an auxiliary counter; in an initial state, the two counters are both kept in a reset state; during edge hopping, the main / auxiliary counter performs ping-pong identity switching work, the main counter circularly counts to generate a bit synchronization clock signal, and the auxiliary counter is reset and kept in a reset state; and the sampling decision device always carries out sampling decision on the baseband signal sign bit of the receiving end by using the bit synchronization clock signal generated by the main counter, and recovers data information. Bit synchronization can be achieved within one code element time, the real-time performance of obtaining the bit synchronization clock signal is high, occupation of hardware resources is reduced, and the method can be used for communication data transmission and electronic navigation.

The invention relates to a three-dimensional intelligent digitalization generation method and system based on two-dimensional shadow information. The method includes the step of generating two-dimensional shadows, the step of building a three-dimensional model and the step of hiding two-dimensional information, wherein in the step of generating the two-dimensional shadows, images or pictures are converted into two-dimensional shadow images through an image processing technology; in the step of building the three-dimensional model, pixels in the two-dimensional shadow images generate two-dimensional point cloud data, third-dimensional coordinate data are given, and three-dimensional point cloud data are sequenced again to form the three-dimensional model; in the step of hiding the two-dimensional information, the three-dimensional point cloud data are itemized in the two-dimensional direction and stacked in the third-dimensional direction to form multiple branches, and perspective projection transformation is carried out on the three-dimensional point cloud data so that information of the two-dimensional shadow images is hidden in the three-dimensional model. By means of the method, the shadow, presented through projection according to the illumination angle, of the three-dimensional model where the two-dimensional shadow information is hidden is of the shape of the original images or pictures.

The invention discloses a method for identifying an oil depot target in a remote sensing image. Firstly, the saliency of the phase spectrum of the entire scene is calculated, and all regions of interest in the scene that may contain the target are extracted according to the saliency of the phase spectrum; in the feature extraction, local regression is adopted The kernel model calculates the local structural features of the region of interest point by point, and generates a feature descriptor that can describe the target structure; in the target detection stage, the cosine similarity is used as similarity measurement to calculate the similarity between the region of interest and the sample image of the oil depot , and use the feature descriptor's ability to distinguish positive and negative samples and the characteristics of the similarity surface to construct a decision network with adaptive capabilities, obtain the preliminary results of target detection through the decision network, and remove redundant preliminary results through the non-maximum value suppression algorithm , to obtain the final target detection result; the general detection method of the oil depot target in the remote sensing image proposed by the present invention has a good effect on multi-scale and multi-view target recognition.