572results about How to "Accurate discrimination" patented technology

The invention discloses a driver fatigue monitoring device based on multivariate information fusion and a monitoring method thereof. The device comprises a digital signal processor, a camera head with a filter, a rectangular ultrared LED light source module mounted on a car windshield, a speaker module and a sensor module comprising an angular sensor and a pressure sensor, wherein the camera head, the rectangular ultrared LED light source module, the speaker module and the sensor module are respectively connected with the digital signal processor. The monitoring method comprises four procedures, namely eye characteristic identification, view line track, driving activity monitoring and fatigue characteristic judgment which merges the fatigue judgment results of the eye characteristic identification, the view line track and the driving activity monitoring to accurately judge the fatigue states of a driver. The invention utilizes an information merging technology to further develop and improve a fatigue driving detection technology, can objectively, rapidly and accurately judge the fatigue state of the driver in real time and avoids traffic accidents caused by fatigue driving.

The invention discloses a visual detection method of shockproof hammer defect detection. The visual detection method comprises the steps that denoising and anti-shaking preprocessing is performed on an aerial photographing image so as to obtain an original image to be detected; the existing original image is expanded by using the method of geometric transformation, scale change and contrast transformation so as to generate more data similar to the original image; samples are acquired, a shockproof hammer in the aerial photographing image is acquired and the size of the shockproof hammer is mainly acquired; a network model to be trained is determined, and the sample data are inputted to perform forward propagation and reverse propagation to adjust the weight so as to obtain the optimized detection network model parameters; the image to be detected is identified by using the trained model and the position of the hammer of the shockproof hammer is determined; and the lead in which the hammer is located is determined, and shockproof hammer defect discrimination is performed according to the relative position of the lead and the shockproof hammer and the constraints of respective directions.

A collision object discrimination apparatus for vehicles includes two collision detection sensors for discriminating kinds of a collision object based on two sensor outputs. The collision object is determined based on a collision width (W) detected by a collision width detection sensor fitted to a vehicle body. Alternatively, the collision object is determined based on a difference between operation characteristics of the two sensors. Further alternatively, the two sensors are integrated into a single unit including a plurality of censor cells.

There is provided a bio-signal detector to detect a heart rate and a respiratory rate so that the determination of a kind of a seat load placed on a seat can be made. That is, whether the seat load is a human or an object is determined, and if it is the human, then the physique size is determined to be small or large (the level of the physique size is determined). Accordingly, as compared with the means based only on a weight of the seat load, the determination result is highly reliable. Besides, with the use of the bio-signal detector, the discrimination between the human and the object is ensured. Moreover, a structure additionally including a weight sensor allows the determination of the physique size in more accurate manner.

An intelligent control method for platform screen doors is provided, which belongs to the field of automatic control technology. In the method, firstly the images of platform in train pull-in and pull-off states are collected by a CMOS camera and then outputted to a signal processor, the signal processor extracts feature information of the images and produces two basic templates and simultaneously establishes equations and indices for determining image threshold grey level of field image and basic template and error and correlation therebetween; then the filed images are collected by the CMOS camera and outputted to the signal processor, the signal processor processes and identifies the filed images and then produces an accurate control command and outputs the command to a controller; and finally the controller converts the received digital control command to an analog control command and the outputs to a driver, and the driver drives a servomechanism of the screen door after the amplification of the analog control command. The invention has high operational reliability and high identification accuracy, and can ensure the comfort and safety of pick-up and drop-off for passengers.

The invention relates to the field of computer vision and discloses a computer vision-based vehicle damage distinguishing method so as to solve a problem that a manual distinguishing approach adopted in conventional technologies is incomplete, inaccurate and inefficient in distinguishing operation. The computer vision-based vehicle damage distinguishing method comprises the following steps: in step a, a binocular image collection system is calibrated; in step b, the binocular image collection system is used for collecting images in a monitored area, and a depth map for the collected images is obtained; in step c, a convolutional neutral network is used for subjecting the depth map to characteristic extraction training operation, and a vehicle damage condition discrimination model is trained; in step d, the vehicle damage condition discrimination model is used for determining a damage degree based on collected vehicle images. The computer vision-based vehicle damage distinguishing method is suitable for distinguishing vehicle damage.

The invention provides a locomotive pantograph-catenary hard point photoelectric vibration comprehensive detection and GPS positioning method. The method includes steps: conversion from arc light to electric signals, analysis and processing of the electric signals, analysis of vibration signals, obtaining of a hard point value via calculation, and GPS accurate positioning. A locomotive pantograph-catenary hard point photoelectric vibration comprehensive detection and GPS positioning system mainly comprises a photoelectric sensor, an acceleration sensor, a signal conditioning device, an isolation transformer, an acquisition card, a control computer, a display, a memory, and a GPS positioner. According to the method and the system, acceleration detection of the hard point and photoelectric detection of the hard point are effectively combined, more accurate discrimination of the hard point can be made via double discrimination, accurate positioning of the hard point of a contact system is realized by employing an advanced GPS positioning module, the search and maintenance of faults are simple, and failure to adapt to the development requirement of high-speed rails and false and missed judgments can be avoided.

The invention discloses a method for generating an incident statement sentence material base. The method comprises the following steps of: converting an article into a set consisting of a plurality of long sentences; aiming at the converted long sentence set, identifying and extracting time points and extracting incident description verbs; identifying and extracting named entities of a personal name, a place name, a mechanism name and a product name from the long sentences obtained in the first step, extracting and marking element information comprising the occurrence time of an incident, theoccurrence place and the type of the incident to obtain a structural result; and extracting the original section of an incident statement sentence and the structural result, and storing into a database so as to generate the incident statement sentence material base. The incident statement sentence material base generated by the method can provide service such as updating, searching, inquiring andthe like in the Internet, and provides application such as writing, editing, subject making and the like for the media information field.

The invention discloses a pedestrian crossing passage traffic light intelligent control system. The pedestrian crossing passage traffic light intelligent control system comprises a pedestrian crossing passage traffic light control module used for controlling the switching of a pedestrian crossing passage traffic light and an automobile traffic light control module used for controlling the switching of the pedestrian crossing passage traffic light, and is characterized by further comprising an image acquisition module used for acquiring pedestrian waiting area images, and a time-waiting equivalent calculation module used for calculating and obtaining a pedestrian time-waiting equivalent according to the acquired pedestrian waiting area images, wherein the pedestrian crossing passage traffic light control module controls the switching of the pedestrian crossing passage traffic light according to the pedestrian time-waiting equivalent calculated by the time-waiting equivalent calculation module, or the automobile traffic light control module controls the switching of the pedestrian crossing passage traffic light according to the pedestrian time-waiting equivalent calculated by the time-waiting equivalent calculation module. The pedestrian crossing passage traffic light intelligent control system is capable of guaranteeing traffic safety, comforting the mood of the waiting people and avoids risky crossing behaviors by reasonably arranging and displaying the waiting time, and improving the crossing efficiency.

The invention discloses an identification method for network access equipment and an implementation system for the identification method. The method comprises the following steps that: a user is connected with and accesses a web service layer through the network access equipment, and software embedded into an application program page of a client acquires information about the network access equipment of the user, wherein the information comprises an equipment attribute, a state packet, a real Internet protocol (IP) address and a user behavior attribute; the network access equipment is connected with an established equipment fingerprint database, and performs retrieval to identify whether the network access equipment exists or not; the user enters an application layer for access or a specific transaction function, and a cloud computing cluster-based equipment fingerprint management platform integrates a machine identification and monitoring network and risk control, performs uniqueness identification on the network access equipment according to the network attribute of the equipment, and judges whether the network access equipment has an unconscious change or a conscious deceptive behavior or not; and the operation of the user over the network access equipment is identified and recorded, and a result is stored in the equipment fingerprint database. According to the technical scheme, the functions of network access equipment query, risk early warning, equipment tracking and the like are provided for the user by fully utilizing the network attribute and the high computing and judgment capability of a cloud computing cluster, roles of two parties of a transaction can be accurately identified, and the credibility of the transaction can be quickly judged, so that the safety of the transaction is improved.

The invention relates to a method for judging the road traffic state based on an annular coil of a signal lamp system. In the method, traffic historical data is acquired by using an annular coil detector and subjected to cluster analysis to obtain clustering centers corresponding to different traffic states; and a Euclidean distance between the real-time traffic data and each clustering center is directly calculated when the real-time traffic state is judged so as to judge that the current traffic state is the traffic state corresponding to the clustering center of which the Euclidean distance is the shortest.

A concentration sensor for detecting a concentration of a liquid reducing agent through heat transfer characteristics between separate two points is provided in a storage tank for storing a liquid supplied to an NO_x reduction catalyst. When the concentration read from the concentration sensor is 0% or less, the liquid is discriminated to be water, that is, it is utilized that when water is converted to the concentration of a liquid reducing agent, it is 0%, and it is also 0% even when the concentration is reduced by convection. When the concentration is greater than a predetermined value, it is discriminated that no liquid is present, since it is considered that the output of the sensor is abnormal because of the absence of a liquid in the tank. On the other hand, when the concentration is greater than 0% and less than a predetermined value, the liquid is discriminated to be the liquid reducing agent, provided that the concentration value may include an error.

A driving device for driving a semiconductor element can include a plurality of protection circuits that detect information necessary for performing protection operation for semiconductor elements that include a power conversion apparatus, a pulse signal generation circuit that sets a pulse sequence signal with a pulse width differently for each of the protection circuits and outputs a pulse sequence signal corresponding to the protection circuit that has been first detected a need for protection operation. The device can also include an alarm signal forming circuit that forms an alarm signal by ON-OFF control of a switching element and externally outputs the alarm signal, and a protection operation state discrimination circuit that discriminates existence of a protection operation state based on the alarm signal from the alarm signal forming circuit, and an alarm control circuit that delivers the pulse sequence signal outputted from the pulse signal generation circuit.

Method and system for controlling deployment of an airbag in which the position of an occupant to be protected by deployment of the airbag is determined, the probability that a crash requiring deployment of the airbag is occurring is assessed and deployment of the airbag enabled in consideration of the determined position of the occupant and the assessed probability that a crash is occurring. Deployment of the airbag may be enabled by analyzing the assessed probability relative to a pre-determined threshold whereby deployment of the airbag is enabled only when the assessed probability is greater than the threshold. The threshold may be adjusted based on the determined position of the occupant.

The invention provides a button differentiating device and a button automatically supplying device. The button differentiating device can accurately differentiate surface, back and defect of a button so as to select buttons even surface and back of the button have no clear concave and convex shapes or other characteristics. The invention has a button supplying part, a button moving unit and a button differentiating device. The differentiating device has a radiating unit, an area array camera, a brightness adjusting unit and a differentiating unit. The radiating unit supplies light to the button in three directions; the brightness adjusting unit respectively adjusts brightness of the lights in three directions and the area array camera takes photo of the button receiving the lights in threedirections. Because the lights in three directions form light and shade on characteristic part of the front face of the button, contrast of the image of the button is strengthened so that the characteristic part is clearer. The differentiating unit compares the image of the button with a differentiating reference image generated by pre-taking the button disposed in an expected direction so as todifferentiate surface, back and defect of the button.

The invention discloses multi-bridge static sounding equipment which comprises a static sounding rod and a force applying device, wherein a probe is arranged at the bottom end of a rod body; the force applying device is positioned above the static sounding rod and applies force to the static sounding rod; and a conical tip resistance sensor for measuring the resistance of a conical tip and a side wall friction resistance sensor for measuring the side wall friction resistance are arranged on the probe. The multi-bridge static sounding equipment is characterized in that the static sounding rod further comprises a microcamera for underground shooting and a plurality of sensors; the sensors are respectively used for measuring the pore water pressure, the temperature, the side pressure, the inclination, water content, the seismic velocity and seismic acceleration; and the sensors can be arranged in a connector and can be sequentially connected according to the requirement of test projects or can be directly and fixedly arranged inside the rod body of the static sounding rod or can be integrated on the probe to form the multifunctional probe. Due to the adoption of the technical scheme provided by the invention, one test process can be used for testing a large amount of projects so as to obtain abundant underground geotechnical information. The invention further discloses a multi-bridge static sounding test method.

The invention provides a processing method, a processing system and maintenance equipment for defects of a liquid crystal display panel. The processing method for the defects of the liquid crystal display panel comprises the steps of: acquiring defect images in a defect area on the liquid crystal display panel; analyzing the defect images and judging defect types; and searching corresponding maintenance schemes according to the defect types. The processing system for the defects of the liquid crystal display panel comprises an image acquisition module, a defect judgment module and a maintenance selection module. The maintenance equipment for the defects of the liquid crystal display panel comprises the processing system, a laser system, an electromechanical system and at least one controlhost computer. The processing method, the processing system and the maintenance equipment for the defects of the liquid crystal display panel can improve the accuracy of judging the defect types and improve the maintenance efficiency by analyzing the defect images, judging the defect types and then maintaining the liquid crystal panel according to the corresponding maintenance schemes.

The invention discloses a KCF-based improved anti-shielding target tracking method. The KCF-based improved anti-shielding target tracking method is based on a KCF algorithm and combined with a Kalmanfilter. The KCF algorithm has the advantages of fast tracking speed, high tracking precision and wide application in the target tracking field, but cannot cope with situations of scale changes and target large-area shielding. The KCF-based improved anti-shielding target tracking method adopts parallel computing to realize multi-scale target detection of a KCF, and designs an effective algorithm for determining target shielding. When determining that the target is severely shielded, the KCF-based improved anti-shielding target tracking method adopts a prediction result of the Kalman filter as acurrent frame target position. Experiments prove that the KCF-based improved anti-shielding target tracking method can effectively cope with the occurrence of tracking drift situation when the KCF algorithm encounters target large-area shielding, so that the KCF algorithm can track the target more precisely when encountering the shielding.

The invention provides a wafer surface defect mode detection and analysis method which can detect surface defects of crystal grains and judge distribution modes and defect mode causes of the defectivecrystal grains. The invention belongs to the field of defect detection in the wafer production and manufacturing process, and aims to solve the problems of high manual labor intensity, low detectionefficiency and the like in the existing defect detection method. The specific process comprises the following steps: acquiring a crystal grain image, and generating a wafer defect mode diagram througha machine vision method; constructing and training a wafer defect mode detection model and a classification model, the detection model being used for determining whether the wafer has a defect mode,and the classification model being used for determining a specific defect mode type; and finally, retrieving the marked sample most similar to the to-be-detected sample in a database according to a similarity measurement algorithm, and inferring the defect mode cause of the to-be-detected sample by analyzing the defect mode cause of the known sample.

The invention discloses a method for processing RFID (Radio Frequency Identification) traffic data. The method comprises the following steps of: arranging an RFID base station of each main road in a city, and sticking an RFID environment-friendly electronic card on each access vehicle. The method specifically comprises the following steps of: by using an adjacency list model as a basis, the RFID base stations as nodes, and paired RFID base stations as base station pairs, forming a road net structure model by using the nodes and the base station pairs; and acquiring information of each access vehicle by each RFID base station to form real-time access data, caching the access data into a memory database, then matching the real-time access data, abnormally processing and regularly counting on matching results to generate real-time traffic parameters and storing results in a data base. According to the invention, the RFID access data is processed through road net modeling, and then accurate traffic parameters are obtained, and the thus a firm foundation is provided for the subsequence accurate traffic management.

Embodiments of the invention provide a telecommunication fraud recognition method and data processing equipment, and aims at solving the technical problem that electronic equipment is relatively low in fraud event recognition rate in the prior art. The telecommunication fraud recognition method comprises the following steps of: obtaining user behavior information from terminal equipment connected with the data processing equipment, wherein the user behavior information is used for indicating an operation behavior carried out on to-be-assessed information in the terminal equipment by a first user; processing the user behavior information on the basis of at least one assessment index so as to obtain an assessment probability, wherein the at least one assessment index is determined on the basis of a history fraud event and is used for representing sensitiveness, for the history fraud event, of the first user, and the assessment probability is used for representing a matching degree between an operation indicated by the user behavior information and a history operation related to the history fraud event; and if the assessment probability is greater than a preset probability, determining the to-be-assessed information as telecommunication fraud information.

The invention discloses a medical image classification method based on collaborative deep learning so as to solve the technical problem of poor classification accuracy of an existing medical image classification method. The technical scheme is characterized in that the method adopts a collaborative learning method between two deep convolution neural networks to carry out training in a mode of learning in pairs; each time, a model receives an image pair as input, and one pair of images are transmitted to the corresponding deep convolution neural networks respectively; the deep convolution neural networks are subjected to initialization and training through a pre-training model fine tuning method; a collaborative learning system is designed to allow the two deep networks to realize collaborative learning; and the collaborative learning is used for monitoring different or same attributes of the image pairs, that is, judging whether the image pair belongs to one category, and carrying out back propagation on collaborative errors generated by the two deep convolution neural networks in real time, and collecting network weight, so that the method can further enhance network learning feature representation capability, and can make an accurate judgment for easily-confused samples more effectively.

The invention discloses an expressway congestion state judgment and traffic flow management and control system and method, and belongs to the field of intelligent traffic management and control. The system comprises a data acquisition module, a data processing module, a back-end service module and a front-end display module. According to the invention, the road congestion state is divided into three types, i.e., no congestion, no serious congestion and more serious congestion, and the traffic congestion index is calculated in combination with the traffic parameters, such as speed, occupancy and traffic flow, acquired by a system detector; the calculated data are compared with a specific threshold value to obtain an initial pre-judgment result; in order to reduce the misjudgment rate and realize accurate judgment, the system maintains the continuity of detection and carries out pre-judgment for multiple times, and finally completes one-time traffic congestion state evaluation; and finally, vehicle management and control are carried out on an entrance ramp by adopting a hierarchical flow control method so as to relieve the traffic pressure of an expressway main line. According to the method, the traffic congestion state of the expressway can be accurately pre-judged, so that hierarchical flow control is better realized.