1295results about How to "Realize real-time detection" patented technology

The invention discloses an ATM security certificate and a pre-warning method based on face recognition. The invention introduces face recognition and video detection into the applications of ATM and bank card business for the first time, which can obviously improve the security of ATM financial system, effectively reduce and overawe various ATM fraud and criminal activities. The complete system operation process in the method of the invention is: firstly, registering the face information of a holder at a service window which arranges an infrared face acquisition equipment or with the registration by the users themselves on ATM in bank, and saving the encrypted face information to a bank database and a bank card. Then, determining whether the ATM operator to be recognized is the holder of the bank card or not through verifying the user's identity by a face recognition engine; determining and recognizing whether existing behaviors that threaten ATM and the users though a behavior monitoring engine; and determining whether to trigger corresponding pre-warning mechanism or not according to the recognition result. The invention can extend to the applications of bank book business, public security monitoring and pursuing escaped criminals, etc.

The invention discloses a senile dementia monitoring system based on a healthy service robot. The senile dementia monitoring system based on the healthy service robot comprises the healthy service robot, an intelligent terminal and a cloud server. The healthy service robot comprises a robot body, a main control unit, a human-computer interaction unit and a medical detection unit; the human-computer interaction unit is connected with the main control unit and comprises a tablet computer, and the tablet computer is arranged in front of the chest of the robot body; the medical detection unit is connected with the main control unit and comprises a brain electrical detection device which is independent of the robot body, and the brain electrical detection device is connected with the intelligent terminal and the tablet computer through Bluetooth signals; the intelligent terminal and the tablet computer are connected with the cloud server through the mobile Internet, and data interaction between the intelligent terminal and the tablet computer is achieved through wireless signals. By means of the senile dementia monitoring system based on the healthy service robot, automatic auxiliary diagnosis and treatment of senile dementia can be achieved, the diagnostic accuracy is improved, prevention and early detection of the senile dementia are facilitated, the pathogenetic condition is mitigated from exacerbating, and the purpose of healing is achieved.

A laser liquid-level measure of floater type consists of screen laser-logging, fixing device of said logging, floater, balance weight and laser distance measurer of reflection type.

The invention discloses a safety helmet real-time detection method based on a convolutional neural network. The method comprises the following steps: decoding a video in a multi-thread I frame extraction mode; realizing end-to-end safety helmet and face real-time detection through a YOLOv3 convolutional neural network algorithm. Provided is a method for judging whether to wear the safety helmet ornot based on the safety helmet and human face real-time detection result. For a person who does not wear the safety helmet, face recognition is realized through a FaceNet algorithm so as to carry outintelligent voice reminding on the person. And a channel pruning and quantifying method is adopted to compress the model, so that the integration of the model on the SoC system is more facilitated. According to the method, the respective advantages of YOLOv3 and FaceNet are combined, the problem of accurate end-to-end detection of the safety helmet worn in real time in the video stream is solved,real-time detection of wearing of the safety helmet in the video stream is realized, and whether a worker wears the safety helmet according to regulations or not can be monitored in real time to reduce the safety risk.

The invention discloses a face detection method and device adopting a cascade classifier, relating to the field of video monitoring. The face detection method comprises the following step of: carrying out face detection on rectangular windows with different positions and sizes in a current image to be detected by adopting a light type cascade-structure classifier to detect face windows, wherein each level of the light type cascade-structure classifier is a weak classifier. For a large quantity of windows to be detected in one image, relative to a scheme that a cascade-structure classifier is bound to eliminate partial non-faces after computing the results of all weak classifiers of a strong classifier, each weak classifier of the light type cascade-structure classifier can eliminate the partial non-faces, and therefore faster detection speed can be achieved, and the real-time detection of faces is further realized on megapixel high-definition videos.

The invention discloses a method for real-time detection of a road vehicle based on a deep learning SSD frame and mainly solves a problem of low detection accuracy of the existing technology under the conditions of traffic congestion and complicated weather. An implementation scheme of the method comprises the following steps that 1.shooting videos of a plurality of running vehicles at a vital communication line and grabbing window information of the vehicle in each frame of image in the video in a manual labelling way; 2.taking VGG-16 in a classification network as a basic network, constructing an SSD300*300 detection frame, extracting features of some feature extraction layers to carry out connection as recognized feature vectors and inputting the feature vectors into a loss function; 3.training through a training sample and taking a trained detection model into the detection frame; and 4.setting a correlation threshold and detecting the vehicles in the test videos by the trained detection model. According to the method, the detection accuracy is greatly improved and an effect of real-time detection is realized, and the method can be used for vehicle detection under the complex scene.

The invention relates to the filed of optical aspheric surface accurate measurement, in particular to a real-time partial zero compensation optical aspheric surface profile detection method. Current zero measurement method only can measure a dynamic range of several wavelengths so that the dynamic range of detection is small. The dynamic range of measurement in the invention is determined by three parts, consisting of a compensation lens, a phase function yGGH which corresponds to a computer generated hologram kinoform loaded by an LCSLM (Liquid Crystal Spatial light Modulator), and a residual phase function yRES. According to the real-time partial zero compensation optical aspheric surface profile detection method, a partial zero compensation technology is adopted, the LCSLM loaded with the computer generated hologram is served as a partial zero compensation element, and a convergent lens playing a role in compensation and an interference fringe analyzing technology are combined, therefore, the dynamic range of measurement is greatly improved, and detection to the profiles of the large aspheric surfaces can be implemented.

The invention relates to a non-covalently modified graphene field effect transistor-based tumor marker detection sensor and a production method thereof. The sensor comprises a silica substrate, a graphene channel and a reaction chamber. The production method comprises the following steps: making a metal electrode, carrying out wet transfer on graphene, making the graphene channel and a protection insulation layer of the graphene channel, and modifying the surface of graphene in a non-covalent mode. The sensor has the advantages of convenient making process, simple modification process, easy operation, and high sensitivity and good specificity in application; and the making method can also be used to make most of biosensors of capture probes with amino groups, and the sensor is used to detect target molecules.

The invention discloses a system for supplying information security operation service to medium-sized and small enterprises. The system comprises the components of a security operation data acquiring module, a security operation service module, a security operation supporting module and a security policy module. The security operation data acquiring module comprises a message detecting module, a terminal security management module, a host security management module and a network equipment security management module. The security operation service module comprises an asset management module and a service management module. The security operation supporting module performs the functions of enterprise user management, identity authentication, authorization charging, service management, terminal management and statistical analysis. The security policy module is used for setting security baseline data, a security event processing strategy, an invasion detecting rule, a security event reporting rule and a service association rule.

The invention discloses a non-rigid multiscale object detection method based on a convolutional neural network, which comprises the steps of firstly designing a deep network, performing deep network learning and training for target detection by infrared pedestrian data sets with certain representativeness cvc-09, cvc-14 and self marked small target pedestrian data, wherein the infrared pedestriandata sets have small targets with the pixel being about 20*30 and big target pedestrians at a close range; then designing different detection models based on a hardware platform, and evaluating the detection models. The designed network applies 7*7 grid partition and applies six candidate boxes in allusion to each grid, target area nomination is accomplished, and pedestrian target detection is achieved through classification recognition and location regression analysis. The neural network is applied in the design, and the feature extraction ability of the detection model is increased. The real-time performance of detection is realized based on the idea of grid partition, the number of the candidate boxes is increased by applying data sets with different target scales and multiscale training, and finally the small target detecting ability is enhanced.

The invention discloses a method for dynamic estimation of vehicle running gradient and relative height. In the method, based on a uniaxial accelerometer and an odometer hardware, an algorithm is used to calculate road grade and vehicle running relative height information, the uniaxial accelerometer is used to measure acceleration accy in the vehicle longitudinal axis direction; the odometer is used to accurately measure vehicle running speed, thus providing the vehicle with vehicle acceleration information a, and according to the geometric relationship between accy and a, gradient value and height change value can be obtained, then whether the vehicle runs upslope and downslope can be identified according to height change and restricting rules. In the process of concrete implementation, installation error of the uniaxial accelerometer, noise processing of acceleration data, alignment of data of two types of sensors and upslope and downslope identification rules are taken into consideration. The method features high measurement accuracy, improvement of stability and reliability of a system, realization of real-time measurement of vehicle running gradient and height, very high sensibility and response speed, small power dissipation, quick startup, simpleness and practicability.

The invention discloses a pointer instrument positioning and identifying method based on machine vision. The position of an instrument panel in an instrument cabinet image is accurately positioned andan instrument reading is automatically recognized. The method comprises the steps of (1) acquiring the instrument cabinet image; (2) carrying out segmentation processing on the instrument cabinet image to obtain an instrument panel image; (3) preprocessing the instrument panel image; (4) acquiring all edge contours in the instrument panel image by using a Canny edge detection method; (5) findingall linear contours in the obtained edge contours by using Hough transform; (6) screening out the linear contours where a pointer in the instrument panel is located through length characteristics of the linear contours; (7) calculating an average value theta of an inclination angle between the two linear contours where the pointer is located; and (8) calculating out a pointer reading according toa linear relation between the linear inclination angle and an instrument scale. By the adoption of a positioning and detecting technology, the method has the characteristics of accurate positioning and reading, good anti-noise performance and efficient and rapid image processing.

The invention discloses an optical remote sensing image ship detection method based on a depth learning single-step detector and mainly solves problems of the limited ability of artificially-designedcharacteristic representation in traditional ship detection methods, insufficient interference resistance capability against the complex background and tedious and slow detection process. The method comprises steps that 1), a remote sensing image data set is established; 2), VGG-16 is utilized as a characteristic extraction portion, a multipath residual block is constructed and is taken as a prediction portion, and a new loss function is constructed; 3), the characteristic extraction portion, the prediction portion and the loss function form a single-step detector; 4), the established data setis utilized to train the single-step detector; and 5), ship detection is carried out thorugh utilizing the trained single-step detector. The method is advantaged in that the method is simple for realization, a ship target in the complex background can be rapidly and accurately detected, and the method is applicable to multiple types of real-time ship detection systems.

The invention discloses a method for monitoring GOOSE communication status of an intelligent substation. The method comprises the steps of obtaining a SCD file of the intelligent substation, configurating a GOOSE configuration file of each IED device of the intelligent substation according to the SCD file, and downloading the GOOSE configuration files to the corresponding IED devices; generating GOOSE messages corresponding to the IED devices according to the GOOSE configuration files; obtaining the GOOSE messages, required to be monitored, of the IED devices in a subscription mode; detecting the GOOSE communication status of the IED devices according to the GOOSE messages; conducting warnings on the corresponding abnormal status according to the detection result. The invention further provides a device for monitoring the GOOSE communication status of the intelligent substation. With the technology of the method and device for monitoring the GOOSE communication status of the intelligent substation, real-time detection on the GOOSE communication status can be achieved, the corresponding warnings can be carried out according to the detection abnormal result, system integration cost can be reduced, the code reuse rate and the communication efficiency are high, and the method and device is convenient to achieve and has good application prospects.

The invention relates to a video detection system and a data processing device and a video detection method thereof, wherein the video detection method comprises the following steps: the traffic video flow is acquired, and the acquired video flow is transmitted to the processing device; the processing device extracts recognition ancillary information from each frame of image; and if a car head is detected in the image according to the recognition ancillary information, the follow-up images are detected and a road traffic information parameter is generated. The invention can achieve full video detection to the road traffic information parameter, avoid false detection and miss detection caused by vehicle driving over lines, vehicle shadow, outside light change and the like, and increase the detection efficiency and accuracy. In addition, the invention can also achieve the real-time display of a dynamic image, and the monitoring of road traffic condition.

The invention specifically discloses a material flow control device applied to a sintering proportioning ore bunker. The device is connected with a blanking device group and comprises an actual flow acquisition unit, a set flow acquisition unit and a blocking material judgment unit, wherein, the actual flow acquisition unit is connected with a weighing device of the blanking device group to acquire the actual value of blanking flow of the blanking device in real time; the set flow acquisition unit is connected with an electric motor to acquire the set value of blanking flow of the blanking device in real time; the blocking material judgment unit is used for outputting a blocking material warning signal when the ratio of the actual value to the set value of the blanking flow of the blanking device is less than the preset percentage. The invention also discloses a material flow control method applied to the sintering proportioning ore bunker. With the device and method of the invention adopted, the material flow state inside the ore bunker can be detected in real time and the sintering production can run smoothly.

The invention discloses an active type direct-current system insulation resistor detection method. The active type direct-current system insulation resistor detection method is applicable to real-time detection of insulation resistors of a direct-current high-voltage source electrical equipment system. The insulation resistor detection method comprises the two stages of fault detection and insulation resistor calculation. According to the fault detection, a parallel-connection resistance value of the positive and the negative insulation resistors can be calculated by measuring voltage values at the two ends of a measurement resistor, and whether the positive and negative insulation resistors break down or not is judged according to the parallel-connection resistance value. According to the insulation resistor calculation, the resistance values of the positive and negative insulation resistors can be calculated by measuring voltage values at the two ends of an access resistor. The active type direct-current system insulation resistor detection method has the advantages that the fault conditions of the positive and negative insulation resistors can be monitored in real time, the whole load is not affected, and the basis is provided for fault diagnosis of the whole system by calculating the resistance values of the positive and negative insulation resistors.

The invention discloses a computer-vision-based self-adaptive control method for a belt conveyor for a coal wharf. The computer-vision-based self-adaptive control method for the belt conveyor for thecoal wharf comprises the steps of during the off-line pre-training stage, through recording a running video of a conveyor belt, extracting an image, making a conveying capacity training data set, a conveying capacity test data set, an abnormal state training data set and an abnormal state testing data set, and training a belt conveyor conveying capacity detection model and a belt conveyor abnormalstate detection model based on a convolutional neural network; during the on-line monitoring stage, using the trained belt conveyor conveying capacity detection model and the trained belt conveyor abnormal state detection model for monitoring the conveyor belt in real time, self-adaptively controlling the conveying speed of the belt conveyor according to the conveying capacity of the belt conveyor, detecting multiple abnormal states of the conveyor belt at the same time, immediately stopping running after the abnormal states are found, and sending an alarm corresponding to the abnormal state.The operation cost of the belt conveyor is greatly reduced, the overhaul and maintenance efficiency of the belt conveyor are improved, the maintenance cost is reduced, the safety is improved, the detection models are built by using the convolutional neural network, the detection accuracy is improved, and the recognition error rate is reduced.

The invention discloses a bluetooth electronic cardiophone which is characterized by comprising a portable terminal and a PC. In the portable terminal, after a sensor conducts primary amplifying, lowpass filtering and secondary amplifying on an acquired cardiac sound signal, a power amplification module conducts real-time cardiac sound playing on one path of the cardiac sound signal, and after a master control module conducts A/D conversion on the other path of the cardiac sound signal, a liquid crystal screen displays a phonocardiogram at the real time, and the other path of the cardiac sound signal is transmitted to the PC by a serial port or a bluetooth transmission module; and the PC displays the phonocardiogram at the real time through a PC screen after received cardiac data is processed by the master control module at the PC end, or the received cardiac data is stored by a storage module and assists a diagnosis module to carry out computer assisted diagnosis according to requires. Compared with the prior art, the invention has the advantages of storing and analyzing the cardiac sound data, displaying the phonocardiogram at the real time, providing more intuitive diagnosis data and giving prompts to common heart diseases, such as tachycardia, bradycardia, arrhythmia, and the like.

The invention provides an array substrate, a display panel and a display device, and belongs to the technical field of display. The array substrate comprises a display area and further comprises a through hole, an anti-cracking dam, a metal lead and a first data line, wherein the through hole is formed in the display area, the anti-cracking dam surrounds the through hole, the metal lead surrounds the anti-cracking dam, the first data line is divided into a first section and a second section by the through hole, the first section of the first data line is connected with one end of the metal lead, and the second section of the first data line is connected with the other end of the metal lead. Cracks at the edge of the through hole can be effectively detected through the array substrate.

The invention relates to a method for multi-channel self-adaptive switch during terminal communication, in particular relating to a method for self-adaptive switch of a wireless channel and an Ethernet channel. The method comprises: a terminal automatically detects a wireless communication module and an Ethernet physical link channel, automatically judges the parametric model of Ethernet, and switches to the Ethernet channel when detecting that the Ethernet channel is available (or else switches to the wireless channel). The method has the following beneficial effects: the real-time detection of the wireless channel and the Ethernet channel can be realized, the switching is automatically carried out according to detection results, the fast switch actions are realized in the whole process, the self adaptability to two types of network channels is fully realized, and the manual intervention is avoided; and when the Ethernet is available, the Ethernet channel is utilized preferably; because the Ethernet channel is adopted, the network bandwidth can be greatly expanded, the network time delay is reduced, and meanwhile the consideration of flow is ensured to be unnecessary, thereby reducing a great deal of communication cost; and when the Ethernet is not available, the wireless module is started for utilizing the wireless channel for communication, so that the smooth communication between the terminal and a master station can be ensured.

The present invention discloses a method and a system for on-line diagnosis of transformer winding deformation. The method comprises: injecting a sweep signal from a high-voltage lead-out line of a transformer winding by way of magnetic field induction; acquiring an excitation current signal induced on the high-voltage lead-out line of the transformer winding; obtaining the response current signal on the neutral ground line of the transformer winding; obtaining the frequency response curve according to the transfer function relationship, which is constructed by using the excitation current signal and the response current signal, of the transformer winding; based on the diagnosis method in the DLT911-2004 standard, diagnosing the frequency response curve; and determining the deformation of the transformer winding. According to the invention, the excitation signal injection and response signal measurement do not need direct electrical connection with the transformer winding, making the high-voltage transformer winding operate in a high-voltage and high-current condition and realizing the real-time detection of the transformer winding deformation online.

The invention relates to a detection method for high-speed train operation dynamics performance state. The method comprises the following steps: A, establishing a numerical simulation model for train dynamics state detection and fixing dynamic response control thresholds of all detection points of a train, B, distributing and optimizing train dynamics state detection sensors, C, measuring dynamic response signals of all the detection points of the train during operation through the sensors, and analyzing and processing detected signal data, so as to obtain detection values of the dynamics state of all the detection points of the train, and D, comparing the detection values obtained in step C to the control thresholds obtained in step A, and further evaluating the dynamics state of the train. The method combines an on-line monitoring technology and a dynamics simulation technology, realizes the real-time detection of high-speed operation state, and provides technical support for safe and reliable transportation of the high-speed train.

The invention provides a forest fire laser radar detection system based on an unmanned aerial vehicle. The system comprises an unmanned aerial vehicle system and a forest fire detection system. The system is mainly characterized in that forest inspection can be performed and the forest fire combustion point and the range of combustion can be timely discovered. Laser radar emits laser to detect the forest through a laser transmitter. The laser is absorbed and reflected through the combustion area. An optical receiver receives the reflected light. A signal converter converts light signals into electric signals. A microprocessor processes the electric signals. The advantages of the system are that the device is convenient to carry and can remotely detect the forest fire in a wide range of area and accurately measure the place and the range of fire. The change of the forest temperature can also be detected by the forest fire laser radar detection system based on the unmanned aerial vehicle so as to achieve the objective of forest fire inspection.

The invention belongs to the technical field of a transport control system of road vehicles, and discloses a highway vehicle behavior semantic analysis and abnormal behavior monitoring system and method. By detecting a video and and tracking a moving vehicle target, vehicle speed detection and vehicle behavior semantic analysis are achieved, whether a vehicle has traffic offence behaviors or not is judged, and then the function of alarming the traffic offence behaviors is achieved; a multi-hybrid Gaussian model is utilized to detect the moving target from the video, a Kalman filter is utilizedto track the moving vehicle target to obtain the track in the vehicle driving process and three major motion characteristics, namely, the position, the direction and the speed, detection is conducted, the vehicle behavior semantic analysis is conducted, through the further utilization of a road driving regulation, whether the vehicle has the traffic offence behaviors or not is judged, and if yes,the vehicle is snapshot and alarmed. Vehicle driving and semantic analysis can be accurately detected in real time in a highway, abnormal behaviors and the traffic offence behaviors of the vehicle are alarmed and snapshot, and the purposes of automatic detection, quick alarming and timely processing are achieved.

The invention discloses a system and a method for intelligently detecting duct piece splicing quality based on image identification. The detection system comprises an identifier, an image acquisition device, an image processing system and a display device. The method comprises the following steps of: drawing a duct piece slab staggering amount control model according to the inner diameter of each duct piece and a slab staggering allowed deviation; binding the control model with a specified identifier in the system, and correctly identifying the identifier to correspondingly load the model; fixing the specific identifier on a contact between adjacent duct pieces, and acquiring the size of the identifier and the distance between adjacent duct piece contacts; arranging a camera on site, and acquiring an image which comprises the identifier and a duct piece entity simultaneously by using the camera; receiving the acquired image by using the image processing system, running a program, forming an image formed by overlapping the duct piece entity and the duct piece slab staggering amount control model, and inputting the image into display equipment; and intelligently judging whether the slab staggering amount of adjacent duct pieces is consistent with specification requirements or not according to an overlapped image provided by the display equipment.