371results about How to "Solve misjudgment" patented technology

The invention discloses a method for preventing the content on a screen from being peeped, and a mobile terminal thereof. The method comprises the following steps of acquiring an image, extracting the facial feature information of a human face in a predetermined frame and judging whether the information is matched with the pre-stored facial feature information or not; if not, extracting the attitude information, and judging whether the attitude information is matched with the pre-stored non-peeped attitude information or not; if not, prompting a user that the content on the screen is peeped. The content on the screen can be initiatively protected, and the judgment conditions are more rigorous and reasonable, so that the problems that people other than the users in front of the screen peeping the content on the screen can not be detected, and the misjudgment problems on pedestrian in the prior art can be solved, so the accuracy of judgment is improved.

The invention belongs to the technical field of building material and structure monitoring, and provides a concrete surface crack real-time monitoring system and a cracking risk dynamic assessment method. According to the cracking risk dynamic assessment method, images of the surfaces of concrete material test pieces are obtained by using CCD lens or digital cameras; processing analysis on acquired and stored digital images is performed via adopting of digital image matching method; digital image processing technique is used for identifying characteristic parameters such as position, area, length, and width of concrete cracks, and weighting classification is performed; quantified concrete cracking risk grading evaluate standards are obtained; and dynamic assessment on concrete cracking risk is realized via comparison with acquired concrete crack characteristic parameters. Existing concrete crack monitoring technique is high in cost and low in accuracy, and misjudgement on crack boundaries is caused when digital image method is used for concrete crack extraction. The concrete surface crack real-time monitoring system is capable of avoiding the disadvantages above, identifying the boundaries of the concrete surface cracks accurately, and evaluating concrete structure safety objectively and flexibly.

The invention discloses a bamboo strip defect online detection method based on machine vision. A camera is used for photographing the four surfaces of a bamboo strip continuously along the length direction of the bamboo strip to obtain the gray images of the surfaces of the bamboo strip. The image treatment is implemented on every captured frame of digital images and whether defects exist on the small section of the bamboo strip in the frame of the images is judged according to the bamboo strip defect detection method. If the small section of the bamboo strip in the frame of the images is provided with detects, the whole bamboo strip is considered to be provided with detects and simultaneously an instruction of the existing defects is sent to a sorting mechanism. If small sections of the bamboo strip in all frames of the images of the whole bamboo strip are provided with no detects, the bamboo strip is considered to be provided no detects. Simultaneously, an instruction with no detects is sent to the sorting mechanism. The method can implement the online detection on the defects of bamboos with different color shades and simultaneously detect the four surfaces along the length direction of the bamboo strip.

The invention discloses a method and system for monitoring the content delivery network (CDN) equipment status, which are used to provide an efficient platform for quick positioning and correct exception handling for operation and maintenance personnel. The method comprises the following steps: periodically reporting equipment heartbeat information to a monitoring server which is taken as a heartbeat receiving end by monitored CDN equipment; periodically and initiatively detecting the status of the monitored CDN equipment on the monitoring server to obtain an initiative detection result; comprehensively monitoring the equipment heartbeat information and the initiative detection result which are collected by the monitoring server, determining the equipment status by combining with the topological structure of the equipment, and sending an alarm aiming at root causes of a failure; and intuitionally displaying a decision result from an equipment status decision device by combining with machine room distribution and the topological structure of the equipment, and displaying the root cause alarm given out by the equipment status decision device on an alarm page.

The invention discloses a gravity inductive type axis-counting device and a method for detecting the state of a tract district thereof, relating to a railway transport management device and a management method thereof. The gravity inductive type axis-counting device comprises an axis detector, an axis-counting information processing system, an input-output subsystem and a power supply, wherein the axis detector is composed of a wheel electronic detector and a gravity inductive type sensor unit; the wheel electronic detector consists of an axis-counting CPU unit, a signal processing unit, a sensor power unit and a state detecting unit; the axis-counting information processing unit comprises a CPU unit, a communication unit, and an I / O unit; and the input-output subsystem consists of a dynamic drive assembly unit, a relay and a button. In the method, a wheel signal is detected by a gravity inductive type sensor arranged on a rail, and is transmitted to the wheel electronic detector and the axis-counting information processing unit for processing, thus judging the state of the tract district. The invention has the advantages of accurate detection result, high reliability, scientific method and capability of realizing automatic judgment on the idle and occupied states of the tract district.

The invention discloses an RGBD image salience detection method based on a hypergraph model. The method includes conducting hyperpixel segmentation for a color image to be detected and a depth image, calculating neighborhood depth contrast graphs for each hyperpixel area of the depth image, constructing a depth background hyperedge according to neighborhood depth contrast ratio, extracting hyper pixel areas on the boundary of the image to construct a boundary background hyperedge, calculating the weight of the two hyperedges, expanding hyperedges according to a hypergraph learning algorithm, building an induce graph, prior calculating boundary background salient map by using the boundary connectivity based on the spatial adjacent relation of the induce graph and the edge weight, and obtaining a final salient detection map based on salience degree updating algorithm of cellular automaton and a fusion algorithm in combination with depth prior. Deficiency of conventional 2D boundary background prior is overcome. The improvement is made based on depth information and a hypergraph model, and better effects are achieved than a conventional image salience detection method that combines color and depth information.

The embodiment of the invention discloses a proximity sensor adjusting method and apparatus. The method comprises the steps of determining whether a proximity sensor is in a target state of not being shielded by a body part of a user; when the proximity sensor is in the target state, obtaining an infrared reflection intensity value obtained by the proximity sensor; and if the infrared reflection intensity value exceeds a preset range of intensity values, adjusting of emission power of the proximity sensor is carried out until the infrared reflection intensity value obtained by the proximity sensor is positioned in the range of intensity values. According to the technical scheme provided by the embodiment, problems of misjudgment of whether the face of the user is close to the proximity sensor are solved.

The invention discloses an odor evaluation method for automotive interior parts. Through larger sample sizes and higher test requirements, discrimination of an odor source is increased, so that contribution of a part assembly to the odor of the whole automobile can be investigated more visually. Compared to odor evaluation on material types, the method can quickly locate the whole automobile odor contribution source and achieve odor change, so that the problem of misjudgment caused by poor local homogeneity of constituent materials of the parts is effectively solved, and detection missing of a small number of the constituent materials or auxiliary materials of the complicated part assembly is prevented.

The invention discloses a regional water body rapid dynamic extraction method combining optics and radar. The method comprises the following steps: obtaining optical remote sensing image data of a monitoring region; obtaining radar remote sensing image data of the monitoring area; preprocessing the optical remote sensing image data and the radar remote sensing image data respectively; obtaining aspectral index based on the preprocessed optical remote sensing image data; extracting an optical remote sensing area water body classification map based on the spectral index; based on the optical remote sensing area water body classification map, extracting a radar remote sensing area water body classification map in the preprocessed radar remote sensing image data; and obtaining a monitoring area water body map based on the optical remote sensing area water body classification map and the radar remote sensing area water body classification map. According to the invention, dynamic extractionand monitoring of water bodies in a regional range can be realized.

The invention provides a method and a device for receiving a preamble, which can be applied to when a transmitting end meets a preset transmitting rule. The method is characterized by comprising steps: whether a preamble exists in a processed baseband signal is judged; the position of the preamble in a physical frame is determined and carried signaling information is analyzed, wherein the step of determination and analysis comprises a sub step in which a time domain main body signal for each time domain symbol is subjected to Fourier transform and effective sub carriers are then extracted; each effective sub carrier and a known sub carrier corresponding to each frequency domain known sequence in a known frequency domain signaling set for the time domain symbol are subjected to predetermined mathematical operation and then inverse Fourier transform, and an inverse Fourier result is obtained corresponding to each frequency domain known sequence; and based on an inverse Fourier selection result selected from one or more inverse Fourier results in a first predetermined selection rule for each time domain symbol, and by using predetermined processing operation between multiple time domain symbols, the signaling information is analyzed based on the obtained inter-symbol processing result.

The invention provides a method for controlling data quality of an automatic meteorological station based on strong weather process correction. The method solves the problems of false judging and missed judging of the traditional quality control method, and can obviously improve the reliability and accuracy of the quality control of the automatic meteorological station; especially for China southeast areas and coastal regions with frequent occurrence of extreme weathers, a smaller judging threshold value can be adopted, so as to realize more accurate judging, and effectively identify the false judging caused by local strong weather process (especially cold front and warm front crossing regions). The method has the advantages that by adopting the secondary reliability test correction, the false judging on the tested correct records caused by false records of the adjacent stations in the test calculation process is effectively avoided, and the accuracy of the quality control is further improved; the more accurate weather actual monitoring data sources are provided for a meteorological service system, and the accuracy of weather precast is further improved.

The invention discloses a train rolling bearing fault and impurity vibration distinguishing method. The method includes: utilizing spectral kurtosis to perform filtering processing on an original bearing vibration signal, and adopting an empirical mode decomposition technique to perform step-by-step decomposition on the preprocessed vibration signal to obtain intrinsic mode functions; subjecting the intrinsic mode functions to frequency spectrum and statistic characteristic analysis, extracting a feature vector composed of frequency spectrum center-of-gravity and characteristic frequency energy ratio as input, establishing a support vector machine based classifier model by training, and distinguishing whether a rolling bearing with abnormal noise is faulted or contains impurities in real time. The method is used for distinguishing the fault or impurity-containing state of the train rolling bearing based on vibration signal analysis, the method effectively improves accuracy in bearing fault identification, the problem of overhigh false alarm rate in the bearing vibration diagnosis method caused by impurities is solved, and good technical support is provided for engineering application of the vibration diagnosis method.

Provided is a light sensing type touching and controlling circuit, comprising a light sensing unit allowed to switch on or off according to an illumination strength to accordingly provide a first signal comprising a first voltage and a second voltage; a first signal reading unit coupled with the light sensing unit and allowed to switch on or off according to a first operation time sequence to output the first signal; a touching and controlling reference unit arranged at one side of the light sensing unit and used for providing a reference voltage; a second signal reading unit coupled with the touching and controlling reference unit and allowed to switch on or off according to a second operation time sequence to output a reference voltage; and a touching and controlling sensing unit coupled with the first signal reading unit and the second signal reading unit, wherein the touching and controlling sensing unit determines whether the light sensing type touching and controlling circuit is touched using the level difference between the first signal and the reference voltage as one of bases.

The invention provides a method and a device for receiving a preamble, which can be applied to when a transmitting end meets a preset transmitting rule. The method is characterized by comprising steps: whether a preamble exists in a baseband signal is judged; the position of the preamble in a physical frame is determined and carried signaling information is analyzed, wherein the step of determination and analysis comprises a substep in which a time domain main body signal for each time domain symbol is subjected to Fourier transform and effective sub carriers are then extracted; each effective sub carrier, a known sub carrier corresponding to each frequency domain known sequence in a known frequency domain signaling set for the time domain symbol and a channel estimation value are subjected to predetermined mathematical operation and then inverse Fourier transform is carried out, and an inverse Fourier result is obtained corresponding to each frequency domain known sequence; and based on an inverse Fourier selection result selected from inverse Fourier results in a first predetermined selection rule for each time domain symbol, the signaling information is analyzed directly and / or by using predetermined processing operation between multiple time domain symbols, the signaling information is analyzed based on the obtained inter-symbol processing result.

The invention discloses a DNS tunnel detection method. The DNS tunnel detection method includes the steps that whether a client side data package sent to a port of a DNS server is a DNS data package is detected; when the client side data package is the DNS data package, whether the format of the client side data package accords with preset format constraint conditions is detected, and if the detected result is yes, whether the query domain name of the client side data package accords with preset domain name constraint conditions is detected; when the format of the client side data package does not accord with the format constraint conditions, and / or the query domain name of the client side data package does not accord with the domain name constraint conditions, the client side data package is correspondingly processed according to preset DNS tunnel processing rules. The invention further discloses a corresponding DNS tunnel detection device. By means of the scheme of the DNS tunnel detection method and device, the normal DNS data package and a DNS tunnel data package can be distinguished, the problem of erroneous judgment of a normal DNS is effectively solved, a DNS tunnel is detected comprehensively, and therefore a firewall is effectively protected.

The invention discloses a control system performer fault diagnosis method based on multi-swarm cooperative chaos simulated annealing particle swarm optimization-support vector machine (MCCSAPSO-SVM). The method comprises the steps of: by means of combined noise reduction and an improved empirical mode decomposition (EMD) method, carrying out feature extraction on collected output signals of a performer; utilizing MCCSAPSO to optimize structural parameters of the support vector machine; adopting a combined Kernel function to ensure good generalization capability and learning capability of the SVM; and utilizing training data to construct a binary tree SVM, and converting a complex polytomous problem into a plurality of dichotomous problems. According to the invention, common state signals, which can be easily obtained, of the control system are processed, whether the control system performer breaks down can be effectively judged in real time according to the output of the binary tree SVM, and the fault type can be accurately determined when the control system performer breaks down. The method is used for real-time fault diagnosis of a high-precision system.

The invention discloses a gas pipeline leakage point detection device and a detection method, and relates to the field of gas pipeline detection. The gas pipeline leakage point detection device comprises a sensor module, a data transmission module, a fixed IP line, a server, a gas pipeline leakage alarm management system and a management computer. The sensor module acquires flow data, pressure data and infrasonic wave data of a gas pipeline. The data transmission module is connected with the sensor module. The fixed IP line is connected with the data transmission module. The server is connected with the fixed IP line, and the flow data, the pressure data and the infrasonic wave data are uploaded to the server through the data transmission module and the fixed IP line. The gas pipeline leakage alarm management system is installed in the server, receives, calculates and analyzes the flow data, the pressure data and the infrasonic wave data, shows the gas running state and gives an alarm.The management computer is connected with the server and used for managing the gas pipeline leakage alarm management system. By the adoption of the gas pipeline leakage point detection device and thedetection method, gas feature graphs of different road sections are provided, the number of influence factors such as the environment is reduced, the false alarm rate of gas alarms is reduced, and the detection accuracy and the alarm efficiency are high.

The invention provides an intelligent paper marking method for subjective questions. The method comprises the steps of respectively preprocessing an examinee answer text and a reference answer text toobtain word vector sequence representations corresponding to the examinee answer text and the reference answer text; extracting semantic vectors of sentences in the examinee answers and the referenceanswers by utilizing a semantic feature extraction network; calculating semantic matching degrees of sentence vectors of the examinee answers and the reference answers by utilizing a fusion attentionmechanism; calculating an examinee answer weighted sentence vector by utilizing the matching result; aiming at the sentence vector sequence of the answers of the complete examinees and a weighted sentence vector sequence based on a fusion attention mechanism; utilizing a semantic feature extraction network to calculate semantic vectors of the complete answers and semantic vectors based on attention of the reference answers; and bonding the two vectors to form a final vector representation of the examinee answer; scoring the final vector representation of the examinee answer by using a multi-layer feedforward neural network. According to the invention, automatic examination paper marking of subjective questions can be realized, and examination paper marking efficiency is greatly improved.

The invention discloses a heartbeat detection device and a heartbeat detection method applied to a distributed system and network equipment. According to the method, subequipment monitors whether an application module on the subequipment normally occupies the processing resource of a central processing unit (CPU) or not; if the application module is monitored to normally occupy the processing resource of the CPU completely and software heartbeat period cannot be timed, the subequipment transmits a software heartbeat message to primary equipment; and the software heartbeat message is still transmitted to the primary equipment under the conditions that the processing resource of the CPU is occupied normally and completely and the software heartbeat period cannot be timed, so that misjudgment on the subequipment by the primary equipment can be reduced, the accuracy and the reliability of heartbeat detection are improved, and the problems of misjudgment, low accuracy and low reliability of heartbeat detection in the prior art can be solved.

The invention discloses a power closed-loop scanning-based maximum power point tracking (MPPT-Maximum Power Point Tracking) method for a plurality of peak values of a photovoltaic array. The power closed-loop scanning-based maximum power point tracking method comprises three stages: firstly, realizing global scanning of a multi-peak value curve on the basis of power closed-loop control, and completing the positioning of a maximum powder point; secondly, performing a three-point synergetic variable-step length disturbance observation process on the basis of a particle swarm optimization algorithm, and performing local searching near the maximum power point; thirdly, performing a fixed voltage steady state tracking process, and meanwhile, activating the processes of the first stage or the second stage according to judgment of environmental change information. According to the method, tracking of the maximum power point of the photovoltaic array can be realized in single-peak or multi-peak situations; the problems of misjudgment and vibration of a disturbance observation method and a conductance increment method are avoided; and meanwhile, the problems that the gradual dynamic tracking function loss is great and the dynamic process is slow in the conventional method can be solved. By the method, the global maximum power point can be effectively searched; quick, stable and accurate tracking of the maximum power point is realized.

The invention relates to an ultrasonic detection method for different technical stages of a composite blade ring. A pulse echo type water-leaching focused ultrasound C scan method is adopted for performing ultrasonic detection on the whole blade ring. The ultrasonic detection method comprises the following steps: firstly, respectively performing water-leaching focused ultrasound C scan detection on interior defects of a titanium alloy inner ring and outer ring forge piece to be used for making the whole blade ring; detecting a joint interface in an integrally formed blade ring after being subjected to cellosilk winding and inner and outer ring hot isostatic pressure, namely with a 5-10MHz water-leaching focused probe, respectively performing C scan imaging on an interface signal and a bottom wave signal; and detecting the bonding quality of a metal / metal interface between the inner and outer rings of titanium alloy and a metal / fiber interface between titanium alloy and silicon carbide cellosilk. The ultrasonic detection method can be adopted for comprehensively controlling the inner quality and interface bonding quality of the whole blade ring of silicon carbide fiber enhanced Ti-based composite and has the advantages of strong operability, high detection sensitivity and direct detection result.

Embodiments of the invention provide a voice keyword recognition method and device, electronic equipment and a storage medium. The method comprises the following steps of extracting an acoustic stateposterior probability distribution vector of each frame in voice data corresponding to a to-be-recognized word, wherein the acoustic state posterior probability distribution vector of any frame comprises a posterior probability of any frame relative to a plurality of acoustic states; and inputting the acoustic state posterior probability distribution vector of each frame in the voice data to a keyword recognition model so as to obtain a keyword recognition result corresponding to the to-be-recognized word and output by the keyword recognition model, wherein the keyword recognition model is obtained by training based on a sample acoustic state posterior probability distribution vector of each sample frame in a sample word and a keyword identification of the sample word. According to the method, the device, the electronic equipment and the storage medium, the recognition precision can be improved, the misjudgment problem of similar words is avoided, the response accuracy is improved, andthe user experience is optimized.

The invention relates to a base-plate laser-repairing machine automatic flaw checkout gear and method. After a base-plate is electrically tested, according to the results of the electrical test, a second view-finding device of an optical examination module replaces eyes to high-speed view-find defect lines and detect out the defects through an image detection mode of a laser-repairing module, which is convenient for subsequent laser-repairing movements.

The invention provides a molecular biology-based misgurnus anguillicaudatus and paramisgurnus dabryanus specie identification primer and an identification method. The misgurnus anguillicaudatus and paramisgurnus dabryanus specie identification primer is characterized by designing a specific primer sequence: upstream primer sequence being 5'-CCGCCCAGGAGTATTTTATG-3'(shown by SEQ ID NO.1); downstream primer sequence being 5'-TGGCATTTCATCAGGGGTG-3'(shown by SEQ ID NO.2). The identification method comprises the following steps: extracting DNA of misgurnus anguillicaudatus or paramisgurnus dabryanus samples by adopting a conventional method; performing PCR amplification; after the PCR amplification is completed, taking a proper amount of amplified products for performing gel electrophoresis detection to determine the size of amplified fragments, and identifying. The identification method is accurate and high-efficiency. The problems that the traditional morphological distinguishing method is difficult to operate and low in resolution success rate can be solved. Especially, the misgurnus anguillicaudatus and paramisgurnus dabryanu at the young and juvenile stages can be difficulty identified from appearance, and the invention provides a convenient and accurate identification method.

The invention discloses a detection method of flame. Since the flame has the characteristics of certain continuousness, and the like at a position, the detection method can be used for acquiring images. Pixel points having the flame color characteristics in the images can be detected by using color information. The flame area can be detected by using a time domain filter detection motion area andthe like. The detection method can be used for fire monitoring at long distance in a wide range.

The invention discloses an EPB automatic release control method and device, a controller and a computer storage medium, and the method comprises the steps: judging whether a driver controls a vehicleor not according to a driver seat image or driver seat door information shot by a camera; when it is detected that a driver seat image shot by the camera comprises human body characteristics, judgingthat a driver controls the vehicle in the vehicle; when it is detected that the driver seat door is in a closed state and an operation signal of a driver exists, judging that a driver controls the vehicle in the vehicle; when it is judged that the driver controls the vehicle in the vehicle, and the EPB release condition is met, controlling the EPB to be released, so the problem of misjudgment about whether the driver is in the driving position or not is avoided through the real camera image or the vehicle door information and the driver operation signal, the response of whether the EPB is automatically released or not most conforming to the current situation and the intention of the driver can be made, and the intelligent experience of vehicle driving is improved.

The invention provides a human-computer interaction method and a human-computer interaction device based on the integration of eye movement tracking and gesture recognition in a virtual assembly. Themethod comprises the steps: carrying out the gaze point tracking according to the obtained eye movement data; performing gesture recognition according to the obtained gesture information, labeling theobtained gesture recognition data and eye movement data to form a training set, and constructing a multi-stream convolutional neural network-long-term and short-term memory network model, wherein thenetwork model performs self-learning by using the training set; and applying the optimal network model obtained by training to a virtual assembly process, obtaining eye movement data and gesture information of the virtual assembly process, extracting eye movement and gesture features, and analyzing according to the feature information to obtain a behavior category of an operator so as to completean assembly task. The problem of misjudgment of similar behaviors in a single mode is solved. The advantages of a deep learning algorithm are utilized. The behaviors of operators in a video are recognized with high accuracy. A virtual assembly task is completed, and the man-machine interaction is achieved.

The invention discloses a method, device, equipment for parking line-crossing detection and a computer readable storage medium. The method comprises the steps that a camera is used for obtaining a monitoring image of a current parking space of a parking lot, and the size of the monitoring image is converted into a target monitoring image with a preset size; inputting the target monitoring image into an image segmentation convolutional neural network which is trained in advance, and extracting a vehicle image in the target monitoring image; wherein the image segmentation convolutional neural network is constructed based on a U-Net network; judging whether the vehicle in the current parking space crosses a pre-calibrated parking space line of the current parking space according to the calibration information of the vehicle image; and if the vehicle in the current parking space crosses the parking space line of the current parking space, sending warning information of the parking line crossing in the current parking space to a staff in the parking lot. By using the method, device, equipment for parking line-crossing detection and the computer readable storage medium, the accuracy of the parking line-crossing detection result is improved.

The invention relates to a parking space detection system and method based on double parking detectors. The parking space detection system based on the double parking detectors is mainly characterized by comprising a parking space detection server, a wireless parking space detection network and the parking space detectors, the parking space detection server is connected with the parking space detectors through the wireless parking space detection network, and each of two sides, close to side lines of each parking space, is provided with one parking space detector. The parking space detection method includes the steps: using the two parking space detectors, on the side lines of each parking space, for detection; judging the state of the parking space is unchanged when detection values, of the two parking space detectors on the side lines of one parking space, hop within different time; using a matching detection method for the detection of the parking space when the detection values, of the two parking space detectors on the side lines of one parking space, hop at the same time. The parking space detection system and method based on the double parking detectors is reasonable in design and capable of effectively solving the problem of misjudgment due to vehicle magnetic field effect of the neighboring parking spaces which are arranged in a parallel mode, and accordingly detection precision is improved.