665results about How to "Improve processing precision" patented technology

The invention discloses a method which utilizes FPGA to carry out the floating-point FFT processing, and relates to the signal processing technical field. The method aims to avoid the deficiency of the prior FFT processing method, exceeds the universal DSP processing method in the operation efficiency, and can finish the whole FFT processing in a shorter clock cycle. The method has the realization process that the input data are processed to be floating-point numbers; floating-point data are stored into an intermediate data memory 1 in order; the data are read from the intermediate data memory 1 to carry out a butterfly operation, and the processing result is stored into an intermediate data memory 2; the data are read from the intermediate data memory 2 to carry out the butterfly operation, and the processing result is stored in the intermediate data memory 1; the two operations are alternatively carried out until the FFT processing is over; the operation result is read from the intermediate data memory 1 or 2 according to the inverted order of the address; the output floating-point data are processed to be fixed-point numbers, and are output together with the floating-point result. The invention is applicable to the digital signal processing technology field, such as radar, communication, images, etc.

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 and a 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 the magnitude-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 the standard 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 characteristic of 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 can be 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 present invention discloses a method of reconstructing a three-dimensional dental model based on photos or videos, and a method of reconstructing the three-dimensional dental model based on RGBD images, and relates to the field of image processing. The first method is composed of an incremental approach and a global approach. According to the incremental approach, in the condition that when a generated dental model is not complete enough, or a certain key area of the generated dental model is not clear enough, photos shot at new angles are additionally provided and subjected to treatment, so that a better generating effect is realized. According to the global approach, in the condition that the key areas of the dental model can be clearly expressed when shot at fixed and complete shooting angles, the dental model is generated through one shot without the manual determination or the manual intervention. According to the RGBD approach, the depth information does not need to be calculated, so that the calculation of non-RGBD images is optimized. When the depth information is calculated, the accuracy of the acquired depth information based on RGBD images is higher than that of to-be-estimated non-RGBD images. According to the technical scheme of the invention, corresponding optimization schemes are respectively provided for the above three approaches. Therefore, process flows are reduced and the processing accuracy is improved.

The invention relates to a double-shaft centrifuge based gyro accelerometer testing method. Two grating strips are arranged on a reversal platform, grating reading heads are arranged on a large arm of a centrifuge, a connection line of the two grating strips passes through the axle center of the reversal platform, after the installation is finished, a main shaft of the double-shaft centrifuge and the reversal platform can operate at different rotating speeds and are always in a reversal synch state. By making the time of integer circle or semicircle rotation of the double-shaft centrifuge as a sampling period, the output pulse number of a gyro accelerometer under test is read, and then a calibration result of each error model coefficient is obtained by using a regression analysis method. The testing method can be used for accurately calibrating high-order error term coefficients and particularly the K2 term of the gyro accelerometer, and meanwhile, the data processing accuracy is increased.

The invention discloses a correlation analysis method and system for wireless network alarm management, and the method comprises the steps: collecting X pieces of alarm information in a current network in real time in a mode of distributed alarm collection, wherein X is a natural number greater than one; carrying out the filtering of the X pieces of collected alarm information, so as to remove thealarm information which is not correlated with the fault alarm management and/or error alarm information, and obtain Y pieces of to-be-processed alarm information, wherein Y is not less than one andis less than X; carrying out the normalization processing of the Y pieces of to-be-processed alarm information, and obtaining Y pieces of alarm information with the consistent alarm information specification; carrying out the correlation analysis of the Y pieces of alarm information, recognizing the types of the Y pieces of alarm information and/or the correlation between the Y pieces of alarm information according to a correlation analysis result, determining fault equipment corresponding to the alarm information, and achieving the fault alarm management of the fault equipment.

The invention relates to an intelligent extraction method of a build-up area on the basis of nighttime light data. The intelligent extraction method comprises the following steps: adopting an adaptive particle swarm optimization algorithm to realize the optimal selection of the selection parameters of a VIIRS (Visible Infrared Imaging Radiometer Suite) nighttime light image sample and a MODIS (Moderate Resolution Imaging Spectroradiometer) normalized difference vegetation index image sample; adopting a region growing algorithm based on SVM (Support Vector Machine) classification to finish SVM model training, and adopting a cross validation method to carry out precision validation on the model; and according to an optimized parameter, determining a city sample and a non-city sample, and adopting the region growing algorithm based on the SVM to extract a city built-area range. By use of the intelligent extraction method, from a sample selection source, the adaptive optimization of a sample selection parameter is carried out, and the SVM and the region growing algorithm are adopted to improve processing efficiency and precision for the nighttime light data to extract the nighttime light data.

The invention relates to an information extraction method applicable to a hyperspectral image, sequentially comprising the following steps of: firstly, acquiring an imaging spectral image; secondly, resampling a ground object sample standard spectral curve; thirdly, finding an obvious wave band of the ground object sample standard spectral curve and judging the wave band; fourthly, resampling theimaging spectral image; fifthly, carrying out secondary resampling on the ground object sample standard spectral curve; sixthly, judging a type of a spectral genetic wave band; seventhly, comparing the ground object sample standard spectral curve with the type of a corresponding wave band in an image element spectral curve of the imaging spectral image, and judging whether the same ground object is represented; and eighthly, obtaining a sum of depths of all the spectral gene wave bands, wherein the larger the sum of the depths of the spectral gene wave bands is, the higher abundance of the ground object on earth surface is. By adopting the information extraction method disclosed by the invention, amount of data processing is reduced, interference of characteristic sum noise of other ground objects is reduced, processing speed and accuracy are improved, and good effect on processing of the hyperspectral image is achieved.

The invention discloses an image denoising method capable of quickly and effectively retaining edge and directional characteristics. After an image is subjected to wavelet transform, a low-frequency subimage centralizes most energy noise in an original image and is subjected to secondary Wiener filtering to highlight the details of a fusion image; since the image noise is mainly centralized in wavelet high-frequency subimage parts in three different directions, the coefficient of the image noise is small, denoising processing is carried out by singular value decomposition, a high-frequency diagonal subimage is subjected to the singular value decomposition with a high-frequency subimage in line direction or a high-frequency subimage in column direction after rotating to the line direction or the column direction, and meanwhile, the high-frequency subimage is subjected to edge extraction and retaining in order to avoid losing edge detail information; and finally, the denoised low-frequency and high-frequency subimages are subjected to inverse wavelet transform to reconstruct a final denoising image. Singular value numbers required for singular value reconstruction images are jointly determined through a peak signal to noise ratio of the image and a traditional method.

The invention provides a similar video processing method and device based on artificial intelligence, electronic equipment and a computer readable storage medium. The invention relates to the big datatechnology in the field of cloud technology. The method comprises the following steps: carrying out multi-dimensional feature extraction processing on an image in a video, and carrying out fusion processing on extracted multi-dimensional feature vectors to obtain an image vector of the image; performing feature extraction processing on the audio in the video to obtain an audio vector; performingfeature extraction processing on the text in the video to obtain a text vector; performing fusion processing on the image vector, the audio vector and the text vector to obtain a video representationvector of the video; taking the vector similarity between the video representation vectors of any two videos as the video similarity between the two videos; and processing the two videos according tothe video similarity. According to the invention, the video processing precision can be improved.

The present invention provides a method for constructing spaceborne SAR signal high precision phase-keeping imaging processing platform, comprising the following steps: 1) solving the spaceborne SAR data imaging processing parameters; 2) reading spaceborne SAR echo data in frame and storing in the form of two dimension complex array; 3) performing Fast Fourier Transform in the oriental direction;4) carrying out Doppler centroid circular shifting in the range-Doppler domain; 5) signal interpolating in range-Dopper domain; 6) performing Fast Fourier Transform in the distance direction; 7) performing distance migration correction and compression on data in wave-number domain, finishing focusing in distance direction and compensating residual complex constant phase terms; 8) performing FastInverse Fourier Transform in the distance direction; 9) focusing in oriental direction in range-Dopper domain and compensating residual complex constant phase terms; and 10) performing Fast Inverse Fourier Transform in the oriental direction and compensating linear phase error caused by circular shifting. The method can fast and flexibly finish the phase keeping-imaging processing of SAR signal.

The invention provides a target detection method and device. The method comprises the steps of obtaining a target candidate area in a to-be-detected image; determining at least two component candidateregions from the target candidate regions through an image segmentation network, each component candidate region corresponding to one component of the to-be-detected target, and extracting the localimage features corresponding to each component candidate region from the to-be-detected image; learning each local image feature of each part candidate region through a bidirectional long short time memory network (LSTM) to obtain a part relation feature for describing a mutual relation between the part candidate regions; and detecting the to-be-detected target in the to-be-detected image according to the component relationship characteristics. According to the invention, the image data processing precision of target detection can be improved, the diversity of application scenarios of target detection is enhanced, and the accuracy of the target detection is improved.

The invention relates to a process for treating oil-containing waste water for a ship. The process comprises the following technical steps of: performing solid and liquid separation, coalescing separation treatment, adsorbing separation treatment and the like on pretreated ship oil-containing waste water sequentially. When the oil content in supplied water is less than or equal to 500 mg/L and the turbidity of the supplied water is less than or equal to 200 NTU, the oil content in discharged water quality is less than or equal to 1 mg/L and the turbidity of the discharged water quality is less than or equal to 5 NTU; and the national standard of the Integrated Pollutant Discharge Standard GB8978-1996 can be met. The method is high in treatment capacity of purifying the oil-containing waste water, namely the waste water treatment amount is up to 10 to 100 m<3>/hr, and is high in purifying efficiency, namely the oil removal rate is more than 99 percent and the turbidity removal rate is more than 90 percent. The process for treating the oil-containing waste water for the ship can be also applied to various types of oil-containing waste water, such as oil-field reinjected water, oil-base oil-containing waste water, metallurgical oil-containing waste water, steel factory oil-containing waste water, machined oil-containing waste water and the like, so that the treated oil-containing waste water can be treated until reaching the standard, and waste oil resources can be recycled.

The invention discloses a method for image stitching based on mean value duplication removal. The method comprises the following steps: firstly respectively separating a first overlapped image and a second overlapped image from a first image and a second image, thus separately conducting mean value processing on the first overlapped image and the second overlapped image, which reduces image processing work compared to direct processing of the first image and the second image and is conducive to increasing image processing efficiency. According to the invention, the mean images obtained after the processing of the first overlapped image and the second overlapped image are intended for covering a stitched image between the first image and the second image through the stitching of mutual covering of the overlapped parts, so that a first feature image and a second feature image undergo transition stitching through the mean value images, which avoids pixel hopping caused by direct stitching with no transition of the stitched image of the first feature image and the second feature image.

The invention discloses an ultra-large type plate shearing machine plate shearing edge curve detection system and method. The plate shearing edge curve detection system mainly comprises an optical imaging unit, an image acquisition and processing unit and a main control computer. The plate shearing edge curve detection method includes the following steps: first, CCD sensors and lenses which are installed on an ultra-large plate shearing machine are utilized to perform real-time image acquisition on a plate shearing edge under cooperation of strip-shaped light sources; after preprocessing of collected images, the images collected by adjacent industrial cameras are spliced, an image edge is detected after edge burrs are removed, and finally a coordinate axis and coordinate points are selected and a fitting technology is utilized to detect an actual machining curve of a sheared plate. The ultra-large type plate shearing machine plate shearing edge curve detection method utilizes image real-time acquisition and image processing technologies to detect the plate shearing edge curve, and the detection method can achieve higher detection precision and detection efficiency compared with a traditional detection method.

The invention discloses a random drift real-time filtering method for a fiber-optic gyroscope. Compared with the prior art, the method comprises the following steps: in a marine strap-down navigation attitude system, de-noising an acquired fiber-optic gyroscope signal in real time; decomposing data in a sliding data window according to the appointed number of layers through a second generation wavelet lifting algorithm to obtain wavelet coefficients of each layer and scale coefficients of the last layer; establishing corresponding threshold rules for the decomposed wavelet coefficients of each layer, and performing threshold quantization; combining the processed wavelet coefficients and the scale coefficients of the last layer, and progressively reconstructing the scale coefficients of each layer to obtain a de-noised signal. By virtue of the random drift real-time filtering method for the fiber-optic gyroscope, the real-time de-noising accuracy and the reaction speed of the fiber-optic gyroscope signal are effectively improved, so that attitude information errors are further suppressed. In an application process, parameters such as the width of the data window and the number of wavelet decomposing layers are easily set and can be set according to an experience value obtained by a simulation experiment.

The invention relates to a frequency measurement device and a method, belonging to the technical field of a power system. The frequency measurement device comprises a central processing unit, a filter circuit and a timer; the filter circuit and the timer are respectively connected with the central processing unit; the filter circuit is used for converting sine-wave electric signals into square-wave electric signals with the same cycle as the sine-wave electric signals; the timer is used for recording the time of occurrence of turning of the square-wave electric signals; and the central processing unit is used for receiving the square-wave electric signals output by the filter circuit, reading the occurrence time recorded by the timer, determining the turning cycle of the square-wave electric signals, and then determining the frequency of the sine-wave electric signals according to the turning cycle. Since the sine-wave electric signals are converted into the square-wave electric signals within the same cycle as the sine-wave electric signals, and the hardware for capturing the turning of the square-wave electric signals can be realized conveniently and has high processing accuracy and speed, the efficiency of measurement of the frequency of the sine-wave electric signals is increased.

The invention relates to an underwater laser time service system and method based on Beidou. The underwater laser time service system based on Beidou comprises an unmanned aerial vehicle mobile base station which is used for receiving navigation signals and time service information t transmitted by Beidou satellites, carrying out navigation according to the navigation information, moving to a position above an underwater time service receiving device, obtaining precise time according to time service time t, generates a laser signal comprising the precise time and transmitting the laser signal to the underwater time service receiving device; and the underwater time service receiving device which is used for receiving the laser signal, carrying out analysis to obtain standard time of the unmanned aerial vehicle mobile base station, detecting depth information h<2> of the position at which the underwater time service receiving device is located, carrying out data processing on height information h<1>, laser transmitting angle information theta, the depth information h<2> and a water refractive index n through combination of the refractive index n of the water area, thereby generating transmission delay tau, and obtaining the time service time T through combination of the time service information t and the transmission delay tau. Compared with the prior art, the system and the method have the advantages that the transmission delay resulting from signal transmission can be reduced, the anti-interference capability is improved, and the underwater time service precision is improved.

The invention provides a method and a system for detecting peak value of a photoelectric signal. The method comprises the steps of performing segmental reading on an original photoelectric signal; interpolating preset amount average value parameters into the front end and the back end of each data field which is segmentally read; selecting a mother wavelet function for performing one-order derivation and two-order derivation on segmental data after processing; extracting a maximum value point of a one-order derivative and a second-order derivative maximum value point of second segmental data; determining a peak position in each second data field; and calculating the peak width and peak height in each second data field according to the peak position in the second data field, and acquiring the peak value of the original photoelectric signal. In the whole process, the average value parameter is interpolated into the front end and the back end of the data field of the original photoelectric signal, thereby improving processing precision to the original photoelectric signal; the mother wavelet function is utilized for performing one-order derivation processing and two-order deviation processing on the data field after interpolation, thereby accurately determining the peak position; and additionally, no complicated filtering processing on the original photoelectric data is required, thereby remarkably improving peak detecting efficiency on the original photoelectric signal.

The invention provides a multi-GNSS (Global navigation satellite system) long baseline near-real-time deformation monitoring method based on multiple base stations. In order to improve the accuracy of long baseline calculation, an ultrafast multi-GNSS satellite orbit product is used to reduce the influence of orbit error on baseline calculation. The influence of abnormal data on baseline calculation is presented as much as possible through multiple posterior residual editing iterations. Near-real-time high-precision baseline calculation can be realized through a sliding window-based batch processing mode. The precision and stability of baseline calculation can be improved through a multi-base-station joint processing mode. The precision of baseline calculation is further improved by using a creative ambiguity processing strategy. In order to improve the efficiency of data processing, a normal equation stacking technology is used and the previous normal equation information is fully utilized to avoid repeated calculation. A shared base station is used for processing, and repeated processing by the same base station is avoided. By using an un-differenced processing mode, the advantage of distributed computing can be fully utilized, and the efficiency of data processing can be improved.