44results about How to "Improve calculation results" patented technology

The invention discloses a sea-gas coupled real-time observation buoy system based on Beidou satellite and Iridium satellite double-satellite communication. The sea-gas coupled real-time observation buoy system comprises a sea-gas coupled observation sensor unit, a data collecting communication control unit, an anchoring unit, a buoy body platform, a power supplying unit and a data processing center. The buoy data collecting communication control unit adopts a Beidou satellite system to transmit conventional block method sea-gas flux, upper-layer sea current profile and buoy operation state parameters which have high requirements on data safety and low requirements on the communication quantity, and adopts an Iridium satellite system to transmit high frequency eddy related and high-resolution seawater skin surface temperature profile data which have high requirements on the communication quantity. A buoy can synchronously observe turbulence flux parameters of an eddy-related method of amovable platform, flux parameter of a block method, the seawater 0-1.0 m high-resolution skin surface temperature profile and the 0-70 m upper-layer sea current profile in real time in real time. Thesea-gas coupled real-time observation buoy system is suitable for being released in deep and open sea, and has the characteristics of diverse sea-gas interface observation parameter, safe data transmission and the like.

The invention discloses a tracked vehicle gun aiming method based on line-of-sight tracking. The tracked vehicle gun aiming method based on line-of-sight tracking aims to overcome the problems of lowapplicability, low precision and unsatisfactory real-time performance in the prior art, and comprises the following steps of: 1) acquiring a human eye image, wherein after a line-of-sight tracker is correctly worn, the position of a human eye CCD camera relative to the center of two eyes is adjusted, and the human eye CCD camera faces right towards the front direction so as to ensure that the line-of-sight tracker can accurately complete acquisition of the eye image of a shooter; 2) pupil detection, wherein (1) eye image preprocessing, (2) pupil center positioning, (3) pupil tracking, (4) Pulchin facula center positioning and (5) P-CR vector obtaining are carried out; 3) fixed-point calibration; 4) line-of-sight tracking, wherein after the mapping relationship between a P-CR vector and thecoordinates of a point of sight of a scene image is obtained, eye movement parameters of a new human eye image are further obtained, and then the position of the point of sight in the scene image isestimated by means of a determined mapping relationship to complete final line-of-sight tracking; and 5) executing control instructions.

The invention relates to the technical field related to notification of websites, in particular to a method and system for carrying out notification on the website projects. The method for carrying out notification on the website projects comprises the steps of (1) obtaining interestingness grades of a plurality of users to a plurality of projects, (2) determining the set of projects adjacent to a targeted project, (3) determining a predicted grade of a targeted user to the targeted project according to the interestingness grades of the targeted user to the adjacent projects, and (4) sending a notice of the targeted project to the targeted user when the predicted grade of the targeted user to the targeted project exceeds a preset grade threshold. Compared with the prior art, the recommendation effect of the grade prediction method of the method and system for carrying out notification on the website projects is better, and quality of recommended commodities is greatly improved.

The invention provides a self-adaption curvelet threshold value earthquake denoising method based on local variance analysis. According to the method, on the basis that deep analysis is conducted on the characteristics of a curvelet coefficient on different scales and at different angles, a self-adaption threshold value technology is provided so that effective separation of a signal and noise can be achieved at any angle, protection conducted on non-noise components is enhanced with the help of local variance information of a curvelet area, and therefore damage to the effective signal especially the weak signal can be reduced. Practices on synthetic seismogram and land seismic data prove that the actual denoising effect of a threshold value iteration method is not ideal, the denoising effect of the method is more real and effective, and the method has the unique advantages in the aspects of noise attenuation and detail protection.

The invention relates to the technical field of power systems, and discloses a short-circuit current calculation method based on power grid characteristics. The method comprises the steps of A) selecting power distribution network characteristics, and taking short-circuit current If_DG flowing through a measurement point after IIDG access as a sample label; B) performing power grid modeling and simulation by utilizing simulation software to obtain a sample set; C) establishing a machine learning model through a machine learning integration method, and performing error analysis; and D) using the trained machine learning model for short circuit current calculation. According to the short-circuit current calculation method, by analyzing the power distribution network characteristics related to the short-circuit current, a power distribution network characteristic combination is provided; power grid modeling and simulation are carried out by utilizing simulation software to form the sampleset; short-circuit current calculation is accurately carried out; the calculation speed is high; and the calculation result accuracy is high.

The invention provides an improved method for selecting a high-voltage circuit breaker in a power transformation and distribution system, which summarizes a relationship between a periodic component and a direct current component of breaking short-circuit current and improves a calculation formula for the breaking capacity of the circuit breaker. Different from the traditional calculation method, the improved method respectively calculates the periodic component and the direct current component of the short-circuit current and then compares with breaking current of the circuit breaker so as to judge whether the breaking current of the circuit breaker can meet requirements or not. The time-periodic component relationship which is relatively simple and easy to search and calculate is found out by using a relational expression, a value range of a time constant in a power distribution system is deduced, and a calculation method is provided; and the breaking capacity of the circuit breaker is determined through comparing the breaking time. The improved method is simple and clear in calculation process, the original calculation data is easy to obtain, and the calculation result is accurate. Electrical equipment with a high performance-cost ratio can be accurately selected because the accuracy is substantially improved.

The invention relates to a photographing correction method and a photographing device of a multi-lens group, comprising: S1. enabling one lens group in the multi-lens group to capture a first image and generating a first sharpness index based on the first image; S2, judging whether the first sharpness index satisfies a preset first threshold condition, if not, enabling another lens group in the multi-lens group to capture a second image; S3. acquiring the number of pixels added to the first image by the second image based on the first sharpness index, and completing correction of the first image. Based on the sharpness index to evaluate whether the first lens composition image is blurred, When part or all of the first image is blurred, a second lens with a focus range different from that of the first lens group is enabled to photograph the second image, and the blurred area in the first image is replaced or synthesized through the second image, thereby greatly reducing the calculationamount and improving the calculation efficiency under the condition of guaranteeing the image quality.

The invention provides a liquid crystal lens imaging method, comprising: driving a liquid crystal lens to be in a first state, using a first optical signal which passes through the liquid crystal lens at the moment of the first state, generating a first lighting graph corresponding to the first optical signal according to the first optical signal; driving the liquid crystal lens to be in a second state, using a second optical signal which passes through the liquid crystal lens at the moment of the second state, generating a second lighting graph corresponding to the second optical signal according to the second optical signal; processing the first lighting graph and the second lighting graph, using the lighting graph obtained by processing as a final lighting graph; and according to the final lighting graph, generating a final digital image signal. Through processing the lighting graphs, the liquid crystal lens imaging method finally obtains the digital image signal, so as to prevent error problems caused by calculation just aimed at the linear response interval of an imaging sensor when the digital image signal is directly used for processing. Thus, a calculation result of the lighting graph is improved.

The invention provides a bridge span arrangement method and device, equipment and a readable storage medium, relates to the technical field of bridge construction, and comprises a new thought of performing span arrangement by using artificial intelligence. Aiming at the requirement of intelligent span arrangement, on the basis of a basic principle and a calculation principle of bridge span arrangement and in combination with excellent learning decision-making capability of reinforcement learning, a strategy value network is optimized and output through a Sarsa(lambda) algorithm, and the strategy value network is added into an AlphaGo algorithm; the simulation stage can obtain accumulated rewards corresponding to global simulation with the node as a root node through a rapid cross arrangement strategy, and can also directly find the corresponding maximum accumulated reward in the strategy value network; and finally the maximum one of the two is taken as the actual accumulated reward after the node simulation is finished, therefore, the final operation result of the AlphaGo algorithm is improved.

The invention provides methods and apparatus for determination of light sources of images and color vision matching of images. The method for determination of light sources of images includes the steps of calculating the indicating value of each of a group of candidate light sources of an image, wherein each candidate light sources is described by corresponding coordinate pairs (p, q) in a chromacoordinate system; determining the threshold of the indicating values of the candidate light sources; identifying a subset of the candidate light sources with the indicating values not greater than the threshold value, and calculating the weight average of the corresponding coordinate pairs for all the candidate light sources in the subset to obtain the average coordinate pairs describing the light sources of the image in the chroma coordinate system.

A computer-implemented method for simulation of an electrical circuit with circuit components by at least one computing unit includes mapping a coupling of the substrate representations in a couplingequation system for exchange of calculated coupling variables between the subcircuits. The method also includes calculating, in an evaluation step, at least one stability parameter on a basis of the coupling equation system, and deciding, in a selection step and depending on the at least one calculated stability parameter, whether the current separation of the electrical circuit into subcircuits will be used as the basis of the simulation. The method further includes performing, after a successful selection, the simulation of the electrical circuit by calculating the substrate space representations on the at least one computing unit.

The present invention relates to an ultrasound diagnosis apparatus (10), in particular for analyzing a fetus (62). An ultrasound data interface (66) is configured to receive 3D (three dimensional) ultrasound data from an object (12). The ultrasound diagnosis apparatus further comprises a measurement unit (70) for measuring anatomical structures of the object based on the segmentation data and a calculation unit (72) configured to calculate at least one biometric parameter based on the 3D ultrasound data.

The invention discloses a fluid-solid coupling field calculation method of a double-reflector antenna, relates to the technical field of reflector antenna analysis, and aims to provide a calculation method which is high in precision, free of conversion errors and high in practicability. The method mainly comprises the following steps: establishing a model and a computational domain in fluid fieldpretreatment software, and obtaining an intermediate format file after conversion; importing the intermediate format file into fluid field software, determining a solver according to the Reynold number, and obtaining a fluid field result; performing data processing on the fluid field result to obtain a resistance coefficient, a lift coefficient and a moment coefficient of the antenna; converting the format of the fluid field result, then importing the fluid field result into solid field software, establishing a relation coupling unit in the solid field, and obtaining a solid field result. According to the method, fluid and solid field analysis is effectively integrated, and the method is particularly suitable for design, index budget and strength check of various reflector antennas.