64results about How to "Reduce the influence of human subjective factors" patented technology

The invention relates to a multi-objective optimization method for collaborative allocation of time slots of multi-runway approaching-departing flights, belonging to the technical field of civil aviation. The multi-objective optimization method is mainly based on a collaborative decision-making idea and comprises the following steps of: firstly, dividing a time interval into a plurality of time slots, determining the approaching-departing capacity of an airport and dynamically creating time slots; secondly, establishing a multi-objective optimization model with minimum total delay cost, fair delay loss of each aviation company and balanced use frequency of each runway as objectives and with the effectiveness and the approaching-departing capacity of the airport as constraining conditions; and finally, solving by using a genetic algorithm to obtain an optimal flight time slot allocation scheme. The multi-objective optimization method disclosed by the invention is suitable for solving the time slot allocation problem of the approaching-departing flights of each airport.

The invention discloses a tire X-ray image detection and identification method and system. The detection and identification method comprises an offline training step and an online detection step. The offline training step comprises: (11), establishing a training sample; (12), performing defect detection on an image in the training sample, and computing defect information of the training sample; (13), performing quantization coding on the defect information of the training sample to obtain a training sample feature vector; and (14), training the training sample feature vector and a grade judging result into a grade judging model. The online detection step comprises: (21), collecting a to-be-detected tire X-ray image in real time; (22), performing detect detection on the to-be-detected tire X-ray image, and computing all to-be-detected tire defect information; (23), performing quantization coding on all the to-be-detected tire defect information; and (24), computing the to-be-detected tire feature vector and the grade judging model. According to the detection and identification method, artificial subjective factors are avoided, the judgment is more objective and unified, and the judgment efficiency and accuracy are both improved.

The invention relates to a gridding monitoring, diffusion early warning and tracing method for unorganized VOCs of an industrial enterprise. The method comprises the following steps: establishing a VOCs unorganized emission source list of the industrial enterprise; vOCs gridding monitoring stations are screened and arranged on the basis of a VOCs unorganized emission source list and local perennial meteorological conditions through a comprehensive rating method of functional region division and gridding point distribution; creating a VOCs management and control platform, displaying the grid VOCs emission level, time change trend and characteristics of the whole plant, and analyzing the total VOCs emission amount, high concentration value and occurrence time of the pollution map; through the optimized gas diffusion model and in combination with VOCs monitoring data of the whole plant, diffusion early warning analysis of VOCs pollution in the whole enterprise plant area is realized. And traceability analysis of VOCs pollution is realized through a source analysis technology. According to the invention, an integrated fine supervision mode of source combing, process monitoring and system management and control early warning of unorganized VOCs pollution of industrial enterprises is realized.

The invention provides a method for classifying color of a color three-dimensional model based on a neural network. A three-dimensional model aiming at an actual object is obtained through three-dimensional scanning equipment or virtual modeling software; a three-dimensional model database is built; the three-dimensional model file obtained in the previous step is stored in the three-dimensional model database; a model data set for training and identification in the three-dimensional model database is read; the read model data set is input to a three-dimensional model character extraction module to extract the color character of the three-dimensional model; the character data in color is read from a model character database; a quantification sensor of the color character is established; a triad color character vector representing the color character of the three-dimensional object is obtained; an analytic module of the neural network is established; the parameters of the network are adjusted; decision-making operation of the color character vector is carried out through the neural network; and according to a discrimination formula of the neural network, an identification result is output. The method can effectively classify the object colors with different classes.

The invention discloses a video image color cast detection method and device. The method comprises the steps that a video image including a color card is acquired; edge detection is performed on the video image to be detected so that pixel points included by multiple color modules of the color card in the video image to be detected are confirmed; and color cast amplitude of colors corresponding to the color modules of the video image to be detected is confirmed according to the pixel points and pixel values included by the color modules and standard pixel values corresponding to the color modules by aiming at all the color modules of the color card. Compared with methods and the devices in the prior art, color cast amplitude of the video image is automatically detected via application of the method and the device so that the whole detection process is high in detection efficiency, data are objective and accurate and detection speed is high.

The invention discloses an annular laser resonant cavity diaphragm adjustment system. The annular laser resonant cavity diaphragm adjustment system comprises a to-be-assembled annular resonant cavitybody fixed on a substrate; a laser pumped source is arranged outside one side of the annular resonant cavity body; a prism is arranged on one corner of the annular resonant cavity body while reflecting mirrors are arranged on the other three corners respectively; a photoelectric detector is arranged in the horizontal direction of a laser beam emitted by the prism; and the photoelectric detector isconnected with a laser frequency control box and a frequency server through a wire in sequence. An adjustment method comprises the steps of enabling the laser beam emitted by the annular laser resonant cavity to pass through the prism to output light at a brewster angle, and enabling the output light to come to the photosensitive surface of the photoelectric detector in an aligned manner firstly,next, opening the laser frequency control box, performing modal state scanning and frequency stabilizing, and detecting the assembling position of the diaphragm according to the voltage change amountof the laser pumped source before and after diaphragm assembling. The system is simple in detection process, quick in processing, and accurate in assembling result.

The invention relates to big data, and provides a service recommendation method, device and equipment based on image recognition, and a storage medium, and the method comprises the steps: obtaining apicture which is collected by a user through a photographing unit and contains a terminal image; detecting whether the photo meets the condition of photo shooting quality or not; if yes, calling a preset terminal image recognition model, and recognizing the target terminal in the photo to obtain attribute information of the target terminal; calculating evaluation information of the target terminalaccording to the attribute information in combination with a data retrieval algorithm and an evaluation algorithm; and according to the evaluation information, calculating a service scheme for the target terminal by using a preset service product recommendation algorithm, and displaying the service scheme. In addition, the invention also relates to a blockchain technology, and the photo containing the target terminal can be stored in the blockchain node. A user takes a picture with the target terminal, and the terminal identification model is used for objectively identifying the attribute information of the terminal, so that related service products can be more objectively and accurately recommended to the target terminal.

A self-generated neural network construction method comprises the following steps of 1 adding a stimulation signal; 2 evaluating the output intensity of a nerve cell, determining the connection direction of the nerve cell, forming the network connection continuously, and finally generating an initial network; 3 calculating the position and the probability connected with a target nerve cell; 4 determining whether the current network generation process stops, if yes, turning to a step 5, otherwise turning to the step 2; 5 optimizing the network connection via an optimization algorithm; 6 determining whether simulation still needs to be added, if not ending, otherwise, turning to the step 1. The network provided by the present invention is a self-generated network, and the influence of the artificial subjective factors can be reduced effectively. In addition, the generation process of the network takes the biological neural theory as basis, so that the neural brain can be searched possibly and further to realize the real intelligence.

The invention relates to a temporary speed-restriction early warning system based on an active tag and a realization method thereof. The temporary speed-restriction early warning system comprises an active tag and a host machine of the temporary speed-restriction early-warning system, wherein the active tag is used for periodically broadcasting temporary speed-restriction information, and the host machine of the temporary speed-restriction early warning system is installed on a train; the tag information of the active tag comprises the arrangement of the active tag along the route direction of the train; the host machine of the temporary speed-restriction early warning system is provided with a reader operation interface, a current speed and position reading module, a main control module, a speed control module and a display module, wherein the reader operation interface reads the active tag; the main control module is used for positioning the active tag according to the read temporary speed-restriction information by combining the current speed and position information of the train corresponding to the temporary speed-restriction information to obtain the position of the active tag, and calculating a train speed control signal; the speed control module carries out train speed control on the train by a braking module; and the display module is used for prompting a train driver to carry out train speed control on the train. The invention adopts the active tag as a transmitting source, adopts the host machine of the temporary speed-restriction early warning system to read the temporary speed-restriction information and realizes temporary speed-restriction automatization.

The invention relates to the field of electrical automation, and discloses a power grid voltage out-of-limit cause diagnosis method based on a data mining technology, which comprises the following steps: A) extracting out-of-limit voltage, and collecting related characteristics of the out-of-limit voltage; B) deleting bad data, and carrying out data preprocessing to obtain sample data; C) constructing a K-means improved clustering model, and obtaining a voltage out-of-limit cause clustering result; D) carrying out classification identification on voltage out-of-limit cause clustering results;e) constructing an SVM multi-classifier model, and optimizing model parameters; and F) inputting a to-be-diagnosed data sample to the SVM multi-classification model, and taking the output of the SVM multi-classification model as the cause of voltage out-of-limit of the to-be-diagnosed data sample. The method is high in efficiency, reduces the influence of human subjective factors, quickly and effectively processes the power grid low-voltage mass data, improves the K-means clustering model, obtains a good clustering effect, carries out the parameter optimization of the SVM model, and improves the accuracy of the power grid voltage out-of-limit cause diagnosis.

The present invention discloses a traffic accident responsibility determination system based on machine learning, and relates to the technical field of traffic accident responsibility determination. The system comprises a traffic information interaction center, an image processing system and an accident responsibility analysis system. The traffic information interaction center comprises an image character receiving module and a determination result sending module; the image processing system comprises an image loading module, a color extraction module, a contour extraction module, a Gaussian Blur module and a distance analysis module; and the accident responsibility analysis system comprises a feature case search module and a comparison computing module. The traffic accident responsibilitydetermination system based on machine learning allow a reporter to replace the traffic police to perform photograph and evidence collection of a scene of a traffic accident and pass back informationin real time for responsibility analysis and determination so as to effectively reduce the time of scene measurement evidence collection, analysis and determination, is high in efficiency and accuratein determination, reduce human subjective factors, can recover the traffic as soon as possible and can avoid traffic congestion.

The invention is suitable for the field of electrocardiosignal identification, and provides an electrocardiosignal identity identification method and an electrocardiosignal identity identification device. The electrocardiosignal identity identification method comprises the steps of receiving the electrocardiosignal signal of a user; generating a cardiac cycle according to the electrocardiosignal; comparing the generated cardiac cycle with a pre-generated cardiac cycle template, thereby defining a cardiac cycle template with highest similarity; selecting an electrocardiosignal characteristic vector in the electrocardiosignal through a deep learning network model, wherein the deep learning network model establishes a matching model based on a sparse self-coded feedforward back neural network; performing matching between the selected electrocardiosignal characteristic vector and the corresponding identity (ID) in the cardiac cycle template with highest similarity, thereby defining a corresponding identification probability vector; and defining the identity (ID) of the electrocardiosignal according to the identification probability vector. The electrocardiosignal identity identification method and the electrocardiosignal identity identification device can improve identification accuracy for the identity (ID).

The invention provides a processing method of ginger processed pinellia. The processing method comprises the following steps: taking ginger slices and decocting; adding pure pinellia and immersing until the center of pinellia is immersed with the soup; adding alum and continually immersing for 48-72 hours; taking out and steaming; then cooling and slicing; and naturally drying or drying at a low temperature to obtain a finished product. The use amount of the ginger accounts for 25wt% of the pure pinellia and the use amount of the alum accounts for 6wt%-8wt% of the pure pinellia. According to the processing method, the use amount of auxiliary materials is reduced and medicinal materials are pre-treated and steamed, so that the use amount of the alum is greatly reduced, and the effects of detoxifying and removing irritation are guaranteed; effective components of pinellia are kept to the greatest extent; furthermore, the ginger processed pinellia processed by the method has a better finished product shape and high yield; modern equipment can be used for realizing standardized production and parameters of a processing process can be accurately controlled; and the influences of manual subjective factors are reduced, so that the quality of the product is more stable and controllable.

The present invention discloses a method for screening an oil gas wildcat target. The method comprises the following steps: obtaining a drilling success probability of a to-be-screened wildcat target and recoverable resource amount; obtaining a development risk scale of the to-be-screened wildcat target based on the drilling success probability and the recoverable resource amount; determining a threshold of the development risk scale based on an exploration target; and screening according to the threshold of the development risk scale, so as to obtain a wildcat target. According to the wildcat target screening method disclosed by the present invention, optimization of wildcats with different regions, different fields and different resource types on a unified platform can be realized, and not only operability and practicality of the screening method can be significantly improved, but also human subjective influences during a screening process are maximally reduced, and more reliable bases are provided for determination of a wildcat deployment direction and sequence, exploration investment optimization and layout plan making in an oil company.

The embodiment of the invention discloses a lie detection method and device. The lie detection method includes: acquiring a detection parameter of a subject, wherein the detection parameter includes afirst sound production record and a first expression picture; determining the first sound production record as a first input value, and inputting the first input value into a preset deep learning based sound detection model, and acquiring a sound characteristic of the first sound production record and / or a first confidence coefficient; determining the first expression picture as a second input value, and inputting the second input value into a preset deep learning based expression detection model, and acquiring an expression characteristic of the first expression picture and / or a second confidence coefficient; and generating a lie detection result according to the sound characteristic and / or the first confidence coefficient and the expression characteristic and / or the second confidence coefficient. The lie detection method and device can improve the accuracy of the lie detection result.

The invention discloses an Internet-based frozen food online monitoring method. The method comprises the following steps: using a magnetic memory detection technology to detect the defects of a frozen food, including the unfreezing status, the color and the smell change; quantifying the defects by a calculation formula; carrying out unfreezing status evaluation, color evaluation and smell change evaluation on the frozen food according to above obtained magnetic memory detection results; obtaining the technical index of the maximum safety score born by the frozen food; and carrying out quantified and qualitative analysis combination through adopting a fuzzy comprehensive evaluation system, practically establishing an evaluation set by combining engineering, and establishing a comprehensive evaluation judgment matrix. The method has the advantages of facilitation of convenient monitoring, improvement of the intelligence degree, strict control of the quality of the food, reduction of the influences of artificial subjective factors, high qualified rate of the manufactured product, and indirectness in promotion of the good credit image of manufacturers.

The invention discloses a method for accurately obtaining a mining influence boundary and determining a building damage grade. The method belongs to the field of mining damage technology identification, and comprises the following specific steps: (1) interpreting by using N-scene SAR image data continuously imaged in a target area through an InSAR time sequence analysis method to obtain a time sequence of deformation of a mining influence area to the ground surface along a radar sight line; (2) solving earth surface three-dimensional deformation information based on the subsidence prediction model by using the related information of the mined working surface and the geological mining parameters of the target area, so as to accurately obtain earth surface subsidence, inclination (curvature)and horizontal deformation; (3) determining a mining influence boundary and a building damage grade based on each deformation index. The steps are simple, the detection effect is good, an effective,real and reliable data basis can be provided for mining damage technology identification work, mine and enterprise (citizen) disputes can be fairly and fairly solved, and social contradictions are relieved.

The invention provides a method and equipment for determining the influence weight of geological parameters on the resource quantity, and the method comprises the steps: obtaining a plurality of groups of sample data of a to-be-surveyed region, the to-be-surveyed region storing unconventional oil and gas resources, and the sample data being data of each target geological parameter used in a preset resource quantity calculation mode; calculating a resource quantity result of each group of sample data based on a preset resource quantity calculation mode and the multiple groups of sample data; performing multiple linear regression fitting processing on the resource quantity result of each group of sample data and the multiple groups of sample data to obtain a linear regression model of the resource quantity and each target geological parameter, and determining a partial regression coefficient of each target geological parameter based on the linear regression model; and based on the partial regression coefficient of each target geological parameter, determining the weight of each target geological parameter on the influence of the resource quantity. According to the method, the influence of human subjective factors can be reduced, and the purpose of improving the accuracy of the shale gas resource evaluation result is achieved.

The invention relates to a product quality characteristic error traceability analysis method based on multi-source perception, which comprises the following steps of establishing causal relationship between processes by using historical data; screening the historical data so as to compose the screened historical data into a sample space; taking the sample space as the analysis standard, and monitoring the real-time data by a T2 control chart method; performing the orthogonal decomposition on the outbound T2 value according to the causality to obtain decomposition terms; carrying out the out-of-bounds analysis on the T2 statistic value of the decomposition term, and locating the problem procedure. The method of the invention utilizes a large amount of multiple historical data to complete the analysis of the correlation relation between the processes, and reduces the influence brought by the subjective factors. The out-of-bounds T2 value is decomposed according to the relationship directed graph obtained from the analysis process of the relationship between processes, so as to determine the problem process, achieve the accurate grasp of the process, and then achieve the purpose of problem traceability.

The invention discloses a malicious feature extraction method and device and storage media. The method comprises the steps that black file sets used for extracting features and white file sets corresponding to the black file sets in number are selected; black files in the black file sets and white files in the white file sets are converted into multi-dimensional vectors; dimension combing and screening are conducted on the vectors of the black files in the black file sets and the white files in the white file sets, and the malicious features are extracted. Vector conversion is conducted on the known black file sets and the known white file sets, dimension combing and screening are conducted on the converted vectors, viruses and other malicious features can be automatically, timely, accurately and effectively extracted, a large number of analysis workers can be omitted, influence from human subjective factors is greatly reduced, and meanwhile reaction speed is greatly increased.

The invention discloses a system and method for detecting thermal ageing degrees of insulating cardboards. The system comprises image acquisition equipment, image reception equipment and analysis andjudgement module, wherein the image reception equipment is connected with the image acquisition equipment; the analysis and judgement module is connected with the image reception equipment; and the analysis and judgement module is used for detecting fiber widths, hole areas and surface roughness of insulating cardboards so as to judge thermal ageing degrees and residual life of the insulating cardboards. By adoption of the system and method provided by the invention, the influences of subjective factors can be reduced and the detection precision can be improved.

The invention belongs to the technical field of tunnel surrounding rock recognition, and discloses a neural network surrounding rock grading prediction method and system based on geophysical prospecting data, and the method comprises the steps: carrying out the advanced prediction of the front of a tunnel face through a geological radar, namely an electromagnetic wave method, and obtaining a geological radar echo map; carrying out wavelet noise reduction processing on the obtained geological radar echo map; obtaining a one-dimensional data matrix based on the geological radar echo map after noise reduction; taking the obtained one-dimensional data matrix as an input value of a full-connection neural network model, taking the surrounding rock grade of the corresponding mileage as a label value, training and predicting the full-connection neural network model, and obtaining an optimal full-connection neural network model; and carrying out surrounding rock grading prediction by using the obtained optimal full-connection neural network model. The tunnel surrounding rock can be effectively graded and identified.

The invention relates to a double-view-field optical coherence tomography imaging system and a material thickness detection method. The swept-frequency optical coherence tomography imaging system comprises a scanning imaging system and a swept-frequency light source for providing a light source for the scanning imaging system. The scanning imaging system comprises a sample arm and a reference armwhich are correspondingly connected with the swept-frequency light source respectively, and the swept-frequency light source provides a sampling light source or a reference light source for the samplearm and the reference arm respectively; the sample arm further corresponds to a sample placing platform, and the sample placing platform comprises a posture adjusting unit for positioning of a to-be-detected sample. Light returned by the sample arm and light returned by the reference arm interfere with each other to form interference signals, and the interference signals are detected by a balancephotoelectric detector and transmitted back to a PC. According to a double-side-view OCT system provided by the invention, the OCT technology is further expanded to the thickness measurement of the opaque material, so that the precision of measuring the thickness of the opaque material is improved.

The invention provides a fault diagnosis method and device, and relates to the technical field of nuclear power plants. The method comprises the steps: obtaining the operation data of evaluation parameters corresponding to each fault symptom in a preset fault symptom set of a control rod misplug fault; determining the confidence coefficient of the fault symptom according to the operation data; determining whether the control rod misplug fault is established or not according to the confidence coefficient of the fault symptom and thecorresponding diagnosis rule of the control rod misplug fault in a preset fault diagnosis confidence rule base, and obtaining a determination result; and outputting a corresponding fault diagnosis conclusion according to the determination result. According to thefault diagnosis method and device, the fault diagnosis confidence rule base is pre-established, the confidence coefficient of the fault symptom related to the control rod misplug fault is calculated,whether the control rod misplug fault is established or not can be automatically diagnosed in real time in combination with the related diagnosis rules in the confidence rule base, the influence of human factors is reduced, the fault misjudgment rate is reduced and emergency operation guidance is provided.

The invention discloses an urban multi-reservoir water source water distribution method based on comprehensive evaluation information, which relates to the technical field of water resource planning and allocation. At first, a multi-reservoir water source comprehensive evaluation system is constructed, the evaluation factors of four evaluation subsystems are determined, and calculate the comprehensive water quality value of the water quality target subsystem. The comprehensive operation value of the operation subsystem and the comprehensive evaluation value of the multi-reservoir water supplysystem are used to determine the sequence of the multi-reservoir water supply and the actual water supply quantity. The principal component analysis method is used to objectively select the evaluationfactors that affect the water quality objectives when selecting the water quality evaluation factors. The comprehensive evaluation method and fuzzy evaluation method are used to calculate the comprehensive water quality value of the water quality objective subsystem and the comprehensive operation value of the operation subsystem respectively. The method of multi-reservoir water supply allocationbased on comprehensive evaluation value can reduce the influence of human subjective factors and provide scientific basis for the sequence of multi-reservoir water supply and actual water supply.

The invention discloses a multi-mode multi-standard path search method based on genetic algorithm, which is realized by computer. The method solves the multi-mode multi-standard traffic path search problem by defining genetic operator, matching corresponding strategy and utilizing the optimal theory. Under various trip standards, the method is used to solve, which does not have to set the weight value of each standard in advance, avoids no-solution risk brought by constraint method and reduces the influence of personal subjective factors. The path output result of the invention can satisfy various trip requirements and has the characteristic of various trip mode combinations, provides personalized and various trip projects for traveler and provides good technical support for the public trip information service.

The invention discloses a power battery screening method for typical feature vectors. The method comprises the following steps: acquiring charging and discharging voltage signals of a plurality of power batteries with known states, and extracting voltage signal characteristics to form characteristic vectors; estimating a probability density function of each feature according to the values of the same type of samples in each feature variable, and generating a plurality of new samples by using a sampling method; aggregating the new samples by using a clustering method to form a plurality of representative typical feature vectors; and calculating cosine similarity between the feature vector of the to-be-detected power battery and each typical feature vector of each type, and identifying the health state of the to-be-detected power battery. By means of the screening method, the problem of overfitting of a traditional classification model under the condition of small samples and deviation caused by data imbalance are avoided, the practicability of power battery screening is improved, meanwhile, the typical characteristic conditions of the power batteries under different faults can be known more clearly, and a foundation is laid for better defining the states of the power batteries.

The invention discloses a server grouping method, which comprises the following steps of: analyzing a received server grouping request to obtain a target grouping task; respectively obtaining each target server parameter corresponding to the target grouping task in each server; and clustering and grouping the servers based on the target server parameters by using a fuzzy C-means clustering algorithm to obtain server groups. By applying the technical scheme provided by the embodiment of the invention, the influence of human subjective factors is greatly reduced, the grouping accuracy is improved, and the grouping efficiency is greatly improved. The invention further discloses a server grouping device and equipment and a storage medium, which have corresponding technical effects.

The invention provides a processing method of ginger processed pinellia. The processing method comprises the following steps: taking ginger slices and decocting; adding pure pinellia and immersing until the center of pinellia is immersed with the soup; adding alum and continually immersing for 48-72 hours; taking out and steaming; then cooling and slicing; and naturally drying or drying at a low temperature to obtain a finished product. The use amount of the ginger accounts for 25wt% of the pure pinellia and the use amount of the alum accounts for 6wt%-8wt% of the pure pinellia. According to the processing method, the use amount of auxiliary materials is reduced and medicinal materials are pre-treated and steamed, so that the use amount of the alum is greatly reduced, and the effects of detoxifying and removing irritation are guaranteed; effective components of pinellia are kept to the greatest extent; furthermore, the ginger processed pinellia processed by the method has a better finished product shape and high yield; modern equipment can be used for realizing standardized production and parameters of a processing process can be accurately controlled; and the influences of manual subjective factors are reduced, so that the quality of the product is more stable and controllable.