36results about How to "Less human intervention" patented technology

A method for improving a scoring process includes displaying an original scanned image of a hard copy page of an answer to a scorer. If the scorer determines that the image is of insufficient quality to read, an electronic request for an improved image is received from the scorer and routed electronically to an archive containing the hard copy page, from which the page is retrieved. The retrieved page is rescanned to form a new image, which is then electronically transmitted to the scorer for scoring. A system includes hardware and software elements, including a processor and a database accessible by the processor containing original scanned images of answer pages. A scorer display and an input device are located at a scoring site, and an archive display and a scanner at an archive site. Software as described above is resident on the processor for implementing the method steps.

The invention discloses a precision testing platform for a small-sized part. The precision testing platform comprises a standard platform plate, a frame body, an inclined amplifying mechanism and leveling mechanisms, wherein the standard platform plate is horizontally supported by a plurality of support columns of the frame body through elastomers; the inclined amplifying mechanism is attached to the bottom surface of the standard platform plate, supported by the leveling mechanisms, mainly comprises a hard tray, a flexible capsule and a balance steel ball, and is used for exerting leveling gravity on the leveling mechanisms; the leveling mechanisms are arranged between the support columns of the frame body and the inclined amplifying mechanism according to a lever principle, and used for performing automatic reset calibration on the standard platform plate which is unbalanced and inclined. The precision testing platform can automatically reset the standard platform plate, which is unbalanced and inclined, on the frame body to a level balance state, and level calibration action is efficient, fast and convenient.

The present disclosure relates to a cavity filter, a connector and manufacturing processes thereof. The cavity filter comprises a cavity, a cover plate and a connector disposed on the cavity or the cover plate; an end of the connector is connected with internal devices inside the cavity filter and the other end of the connector is connected with external communication devices; and the connector comprises an inner conductor and a metal enclosure disposed coaxially and an insulation medium disposed between the metal enclosure and the inner conductor, and a non-metal layer is disposed on an outer peripheral surface of the metal enclosure. The connector of the present disclosure is formed with a non-metal layer on the outer peripheral surface of the metal enclosure thereof, which can improve the moisture-proof capability, the salt-mist-proof capability, the mould-proof capability and the reliability of the connector and the cavity filter.

The invention discloses an unmanned aerial vehicle abnormal behavior recognition method based on a convolutional neural network. The unmanned aerial vehicle abnormal behavior recognition method comprises the steps: training data and prediction inspection, wherein the training data carries out the feature extraction of flight history data of an unmanned aerial vehicle as an input parameter, importsthe input parameter into a constructed CNN model, and carries out the training; and after the training is completed, new unmanned aerial vehicle flight data is imported into the CNN model to obtain ajudgment result, thereby performing prediction inspection. The unmanned aerial vehicle abnormal behavior recognition method has the advantages that 1, the recognition accuracy is much higher than that of a conventional method; 2, aiming at the conventional illegal problem caused by manual operation of an operator, the illegal problem can be well detected; 3, the system is fully automatic, and thedegree of human intervention is greatly reduced;

The invention discloses a seed activity detecting device. The device comprises a rotatable turnplate, an activity detecting mechanism and a control system, wherein the turnplate is used for driving to-be-detected seeds to rotate to a preset detecting position, so that the activity states of the to-be-detected seeds are detected; the activity detecting mechanism is matched with the turnplate and used for obtaining spectral information data of the to-be-detected seeds on the turnplate; and the control system is connected with the turnplate and the activity detecting mechanism, and used for controlling the turnplate to rotate and controlling the activity detecting mechanism to start and stop. The invention further discloses seed sorting equipment comprising the seed activity detecting deviceand a seed sorting method. According to the seed activity detecting device, on the premise that the seeds are not damaged, the activity of the seeds can be effectively recognized, the seeds can be stored easily after being detected, and meanwhile, the problems that manual recognition is low in efficiency and low in recognition success rate can be effectively solved.

The invention relates to a plug and play sensor monitoring system and an acquisition unit. The acquisition unit comprises an FPGA processor and a memory. The FPGA processor is connected to the memory.The FPGA processor is also provided with an acquisition interface used for connecting a sensor and a communication interface used for communicating with a monitoring unit. The memory is used for storing the self-described information of the sensor connected to the acquisition interface. The communication interface is used for sending the self-described information of the sensor to the monitoringunit. In the invention, the acquisition unit stores the self-described information of the sensor and sends the self-described information to the monitoring unit. The monitoring unit receives and stores the self-described information of the sensor. According to the self-described information of the sensor, the sensor, the acquisition unit and the monitoring unit are matched with each other. The plug and play of the sensor is realized, the expansibility of the system is increased, and problems that a configuration work amount is large, efficiency is low and an error is easy to generate are solved.

The invention discloses a one-object-one-code-based user consumption behavior automatic clustering method and an application system, and relates to the field of data mining, and the method comprises the following steps: obtaining user code scanning data, and carrying out the preprocessing of the data, and generating first data; labeling the first data by using a labeling algorithm, performing automatic group division by using a clustering algorithm in combination with consumer historical data, and summarizing other behavior records of the user to generate second data; storing the second data, and portraying the user according to the second data; and according to the user portrait, actively touching the user and/or passively touching the user, and determining a marketing scheme when the user scans the code next time. The system further comprises a data acquisition module, a data processing module, a label and clustering module, a data storage module, a user portrait display module and a precision marketing module. The method has the advantages of being high in matching degree with user hobbies, good in recommendation effect, high in recommendation accuracy, good in user experience and the like.

A system and method for disrupting an information security threat that constitutes an attack on a computer asset in a computer network is provided. The provided system and method disrupts this information security threat after the attack on the computer asset has been detected by at least one of the monitoring devices on the affected computer network. An intermediate upstream gateway of the affected computer network is then utilized to disrupt this information security threat. As the detected attack is being disrupted, a mitigation action will be automatically initiated if a mitigation action associated with the attack is stored in the system's database; else information about the attack will be sent to a central command centre for further assessment. At the central command centre, a mitigating action will be further developed and executed to address the intention of the attack.

A method for cutting a PVC board based on a Mark point position function features that an industrial camera is installed near that main axis of engraving and mil machine, and the relative deviation between the camera and the main axis is calculated by the function of biasing cutter of the camera. The engraving and milling machine formulates the strategy of searching Mark point according to the cutting procedure file; By searching Mark point strategy, four pairs of Mark point center coordinates and the current spindle coordinates fed back when the control board finds Mark point are obtained byindustrial camera respectively, and the coordinates of current camera coordinate system are converted into the coordinate system of cutting program file according to the Formosan rock; and the coordinate system of current camera coordinate system is converted into the coordinate system of cutting program file by industrial camera. The coordinates of the four converted Mark points are compared withthe coordinates of the Mark points in the coordinate system of the cutting program file. If no offset occurs, the position of the PVC material is normal. If there is an offset, the position of the PVC material is corrected. The method of the invention can improve the running efficiency of the machine.

The invention relates to a transverse wave velocity extraction method for seabed shallow sedimentary layers and a processing terminal. The method comprises the following steps: step 1, generating a plurality of horizontal stratified geological models of the seabed shallow sedimentary layers; step 2, simulating each horizontal stratified geological model by adopting a seismic wave field numerical simulation method to obtain corresponding OBS data including a vertical component and a horizontal component; step 3, preprocessing the horizontal component of the OBS data, including performing time correction on the OBS data to obtain preprocessed OBS data; step 4, combining the OBS data and the transverse wave speed into label data, and inputting the label data into a neural network for training; and step 5, acquiring and preprocessing actual OBS data, inputting the actual OBS data to the neural network for processing, and multiplying an output result by a normalization factor to obtain a final transverse wave velocity. Manual intervention components are few, the calculation amount is small, and the transverse wave speed accuracy is higher.

Disclosed is direct ink write (DIW) print extrusion head for 3D printing of viscous elastomers. The disclosed print extrusion head comprises a mixer assembly, comprising a fluid distribution cap coupled to a carrier, an in-line mixer coupled to the fluid distribution cap. A cooling jacket surrounds the in-line mixer. A nozzle is coupled to the in-line mixer and protrudes below the cooling jacket over a work surface. The position of the nozzle relative to the work surface is changeable. At least one heat source is on the chassis and disposed adjacent to the fluid distribution cap. The at least one heat source comprises a heat guiding element to direct heat to a region onto the work surface below the nozzle.

Described is a method for the production of a moisture curable composition whereby the reactive end groups of a reactive polymer are reacted with a cross-linking agent for forming a prepolymer, whereby the method is operated in batch mode and the production of the prepolymer occurs in a mixing vessel, characterized in that the reactive polymer and the cross-linking agent, as well as the additional ingredients necessary for the production step in the mixing vessel, are introduced into the closed mixing vessel and react without opening the mixing vessel, and the mixing takes place without opening the mixing vessel, and the product is removed from the mixing vessel without opening the mixing vessel.

The invention relates to the field of education informatization, and discloses a multi-level academic score identification method and system. Analyzing and identifying the declaration content and thedeclaration proof material based on a predetermined standard; and calculating student scores according to the declared content and the analysis and determination result of the declared proof material.According to the method, automatic recording and evaluation are carried out on different levels of expansion activities, the manpower input of manual record and calculation of academic scores is reduced, the condition of human intervention is reduced, and the credibility of a determination result is high.

The embodiment of the invention discloses a method for charging unlicensed vehicles entering and exiting a parking lot and a related device. The method and device are used for charging parking fees for the unlicensed vehicles. The method comprises the steps that a cloud platform receives an exit request sent by user equipment, wherein the exit request comprises a parking lot number; the cloud platform determines a parking lot management terminal according to the exit request; the cloud platform sends exit information of the user equipment to a parking lot management terminal through an integrated platform, so that the parking lot management terminal calculates parking fees according to entrance information and exit information of the user equipment, the entrance information is informationrecorded by the cloud platform when the user equipment enters the parking lot, and the entrance information is sent to the parking lot management terminal by the cloud platform through the integratedplatform; the cloud platform receives parking fees sent by the parking lot management terminal through the integrated platform; and the cloud platform generates a payment channel for the user equipment.

The invention belongs to the technical field of electrical element manufacturing, and discloses surface-mount-packaged laser diode manufacturing equipment which comprises an electrical equipment workbench and an electrical equipment part body. An electrical equipment part conveying assembly used for conveying electronic elements is arranged in the middle of the top end of the electrical equipment workbench; at least one electrical equipment part welding machine is arranged at the top end of the electrical equipment workbench and located on one side of the electrical equipment part conveying assembly. The automation of electronic component manufacturing is greatly improved, the top of the electrical equipment workbench is provided with at least one electrical equipment part manufacturing assembly, the electrical equipment part manufacturing assembly has a double-steering effect, automatic angle alignment can be achieved through an electrical element first-stage motor and an electrical element second-stage motor, and the manufacturing efficiency is improved. After the electrical element clamping structure clamps the electrical equipment part main body, the rotation angle of the electrical element primary motor can be reversely rotated by the electrical element secondary motor.

The invention provides a deep-water underwater oil leakage collection-storage device. The deep-water underwater oil leakage collection-storage device is composed of an underwater oil leakage collection cover, an underwater automatic bag supply rack, a sewage pump, soft oil storage bags, fixed self-sealing joints, movable self-sealing joints, a self-sealing joint docking clamping seat, a docking-and-releasing actuator, a capacity detection circuit, and a release control circuit. The underwater oil leakage collection cover is provided with an oil-water interface instrument and a deflation sealing liquid valve, and a driving mechanism self-sealing joint is arranged at one end of the underwater automatic bag supply rack; the docking clamping seat is provided with a locking-and-releasing movable assembly, the capacity detection circuit comprises an independent power supply which does not rely on external connection, and the release control circuit comprises a proximity switch; a suction port of the sewage pump communicates with a drain outlet of the underwater oil leakage collection cover through a pipeline, the bottom ends of the fixed self-sealing joints are rigidly connected with anoutlet of the sewage pump, and the top ends of the fixed self-sealing joints are clamped in the self-sealing joint docking clamping seat, one side of the self-sealing joint docking clamping seat is connected with the underwater automatic bag supply rack, clamping grooves of the self-sealing joint docking clamping seat and the underwater automatic bag supply rack are smoothly aligned to each other,the locking-and-releasing movable assembly of the self-sealing joint docking clamping seat is mechanically connected with the docking-and-releasing actuator, the soft oil storage bags are bonded to the complete capacity detection circuits and connected with the movable self-sealing joints, and the soft oil storage bags are clamped into the clamping grooves of the underwater automatic bag supply rack through movable self-sealing joints connected to the soft oil storage bags. The invention further disclosed an application method of the deep-water underwater oil leakage collection-storage device.

The invention relates to the field of data processing, in particular to a computer-based data picture labeling method, system and device. The method mainly comprises the steps of obtaining a pluralityof to-be-reviewed pictures needing to be subjected to data labeling; sequentially encoding the plurality of to-be-audited pictures to form a slice task queue, and performing a slice task on the to-be-audited pictures according to an encoding sequence to obtain a plurality of to-be-classified small pictures; sequentially encoding the small pictures to be classified to form a classification task queue, classifying the small pictures to be classified according to an encoding sequence, and outputting a classification result; The classification result output by the classification task is audited;and outputting the picture for data annotation and the corresponding annotation information. By means of the data picture marking method, system and device, marking tasks of a large number of data pictures can be completed accurately and efficiently.

The invention discloses a double-layer orderly arrangement machine. The double-layer orderly arrangement machine comprises a machine frame, a pair of transverse conveying mechanisms, a pair of longitudinal conveying mechanisms, a transverse pushing mechanism and a longitudinal pushing mechanism, wherein the pair of transverse conveying mechanisms are adjacently arranged in the vertical direction and each comprise a conveying assembly and a counter; the pair of longitudinal conveying mechanisms are located on the downstream portion of a pair of transverse acceleration conveyor belt and oppositely arranged in the vertical direction, and each longitudinal conveying mechanism comprises a longitudinal conveying assembly and a longitudinal intercepting telescopic rod; the transverse pushing mechanism is located on the downstream portion of the longitudinal intercepting telescopic rods and comprises a pair of limiting parts, a transverse power part and a pair of transverse pushing part sequentially arranged in the material conveying direction; and the longitudinal pushing mechanism is located on the opposite side of the transverse pushing mechanism and comprises a packaging table, a longitudinal power part and at least one longitudinal pushing part. The invention further discloses an application method of the double-layer orderly arrangement machine. Through the double-layer orderly arrangement machine and the application method thereof, the packaging efficiency and the packaging effect can be improved at the same time, the automation degree is high, the production cost is reduced, and the production benefits are increased.