77results about How to "Not easy to misjudge" patented technology

The invention relates to a valueless image removing method based on deep convolutional neural networks. The valueless image removing method comprises the steps of firstly, after performing whitening preprocessing on an image sample set, performing pre-training on a sparse autocoder to obtain the initialization results of deep convolutional network parameters, secondly, building a plurality of layers of deep convolutional neural networks and optimizing the network parameters layer by layer, and finally, classifying a plurality of classes of problems by use of a realized multi-classification softmax model and then realizing the removal of valueless images. Due to the automatic image learning characteristic of the sparse autocoder, the classification correction rate of the valueless image removing method based on the deep convolutional neural networks is increased. The plurality of layers of deep convolutional neural networks are built on the basis of the automatic image learning characteristic of the sparse autocoder, the network parameters are optimized layer by layer, the characteristic of each layer after learning is the combination result of the characteristics of the previous layer, and the multi-classification softmax model is trained to judge images, and consequently, the removal of valueless images is realized.

The invention relates to a mobile phone and an incoming call processing method thereof. The mobile phone comprises a proximity sensor electrically connected with a main control chip. The method comprises the following steps of: automatically putting a user through according to a process that the mobile phone approaches to an object or the mobile phone keeps away from and then approaches to the object, or automatically hanging up a call in process according to the process that the mobile phone keeps away from the object. The mobile phone and the incoming call processing method thereof have the advantages of completely according with the answering habit of the user and further selecting automatic answering and/or automatic hanging up according to user setting to meet the requirements of different people, along with high correctness and rare misjudgement.

The invention discloses an earthquake prediction method based on resonance package monitoring data. The earthquake prediction method based on the resonance package monitoring data comprises the steps that resonance package monitoring data in one statistical cycle are acquitted continuously, wherein the statistical cycle refers to the period from the day before the day a tidal wave resonance package appears for the first time to the day after the day the tidal wave resonance package appears for the last time; the amplitudes, the cycles and the travel time sequences of resonance waves and tidal waves are obtained according to the resonance package monitoring data, and the earthquake magnitude, the seismic center and the earthquake occurrence time are determined by recognizing three element signals of an earthquake according to the earthquake burst kinetics precursor theoretical method. By the adoption of the earthquake prediction method based on the resonance package monitoring data, an amplitude signal, a resonance signal and a time travel time signal can be extracted to different precision extents by means of any methods which can record earthquake precursor information, misjudgment is unlikely to happen, and the reliability of earthquake predicting and forecasting is remarkably improved.

The invention provides an engine fire fault diagnosis method, which comprises the following steps of: judging whether an engine has an influence on a fire diagnosis fault, updating the information index value of the fire diagnosis fault, judging whether the engine satisfies the operating condition of engine fault diagnosis, judging whether the current working condition is in an area of prohibiting fire diagnosis, determining a fire cylinder according to the acceleration characteristic value in an acceleration mode, calculating a torque characteristic value in a torque mode, judging whether a catalyst damage type fire exists according to the fire rate, and judging whether emission degradation type fire exits according to the fire rate. By the method, the fire of the engine can be timely and accurately discovered, the accuracy of the fire diagnosis is improved, erroneous judgment is avoided, and the fire caused by emission damage or catalytic converter damage can be determined.

The invention discloses a system for monitoring wear of a pantograph sliding plate. The system comprises a three-dimensional (3D) camera and a laser, wherein the three-dimensional camera and the laser are connected with a PLC (Programmable logic controller), and are controlled to be switched on and off by the PLC; the PLC is also connected with an industrial control computer; and the industrial control computer is connected with the 3D camera through a broadband network, and can be used for receiving and storing photos transmitted by the 3D camera through the broadband network. During the passage of a pantograph, the 3D camera can be used for transmitting shot photos of the top of the pantograph through the broadband network, the industrial controllable computer is used for analyzing 3D photos, and the 3D coordinate values of the sliding plate on the top of the pantograph can be obtained clearly, so that the wear condition of the sliding plate can be obtained clearly and accurately, monitoring of the wear condition of the pantograph sliding plate can be automatically realized, and high efficiency and high accuracy are achieved.

The invention discloses a pantograph monitoring system, which comprises a CCD (Charge Coupled Device) camera, a laser group and a flash light group, wherein the CCD camera, the laser group and the flash light group are connected with a PLC (Programmable Logic Controller) and are controlled to be switched on/off by the PLC controller; the PLC controller is also connected with an industrial computer; and the industrial computer is connected with the CCD camera through a broadband network, and can receive, store and process photos sent by the CCD camera through the broadband network. The industrial computer is used for analyzing and processing an image of a part of the side face of the pantograph head of the pantograph, taken by using the CCD, so that the wearing situation of a sliding plate on the top of the pantograph head of the pantograph and the deformation situations of cleats on two sides can be known clearly and vividly. By adopting the pantograph monitoring system, the wearing situation of the sliding plate of the pantograph and the deformation situations of the cleats can be automatically monitored, high efficiency and high accuracy are achieved, error judgment does not occur easily, and the bow net accident is avoided more reliably.

The invention relates to a flatness detecting device. The flatness detecting device includes a work table, a total assembly rack arranged on the work table, a detection platform which is arranged on the work table and includes a bearing surface for bearing and fixing products to be tested, a CCD detector which is arranged on the total assembly rack and is located above the detection platform, and a control device used for turning on and controlling the CCD detector to perform flatness detection; the bearing surface is provided with a plurality of upright columns which protrude along a Z-axis direction; and the top surfaces of the upright columns are provided with tilt light reflecting mirror surfaces, included angles being formed between the tilt light reflecting mirror surfaces and a horizontal plane being acute angles. The flatness detecting device has the advantages of high detection efficiency and little possibility of causing misjudgment.

The invention relates to a heating device which comprises a furnace surface plate, a heating component, a shell and a blower, wherein the furnace surface plate is mounted on the shell; an accommodating cavity is formed between the furnace surface plate and the shell; the heating component and the blower are mounted in the accommodating cavity; the heating component is positioned below the furnace surface plate; a ventilation spacing is formed between the heating component and the side wall of the shell; the blower is arranged apart below the heating component; the air-out direction of the blower faces the heating component; a ventilation spacing is formed between the blower and the side wall of the shell; the top wall of the shell is provided with a plurality of air outlets distributed at intervals in the circumferential direction; the lower end of the shell is provided with an air inlet; and the air inlet, the ventilation spacings and the air outlets constitute a basin-shaped heat-radiating channel. According to the invention, in the working process of the blower, the heat-radiating component is always wrapped by the basin-shaped heat-radiating screen, so that the heat-radiating effect is favorable, and the efficiency is high.

The invention provides a resonance packet wave spectrum monitoring data-based earthquake prediction method. The method includes the following steps that: monitoring precursor stations are established globally; resonance packet monitoring data generated by the monitoring precursor stations for regions where the monitoring precursor stations are located are acquired continuously; acquired resonance waves are aggregated and drawn into a resonance packet wave spectrum by using the waveform characteristics of resonance waves, and at the same time, the time difference of a time when fast waves arrive at the monitoring precursor stations and a time when slow waves arrive at the monitoring precursor stations in the resonance waves is recorded; the scale of an earthquake is predicted through calculating the periods of the resonance waves in a resonance packet, and an earthquake generating time is predicted through the growth waveforms of the resonance waves, and epicentral distances are determined through the time difference of the fast waves and the slow waves in the resonance waves; and the epicentral distances determined by the monitoring precursor stations are drawn into corresponding straight lines, and the intersections of the straight lines are the centers of earthquake regions where earthquakes are predicted to occur, and therefore, early warning notices can be emitted timely. The resonance packet wave spectrum monitoring data-based earthquake prediction method has the advantages of wide applicability and universality. With the method adopted, a long-awaited new era of earthquake prediction will come.

The invention relates to an infrared detector blind pixel detection system and method. The method comprises the following steps: S1) carrying out infrared detection to generate infrared detection data; S2) collecting the infrared detection data according to the set time interval, and generating a detection image correspondingly according to the collected infrared detection data; S3) carrying out singularity detection on each pixel in the detection image, and recording the singularity appearing probability of each pixel to a singularity probability table; and S4) taking the pixels, the singularity appearing probability of which is higher than or equal to the set value in the singularity probability table, as blind pixels, and carrying out blind pixel compensation. Compared with the on-site repair and user-calibrated modes, the method is high in efficiency and detection precision, and not easy to cause error judgment.

The invention provides an automatic judgment system of long jump pedal foul. The automatic judgment system of the long jump pedal foul aims at overcoming the defects of poor accuracy and strong subjectivity of the existing judgment system of the long jump pedal foul. The automatic judgment system comprises a pedal and an automatic judgment device, wherein the pedal comprises a foul line. The automatic judgment system is characterized in that the automatic judgment device comprises a microprocessor, a memory, a high speed camera, an operation panel, an indicator light, a display screen and a power supply; the microprocessor is connected with the memory, the high speed camera, the operation panel, the indicator light and the display screen respectively; and the power supply is connected with the microprocessor, the memory, the high speed camera, the operation panel, the indicator light and the display screen respectively. The automatic judgment system of the long jump pedal foul has the advantages of enabling the long jump pedal foul to be automatically judged, being high in accuracy and not easy to cause misjudgment, enabling judgment to be intuitive and enabling athletes to be sincerely convinced for self-behaviors of the pedal foul through playback and capturing functions.

The invention discloses a level deciding device for pulse-width signal and consists of a pulse-width signal generating circuit, a reference voltage reference generating circuit and a deciding circuit. The pulse-width signal generating circuit generates a first pulse-width signal; the reference voltage generating circuit generates a reference voltages signal and the deciding circuit is electrically connected with the pulse-width signal generating circuit and the reference voltages generating circuit and decides the high and low level value of the first pulse width signal according to the reference voltage so as to generate a second pulse-width signal.

The invention discloses an open-circuit fault diagnosis method for a power tube of an electric drive system based on a grey prediction theory. The electric drive system comprises a double-winding permanent magnet motor, two sets of three-phase bridge inverters and a fault diagnosis module. The method comprises the following steps of firstly, forming an original sequence of a gray prediction modelby collecting current data; predicting the current value of the next moment by using a mathematical model of grey prediction; and then the difference between the prediction data and the original datais calculated through a fault diagnosis device and compared with a set threshold value so that whether a fault occurs or not is judged, a fault signal is fed back, and the current values at the time of 3T/4 and the time of T/2 from the beginning of the fault are detected to judge the fault position. The method is high in reliability, does not need an additional sensor, and can achieve the quick and accurate diagnosis and positioning of the open-circuit fault of the power tube of the electric drive system.

The invention discloses a method for predicting the photographing time point and a device adopting the method. The method comprises the following steps: detecting, judging the face, classifying the face, generating a difference value if the face is the non-frontal face so as to calculate the difference value, calculating the difference value, calculating the difference values which are adjacent in sequence by using a calculating unit, judging the change, increase and decrease of the difference values which are adjacent in sequence so as to predict the photographing time, and informing the shutter of photographing. The device adopting the method comprises a face judging unit, a calculating unit, a training database and a shutter. The method and the device can be used for predicting the point of time when the non-frontal face rotates and changes into the frontal face, so that the frontal face can be photographed more successfully, and the non-frontal face can not be misjudged easily as the frontal face.

The invention discloses a human body movement measuring device, comprising a body, a human body impedance measuring unit, and a movement analysis unit. The body is used for fixing the units herein andbearing a human body. The human body impedance measuring unit is used for measuring biological impedance of the human body. The movement analysis unit is used for carrying out statistical analysis onthe human movement according to a waveform of the biological impedance of the human body. The invention also discloses a human body movement measuring method based on the human body movement measuring device; the human body movement measuring method statistically analyzes human body movement via biological impedance wave of the human body. The invention additionally discloses a human body movement measuring system based on the human body movement measuring device; the human body movement measuring system comprises at least two human body movement measuring devices, a remote server, a statistical unit and a display unit.

The invention discloses an ultrasonic atomizer and a water shortage detection device. The ultrasonic atomizer comprises a water tank (1), an atomizing sheet (2), a sensing antenna (3) and a water shortage detection device (9), wherein the sensing antenna (3) is in contact with a bottom shell of the water tank and is used for sensing whether water exists or not; the atomizing sheet (2) is mounted at the bottom of the water tank; the water shortage detection device (9) comprises a microprocessor (8), a boosting module (7), a signal rectifying and filtering module (4), a signal amplifying module(5) and a signal shaping module (6); the microprocessor (8) outputs a 3MHZ frequency to the boosting module (7); a signal is transmitted to the atomizing sheet (2) through a capacitor after voltage inversion; and when the water exists in the water tank, the high-frequency voltage signal is radiated to the sensing antenna (3) through the water as a carrier, the sensing antenna (3) obtains the signal, a direct current level is obtained through the signal rectifying and filtering module (4), and a level signal with the voltage higher than 2.5V is obtained after the direct current level passes through the signal amplifying module (5).

The invention discloses a primer pair and a kit used for carrying out early stage gender authentication of a chicken, and application of the primer pair and the kit. The invention provides a specific primer pair which consists of a single stranded DNA molecule disclosed by SEQ ID NO.1 and a single stranded DNA molecule disclosed by SEQ ID NO.2. The invention also provides a method for authenticating the gender of the chicken. The method comprises the following steps of: taking the genome DNA of a chicken to be detected as a template, adopting the specific primer pair to carry out PCR amplification, if an amplification product is obtained, proving that the chicken to be detected is a cock or is a candidate cock, and if two amplification products are obtained, proving that the chicken to be detected is a hen or is a candidate hen. Compared with the prior art, the invention has the following advantages that the hen and the cock have a difference of band amounts, and the difference of characteristic bands is 400bp. Through agarose electrophoresis detection, operation is simple and convenient, wrong judgment is unlikely to occur, an authentication result is quick and accurate, and therefore, the primer pair, the kit and the method can be conveniently popularized and used by the grass roots and have a wide market application prospect.