50 results about "Validation test" patented technology

The invention relates to the technical field of depth learning and computer vision, in particular to a human posture estimation method based on a deformable convolutional neural network. The method comprises the following steps of 1, preprocessing an input dataset, wherein expanding and data set picture cutting are included; 2, performing K-means clustering on training set pictures; 3, adopting the K-means clustering result as a label, and inputting a training set into the deformable convolutional neural network to be trained; 4, constructing a score function F for training an SVM for joint classification; 5, performing outputting treatment. By reading the positive data sets, some data sets are adopted as the training set, some data sets are adopted as the validation set, negative data isread for comparison, and the negative data set is introduced for training, so that the background can be effectively filtered away to obtain interference, and the recognition rate is increased.

The disclosed embodiments provide a system for performing multi-tenancy distribution of a graph database cache. During operation, the system obtains a cache of a set of records in a graph database storing a graph, wherein the graph includes a set of nodes, a set of edges between pairs of nodes in the set of nodes, and a set of predicates. Next, the system identifies one or more records in the cache that fail a validation test. The system then merges a remainder of the cache with online updates to the subset of records to produce a merged version of the cache. Finally, the system distributes the merged version to one or more instances of the graph database for use in processing queries of the graph database.

Apparatuses, methods, and systems pertaining to the verification of portable consumer devices are disclosed. In one implementation, a verification token is coupled to a computer by a USB connection so as to use the computer's networking facilities. The verification token reads identification information from a user's portable consumer device (e.g., credit card) and sends the information to a validation entry over a communications network using the computer's networking facilities. The validation entity applies one or more validation tests to the information that it receives from the verification token. If a selected number of tests are passed, the validation entity sends a device verification value to the verification token, and optionally to a payment processing network. The verification token may enter the device verification value into a CVV field of a web page appearing on the computer's display, or may display the value to the user using the computer's display.

A modular vehicle assembly system and methods for increased flexibility and adaptability of a high volume assembly facility which builds several vehicle models. In one example, the invention includes modular assembly equipment (AE) support pallets which are shipped to vendors for installation of selected AE equipment devices that are specific to a predetermined assembly operation and then validation tested prior to shipment. The modular AE pallets and AE devices are quickly installed and easily removable for maintenance or replacement. In other examples, modular AE support platforms and safety fencing are used to support ground level assembly operation and safer working environment.

A method and system for allowing access to computer functions such as websites that utilizes a user's ability to recognize sounds is described. The method presents a user a series of sounds. Some of the sounds presented in the series are labeled as validation sounds. The user is asked to provide an input every time he or she hears the validation sound. The user must identify the sound within a specified length of time. The system disclosed comprises a user interface, a sound database module, a generation module, and a sound database module. The generation module creates the validation test file and expected answer. The answer confirmation module checks the input from the requesting computer and provides access to the computer function if the computer input from the requesting computer meets the required parameters.

The invention discloses a GPU-based modeling method for automatic machine learning. The method comprises the steps of establishing a model library; preparing data; conducting the normalization treatment on the feature part of the input data; selecting a plurality of models in the common mode to subject the models to the GPU parallel computation; automatically preparing data for cross verification according to the input data, wherein the default is 10 folds: 9 folds of training and 1 fold of verification test; respectively calculating the cross verification for the plurality of models through the GPU to obtain the errors of all the models respectively; selecting one model smallest in error as a finally selected model; adopting all data as training data; training the finally selected model to obtain a final model, and marking the final model as Ypre = f (x). In this way, an optimized model is automatically calculated and obtained for a common user. Therefore, the complexity of the machine learning application for the common user and the excessive dependence of the common user on experts are reduced.

The invention provides a kit for evaluating prognostic risks of breast cancer. Big-data clinical validation test results show that 12 miRNAs are combined to be used as a molecular marker, and the molecular marker has an outstanding advantage on the aspect of evaluation of prognostic risks of ER/PR(+)HER2(-) breast cancer. The molecular marker formed by combining the 12 miRNAs is firstly applied to development of the miRNA detection kit of ER/PR(+)HER2(-) breast cancer. The breast cancer miRNA detection kit can realize prognosis evaluation of ER/PR(+)HER2(-) breast cancer on the basis of a real-time fluorogenic quantitative PCR method, and can screen patients with high relapse risks, guide clinical individualized treatment and improve treatment accuracy.

The invention relates to a rice detection and classification method based on deep multi-view feature. The method comprises the following steps: 1) division of a data set: dividing a rice drop image data set through a cross-validation method to obtain a training set and a test set; 2) construction of a network model: carrying out feature extraction on an input image through a deep learning method,and meanwhile, continually updating model parameters through stochastic gradient descent; and 3) model analysis and performance evaluation: inputting the test data into a trained rice classifier, andcomparing probability predicted value and classification threshold to obtain a rice detection and classification result. Compared with the prior art, the method is greatly improved in accuracy, and improves defects of rice feature extraction of a conventional method; and the method finishes calculation of husked rice yield of the rice according to the classification result.

The invention relates to the technical field of mechanical control, and particularly relates to a method for verifying a center deviation position of a dynamic wafer in an AWC (Active Wafer Centering) system. The method comprises the following steps that: (1) a manipulator moves to a demonstration position above a station, an initial target sphere position is recorded by utilizing a laser tracker, and simultaneously, a current joint value on a demonstration box is recorded; (2) the manipulator drives the wafer to carry out translation, and a joint value on the demonstration box after translation is recorded; (3) the manipulator retracts and lifts the wafer to the demonstration position above the station again, and a new target sphere position is recorded; (4) an offset of a wafer center is detected by utilizing an AWC function; (5) a coordinate system is set up; and (6) according to the coordinate system, wafer deviation results obtained by display of the demonstration box, AWC detection and actual measurement of the laser tracker are respectively calculated and are compared. According to the method disclosed by the invention, the problem of translation of the wafer in the verification test is solved, no instrument or equipment is additionally added to operate the wafer, and verification test efficiency is improved.

A video brokerage system provides a subscriber of a cable system access to a program in response to the subscriber entering a coupon code, where the program would otherwise not be viewable, or a discount is provided for viewing the program. In one embodiment, a coupon system interacts with a computer system operated by a merchant, which distributes the coupon code to the subscriber when acting as a customer of the merchant. The subscriber then provides the coupon code to the cable service provider when selecting a video-on demand program, allowing the subscriber to either receive a discount on a pay-per-view movie, or accessing a program that otherwise would not be accessible. Various validations tests are defined in conjunction with the cable service provider processing the coupon code, including checking: an expiration date of the coupon code, rating qualifiers, content provider qualifiers, etc.

The invention provides a microgrid semi-physical testing circuit and method on the basis of a feedback type load simulator. A semi-physical way means that tested objects are material objects and load for testing is simulated. By means of a characteristic that the feedback type load simulator simulates constant-impedance load, nonlinear load and load of an asynchronous motor to substitute for actual constant-impedance load, nonlinear load and load of the asynchronous motor, dynamic characteristics and power quality of a main power source of an energy storage system of a microgrid can be verified and tested. During testing, power balance between a load characteristic simulation input end of the feedback type load simulator and a unity power factor feedback converter is maintained, and load characteristic simulation and energy feedback are realized. The problems of many load types, high cost, energy waste of feedback and reduction of long-time running adjustment accuracy when dynamic characteristics and power quality of a main power source of an energy storage system of a microgrid are verified and tested are solved.

The invention discloses a hydraulic motor overhaul testing method. A hydraulic overhaul operation table is utilized, and overhaul selected test items comprise a displacement validation test, an efficiency test, an impact test, a variable property test, a pressure oscillation examination, a self-sucking test, a high temperature test, an overspeed test, an overload test, a full-load test and an outward leakage examination test. By selecting the test items, a hydraulic axial plunger pump, a plunger motor, a hydraulic gear pump and a hydraulic blade pump can be overhauled on the hydraulic motor overhaul operation table through the method, compatibility of overhaul equipment is achieved, the operation method is simple, test performance is stable, and overhaul quality and maintenance efficiency can be greatly improved.

The invention discloses an antenna impedance matching method on the basis of key index data of a power amplifier. The method mainly comprises the steps of calibrating a vector network analyzer; measuring output impedance value of an antenna; acquiring an impedance matching network value through Smith-Chart software; creating an impedance matching network; acquiring testing data of the created impedance matching network; verifying whether the testing data meet the index requirement; comparing the impedance matching network with the key index data of the power amplifier; determining whether the created impedance matching network reaches the optimal state; if not, debugging the impedance matching network; if so, passing the debugging. With the adoption of the antenna impedance matching method on the basis of the key index of the power amplifier, the accuracy can be ensured on the premise that an intuitive and physical conception is clear, thus the calculation speed is greatly raised, the performances of the antenna are optimized, and meanwhile, the shortages of computer simulation method and the manual calculation method are overcome, and as a result, the performances of the antenna are further optimized.

The invention belongs to the technical field of structural strength of airplanes and relates to a ground validation method of a main load-bearing structural member of a fuselage under large load. The method comprises the steps: 1.1) analyzing the deformation situation of a transverse beam 1 in the fuselage; 1.2) simplifying the support of the fuselage to the transverse beam 1 and forming the scheme of primary simulation support structure; 1.3) analyzing the deformation situation of the transverse beam 1 under the primary simulation support structure; 1.4) matching the deformation situations of a test sample before simplification and obtaining the parameters of a final simulation support structure; 1.5) performing the manufacturing the final situation support structure and a test fixture and performing ground validation test. According to the ground validation method of the main load-bearing structural member of the fuselage, the scheme of the test sample (transverse beam 1) support structure is simplified, the usage of auxiliary parts in the test is reduced and the experimental cost and period can be saved greatly.

The invention provides a modular AOI (automated optical inspection) positioning method, a modular AOI system and burning IC (integrated circuit) equipment. The burning IC equipment is subjected to IC positioning in a modularization mode, and the modularization refers to the operation that each module for controlling an IC positioning system is packed into a dynamic command library through a certain software development platform, and the positioning operation of IC burning is controlled through calling an interface function of the dynamic command library. The system has the advantages that the multi-direction calculation function is adopted, the detection speed is accelerated, and in addition, the detection precision is improved by adopting a standard image correcting method; meanwhile, through the modular development framework, the verification test time in the development process is also reduced, and the product development speed is accelerated.

The invention provides an optimal configuration method for test resources of a spacecraft. The method comprises the following steps: 1, utilizing a method of weighted assignment to primarily determine testing sets of a product; 2, designing a multi-single model of the product and generating matrices related to tests of the product; 3, verifying whether test sets can satisfy the requirements for fault monitoring and diagnoses or not, adjusting value affecting factors in the method of weighted assignment if the test sets do not satisfy the requirements and repeating steps 2 and 3 till the requirements can be net; 4, designing an optimal target for test resources; 5, utilizing an optimal distribution method for optimal configuration of test resources; 6, verifying whether optimized test resources can satisfy the requirements for fault monitoring and diagnoses or not, taking an optimal test set as a complete test set of the product if the optimized test resources can satisfy the requirements for fault monitoring and diagnoses, or otherwise repeating steps 5 and 6 till the complete test set is obtained. The optimal configuration method for test resources of the spacecraft has the following beneficial effects: the engineering experience and theory guidance are integrated so that the repeated design during production realization is avoided and waste of test resources is avoided.

The invention relates to a browser fuzzy test method based on AST mutation. By parsing a validation test sample of vulnerabilities in older browsers into an AST syntax tree, new AST syntax tree is obtained after AST syntax mutation, and by reparsing to generate new validation test samples, the browser loads the new validation test samples to monitor the browser's running. If the browser crashes, the validation test samples at the crash time are collected and uploaded to the WEB server, and the validation test samples at the run time are deleted. Otherwise, the validation test samples at the run time are deleted. The method of the invention can mutate the original verifiability test sample, regenerate the new verifiability test sample, let the browser run, monitor the state of the browser,collect the verifiability test sample that causes the browser to collapse, thereby greatly improving the efficiency of the browser fuzzy test and the efficiency of the vulnerability mining.

The invention discloses a test case verification method, which is applied to the technical field of tests, mainly aims to solve the problem that the existing test case verification consumes much time.The method mainly comprises the following steps of: determining a target test case which is a test case needing to be verified, obtaining a test condition set corresponding to the target test case according to the target test case, combining according to the plurality of test conditions to generate a verification rule, and generating a verification rule set according to the plurality of verification rules, constructing a case verification table according to the test condition set and the verification rule set, the case verification table comprising all test conditions and all verification rules in the target test case, and determining a verification result of the target test case according to a judgment result of each test condition in the case verification table based on verification under a verification rule. The method is suitable for verifying the test cases.

An embodiment of the invention provides a three-dimensional channel model spatial correlation verification method and system. According to an acquired three-dimensional radiation pattern of a reference measurement antenna in a multi-probe darkroom, a receiving antenna is determined as a spatial correlation factor between an optional pair of spatial positions of the reference measurement antenna. According to the three-dimensional channel model spatial correlation verification method and system, the influence of the three-dimensional radiation pattern of the reference measurement antenna on channel verification testing results is taken into consideration, misjudgment of system channel simulation results is avoided, and three-dimensional channel model verification conclusions are more accurate.

A switch-operated therapeutic agent delivery device. Embodiments of the operated therapeutic agent delivery device my include a switch that can be operated by a user, a device controller connected to the switch through a switch input where the device can actuate the device when certain predetermined conditions are met, following performance of both a digital switch validation test and an analog switch validation test. The switch operated therapeutic agent delivery device may have two parts, which are assembled by a user prior to use. These devices may be configured to determine if a current is present between the anode and cathode when drug is not intended to be delivered by the device. These devices may indirectly control and/or monitor the applied current without directly measuring from the cathode of the patient terminal.

The invention discloses a transient characteristic validation device for a DC-related circuit. The device comprises a master control unit, a signal processing unit, a power amplification unit, and anacquisition unit, wherein the acquisition unit is used to acquire signals output by the power amplification unit to obtain a signal waveform of the output signals, and is also used to compare the acquired signal waveform, which is output by the power amplification unit, with a transient waveform stored in the master control unit to obtain a validation result. The device disclosed by the inventionhas the characteristics that advantages and disadvantages of a linear power amplifier and a switching power amplifier are fully considered, independent output ports are designed according to differentrequirements of actual output current amplitudes and frequencies, and automatic adjustments can be carried out according to actual onsite demands; and the device is small in size and light in weight,and can be simply maintained and conveniently operated, so as to be highly applicable to a transient characteristic validation test of DC measuring equipment on an installation and operation site.