65results about How to "Judging objectively" patented technology

The invention discloses a tire X-ray image detection and identification method and system. The detection and identification method comprises an offline training step and an online detection step. The offline training step comprises: (11), establishing a training sample; (12), performing defect detection on an image in the training sample, and computing defect information of the training sample; (13), performing quantization coding on the defect information of the training sample to obtain a training sample feature vector; and (14), training the training sample feature vector and a grade judging result into a grade judging model. The online detection step comprises: (21), collecting a to-be-detected tire X-ray image in real time; (22), performing detect detection on the to-be-detected tire X-ray image, and computing all to-be-detected tire defect information; (23), performing quantization coding on all the to-be-detected tire defect information; and (24), computing the to-be-detected tire feature vector and the grade judging model. According to the detection and identification method, artificial subjective factors are avoided, the judgment is more objective and unified, and the judgment efficiency and accuracy are both improved.

The invention discloses a fatigue driving detection method based on electroencephalogram signals, comprising the following steps: S1, collecting the electroencephalogram signals of a driver in real time during driving, and removing blink artifacts to get EEG brain wave signals; S2, converting the EEG brain waves of the time-domain signals to a frequency domain, calculating the energy value of the feature wave in each frequency domain section in the brain waves, and determining the fatigue degree according to the amount of relative energy; S3, designing a BP neural network classifier to identify the feature signal of the fatigue degree; and S4, estimating the fatigue index and the fatigue degree. According to the fatigue driving detection method based on electroencephalogram signals put forward by the invention, the fatigue degree of a driver is judged in real time by analyzing the electroencephalogram rhythm of spontaneous electroencephalogram signals, and judgment is accurate, objective and direct.

A medical question contents automatic selection system for automatically selecting medical questions allocated to a patient from the medical questionnaire made by a medical professional. The medical question contents automatic selection system compares the correct answer rate threshold of the medical questions preset by the medical professional with the correct answer rate of the questions to the patient so as to estimate the educational level of the patient about health, compares the vital sign measurement value threshold preset by the medical professional with the vital sign measurement value of the patient so as to estimate the condition level of the patient, and automatically selects medical questions allocated to the patient depending on the educational level and the condition level of the patient. The medical question contents automatic selection system can allocate medical questions to the patient depending on the educational level of the patient about health and disorder condition level of the patient without troubling the medical professional.

A voice packet communications system which objectively evaluates the voice quality in real time. Voice packets received from a network are input into an adjustment circuit. The adjustment circuit evens out the periods of the voice packets, detects the drop-out of voice packets, and inserts substitute packets in the voice packet sequence. Furthermore, the adjustment circuit outputs the number of voice packets that have dropped out, the number of substitute packets that have been inserted, and the number of voice packets that have accumulated inside the circuit, as quality information in each monitoring period. A decoding circuit converts the voice packets adjusted by the adjustment circuit into voice information. A judgement circuit calculates an index that indicates the communications quality of the voice packets from the quality information using a specified calculation formula.

A voice packet communications system which objectively evaluates the voice quality in real time. Voice packets received from a network are input into an adjustment circuit. The adjustment circuit evens out the periods of the voice packets, detects the drop-out of voice packets, and inserts substitute packets in the voice packet sequence. Furthermore, the adjustment circuit outputs the number of voice packets that have dropped out, the number of substitute packets that have been inserted, and the number of voice packets that have accumulated inside the circuit, as quality information in each monitoring period. A decoding circuit converts the voice packets adjusted by the adjustment circuit into voice information. A judgement circuit calculates an index that indicates the communications quality of the voice packets from the quality information using a specified calculation formula.

The invention provides a real time detection system and detection method for camera definition. The system includes a definition detection environment, a comprehensive access platform, a video collecting platform and definition detecting client terminal software. The detection method includes steps of 1, obtaining a video stream; 2, initializing related parameters of a to-be-analyzed image; 3, selecting a to-be-analyzed area; 4, transmitting the selected image of the to-be-analyzed area and the initialization parameters to a computer; 5, converting the image of the to-be-analyzed area to a standard input image according to the initialization parameters; 6, detecting an initial line STL of a wedge-shaped image for the standard input image; 7, detecting the limit line LML of the wedge-shaped image; 8, detecting the end line ENDL of the wedge-shaped image; 9, calculating a resolution line number Res; 10, performing data treatment and displaying in real time. The system and method provided by the invention is fair and accurate in judgment.

The present invention relates to a portable brainwave measuring method, which measures a weak brainwave signal detected from a human scalp through a noninvasive method to measure a degree of concentration through an analysis and process of the detected brainwave signal, and a control method enabled for short distance control of an electronic device or remote monitoring through the internet, by using the brainwave signal. An acceleration sensor is put on the head of a user, and a brainwave detecting means put on the head of a user to detect a brainwave and the acceleration sensor that outputs an acceleration value of three axes including XYZ axes as a signal of predetermined data are used to detect movement of the head, to thereby control the direction and speed of a brainwave-related device. In more detail, after a signal outputted from the brainwave detecting means is converted into a wireless signal and transmitted, a value of the wireless signal is inputted to a receiving unit of a display device and is expressed numerically. By setting the value inputted to the display device, a portable brainwave measuring device provides accurate device control. A control system is characterized in that a signal according to the slope direction of the head is detected and analyzed, and then numerically expressed by using 6 brainwave signals such as a delta wave (δ), theta wave (θ), alpha wave (α), SMR wave, beta wave (β) and gamma wave (σ), which are measured through the portable brainwave measuring device and the acceleration sensor put on the head. A wireless system, including a short distance wireless module for transmitting / receiving the wireless signal, includes a wireless receiving module, a signal analyzing unit, and a control output unit. Additionally, the wireless system is connected to a PC or controls a short / long distance external device.

The present invention discloses a phonetic recognition method includes the specific steps of: recording standard phonetic data in phonetic data base; indexing and selecting standard voice segment to be displayed, prompted or broadcast; outputting standard voice segment, inputting new voice segment; recognizing through comparing the input voice signal and that in the data base; judging and feedback displaying. The present invention utilizes the recognizable and un-recognizable voice simultaneously in accurate judgment.

The present invention relates to the technical field of culture and particularly discloses a simple method for comprehensively evaluating the quality of boar semen. The method comprises the following steps of: firstly, detecting four indexes, namely sperm density, semen quantity, sperm viability and malformation rate; secondly, calculating the total number of active sperms according to a calculation formula; and thirdly, evaluating the quality of the boar semen according to the total number of the active sperms. According to the method, the quality of the boar semen can be objectively and accurately evaluated.

The invention discloses an intelligent supervising system and supervising method for a construction site mud head vehicle. The supervision system comprises an intelligent control system. wherein the camera, the intelligent roadblock and the client are arranged above and on the side of a channel, the intelligent roadblock is arranged at an exit of the channel, the top camera collects license plateand / or cargo compartment sealing cover images, the side camera collects wheel and / or license plate images, and the intelligent roadblock receives a control instruction of the intelligent control system to open and close the roadblock; The intelligent control system judges the license plate, the wheel cleanliness and the conformity of the cargo compartment sealing cover and issues a control instruction; And the client displays the data of the intelligent control system to a supervisor. Particularly, the image recognition module of the intelligent control system is arranged on the 4G / 5G base station MEC system, so that the network congestion and the network delay can be reduced. The supervision method comprises the steps of data acquisition, data transmission, image recognition, control instruction generation and supervision management. The method has the advantages of being high in supervision efficiency, good in timeliness, objective in judgment and convenient to prevent and control.

The invention discloses a defect detection method and device and a computer readable storage medium, and the method comprises the steps: carrying out the image enhancement of a specified image, and obtaining a defect enhanced image; traversing pixel values on the defect enhanced image according to a defect threshold to obtain pixel values exceeding the defect threshold; performing connected domainsegmentation based on the pixel value exceeding the defect threshold to obtain a defect area; and performing defect judgment on the defect area through a linear threshold value to determine a defectstate corresponding to the defect area, so that the method has the characteristic of objectively and accurately evaluating the screen defects.

The invention relates to an inhibition quantitative detector and an inhibition quantitative detection method using the inhibition quantitative detector. A binocular dichoptic effect is achieved by adopting a polarizer, so that the influence of a color lens on an examination result is reduced; a sighting mark with content rather than a spot light source is adopted as a viewing target, so that whether two eyes are balanced can be easier and more objectively judged by a detected person; and the same visual effect of the two eyes is achieved by changing light transmittance of an ND filter, so that a visual inhibition degree can be conveniently and rapidly quantitatively detected.

The invention provides a method and device for analyzing quality of advertising traffic. The method comprises the steps of receiving an analysis request for the quality of advertising traffic with an advertising mark, obtaining traffic data corresponding to the advertising mark, and analyzing and processing relevant information corresponding to a pre-configured analysis field identifier in the traffic data corresponding to the advertising mark according to the pre-configured analysis field identifier to judge whether a corresponding exception exists in the relevant information corresponding to the pre-configured analysis field identifier or not; when judging that the exception exists in the relevant information corresponding to the pre-configured analysis field identifier, querying the exceptional relevant information corresponding to the pre-configured analysis field identifier exists in a white list or not; and if all exceptional relevant information corresponding to the pre-configured analysis field identifier exists in the white list, judging that the quality of the traffic data corresponding to the advertising mark is normal.

The invention discloses a determination method for emotional states. The method comprises: performing interactive dialogue with a user through a virtual character, obtaining voice information and hand motion information of the user, according to the obtained voice information and hand motion information, comprehensively determining emotional states of the user, wherein scene of the interactive dialogue between the virtual character and the user is provided by a VR module, the voice information is provided by a voice module, the hand motion information is provided by an auxiliary module, and a processor giving comprehensive determination according to the obtained information. The method develops initiative of the user, and a determination result is objective and clear, and the method is easy to realize, and has very high application and promotional value.

The invention provides a fraud possibility analysis method and device and a computer readable storage medium. The method comprises the following steps that sample videos are collected and the fraud marks are allocated to the sample videos; the image features and the audio features of each sample video are extracted, and the video features of each sample video are obtained through combination; a neural network is constructed according to the sequence length of the sample videos and the dimension of the video features; the neural network is trained by using the video features and the fraud markof each sample video and the training parameters are optimized so that a fraud possibility analysis model is obtained; the preset duration of facial video of an object to be analyzed is acquired; theimage features and the audio features of the video are extracted so as to obtain the video features of the video through combination; and the video features are inputted to the fraud possibility analysis model, the fraud probability and the no-fraud probability of the object to be analyzed are outputted and the output result of the highest probability value is taken as the fraud possibility analysis result of the object to be analyzed. Whether the character is suspected of fraud can be objectively judged by using the fraud possibility analysis method and device.

The application discloses a signal processing method and device. The method comprises the following steps: deploying a Wi-Fi probe in a preset region; collecting a Wi-Fi signal of to-be-detected equipment in the preset region; computing average signal intensity of the Wi-Fi signals of the to-be-detected equipment in the setting period; determining the dynamic threshold, wherein the dynamic threshold is used as the average signal intensity at the preset processing location; and judging whether the target group enters the target place in the preset region according to the dynamic threshold value. Through the signal processing method disclosed by the application, the problem that large error is existent in a judgment result of judging whether the target group enters the target place by usingthe detection data of the unprocessed Wi-Fi probe and the experience threshold in the prior art is solved.

An embodiment of the present application provides a fingerprint registration method and apparatus, and an electronic terminal. The fingerprint registration method includes: obtaining, according to a plurality of input fingerprint images, a plurality of registration evaluation data corresponding to multiple fingerprint images respectively, where the registration The evaluation data comprises at least one of the following: the image quality data of the fingerprint image and the fingerprint area data in the fingerprint image; calculating a variation range of a plurality of the registration evaluation data; and judging whether the variation range and a preset amplitude variation threshold. In the method, fingerprint registration is performed according to the determination result to determine whether the fingerprint image can be registered by calculating the relationship between the variation amplitude of the plurality of registration evaluation data and the preset amplitude variation threshold. The success rate of fingerprinting registrations for difficult-to-register populations increases the experience of such groups.

An electronic bank flow authenticity identification method is characterized by comprising the steps of obtaining to-be-verified electronic bank flow information provided by a user; converting the to-be-verified electronic bank pipeline information into character type data through coding, and storing the converted character type data in a hash table; obtaining a real data source randomly provided by a third party with public trust at least once; judging whether the transaction date and the transaction amount of the real transaction data of the fixed number exist in the hash table one by one, ifyes, judging that the electronic bank flow information to be verified is true, and if not, judging that the electronic bank flow information is false. According to the invention, the random real transaction data with a fixed number of transactions are obtained only from a third party; a large amount of real transaction data is acquired instead of the real transaction data, so that the economic cost can be saved, the real transaction data currently used for judging the electronic bank flow is unknown through a random acquisition mode of the real transaction data, and the authenticity of the electronic bank flow provided by a user can be judged more objectively, fairly and fairly.

The invention discloses an automobile driver brain visual load evaluation method and system based on a subtask, which belong to the technical field of brain load measurement. The problem that the brain load of the driver cannot be objectively evaluated in real time in the prior art is solved. The automobile driver brain visual load evaluation method based on a subtask comprises the following stepsthat when a simulated rear automobile and an automobile where a driver is located move in the same direction, and when the interval between the simulated rear automobile and the automobile where thedriver is located reaches a set threshold value, whether the driver makes the automobile decelerate or accelerate or not is judged; according to whether the action of decelerating or accelerating thevehicle is carried out or not and the reaction time under the condition of carrying out the action of decelerating or accelerating the vehicle, whether the time used by the simulated vehicle, the eyemovement angular velocity and the color of the simulated vehicle are correctly recognized or not is found, and the brain load of the driver is evaluated. According to the method, the brain load of thedriver can be objectively evaluated in real time.

The invention discloses a chemical indicative card identification and interpretation system. The system is characterized by comprising an image acquisition unit that obtains information of a chemical indicative card through a probe, a key unit used for turning on or off a device, activating and hiding a menu and adjusting device parameters, a signal processing unit that processes the information acquired by the probe through a mainboard chip and makes preparation for outputting a signal (graph) to a display screen, an image processing unit that judges a result of the current indicative card through an algorithm according to the information provided on the indicative card, a display unit used for displaying and outputting a real-time image of the chemical indicative card, displaying a menu operation of a key and displaying a judgment result, and a storage unit that automatically stores the image and the judgment result of the current indicative card according to a preset frequency. According to the system, the result of the chemical indicative card can be subjected to objective, accurate and real-time judgment; and the system is small in volume, light in weight and convenient in operation, has an integrated visual interactive interface, and provides a convenient tool for monitoring and sterilization work of hospitals.

The invention discloses a method for detecting the water dispersibility of basalt fiber wet-process chopped strands. The method comprises the following steps: taking fibers in a basalt fiber wet-process chopped strand suspension by adopting a filter screen; drying the filter screen and fibers on the filter screen; acquiring high-definition images of the fibers on the filter screen; and comparing characteristic parameters of the images with corresponding characteristic parameters of a standard image by an image processing unit to determine the dispersity of the fibers. The suspension can be prepared independently or can be extracted from a slurry pool; or even the fibers can be extracted and obtained from the slurry pool directly by using the filter screen. The detection time in the whole process is shortened and errors introduced in all links are reduced; the dispersion state of the basalt wet-process chopped strand is closest to the dispersion state of the basalt wet-process chopped strand in the wet-process felt production process; the detection of the dispersity of the basalt fiber wet-process chopped strands is realized by using the image processing technology; the errors caused by human factors are reduced; and the water dispersibility of the basalt fiber wet-process chopped strand is objectively, accurately and efficiently determined.

An image generation system and contactless communication media, a method for selling commodities, and a complex store provide that the purchase of commodities can be considered based on actual commodities offered at shops and images including the use state of the commodities. The system includes a tag corresponding to a commodity, a portable terminal and an image generation device. The tag stores image information of the commodity, and transmits the image information to the portable terminal when the portable terminal approaches. The portable terminal receives the image information and outputs the received image information to an image generation device. The image generation device generates an image including the use state of the commodity based on the image information of the commodity outputted from the portable terminal, displays the image including the use state of the commodity generated.

The invention discloses a method and device for determining effective points of a ternary compound oil flooding system. The method comprises the steps of receiving oil well actual production data andactual moisture contents of two stages of a polymer flooding stage and a ternary compound oil flooding stage respectively; fitting the actual production data to obtain moisture content model parameters, and calculating the water flooding fitting moisture content; calculating the difference value of the actual moisture content and the water flooding fitting moisture of the polymer flooding stage and the difference value of the actual moisture content and the water flooding fitting moisture of the ternary compound oil flooding stage respectively; calculating the difference value of the actual moisture content and the fitting continuous polymer-flooding moisture content of the whole stage; drawing a curve on the same graph for the calculated difference value data in sequence, and determiningthe effective points. When a water-flooding curve method and a water-flooding derivative curve method cannot find inflection points or the inflection points are not obvious, the method for determiningthe effective points of the ternary compound oil flooding system can be utilized to judge the effective time, and can also be combined with a direct observation method for comparison and judgment, and the result is more objective and more accurate. The calculation method is convenient and practical, and is suitable for being applied and popularized in a mine field.

The invention discloses a method for accurately representing material machining strength in a cigarette loosening and moisture-regaining process. The method comprises the following steps of: respectively acquiring tobacco before and after loosening and moisture-regaining, balancing and subsequently carrying out spectrum scanning, carrying out multiplicative scatter correction as well as two-order partial derivative and Norris derivative filtering on the spectrum data, carrying out main component analysis, respectively establishing class models of non-loosened and moisture-regained tobacco and tobacco with different loosening and moisture-regaining machining gradients, and finally calculating so as to obtain the class average values of the mahalanobis distance between the class models of the tobacco with the loosening and moisture-regaining machining gradients and the non-loosened and moisture-regained tobacco, and thus judging the machining strength. The qualitative and quantitative expression of the tobacco loosening and moisture-regaining strength is realized, the parameter adjustment and the control to the tobacco shred making line loosening and moisture-regaining process are facilitated, the degree that the conventional machining strength judgment is easy to be affected subjectively is reduced, and the internal quality change of the cigarette can be subjectively judged; and the operation method is simple and easy to carry out, and the technical effect is remarkable.

The invention provides an evaluation method and device based on a driving simulation system and a medium, and relates to the technical field of simulated driving. The evaluation method comprises the following steps: receiving a virtual driving task instruction; according to the virtual driving task instruction, determining at least one execution process, wherein the execution process comprises anexecution sequence identifier; generating virtual driving operation data in response to the virtual driving operation instruction; by applying the embodiment of the invention, when a student is subjected to practical training driving examination, the evaluation result can be obtained according to the at least one execution process, the virtual driving operation data and the preset judgment file, and whether the virtual driving operation data meets the preset requirements of the execution process or not is judged through the evaluation result, so that automatic judgment can be realized; the human cost is reduced, compared with a manual judgment mode, the method can ensure that the evaluation is objective, fair and comprehensive.

The invention relates to a public transport station judgment method and device, and belongs to the technical field of intelligent transportation. The public transport station judgment method comprisesthe steps of obtaining user behavior states of a plurality of mobile terminals; determining a target position; determining change information of the user behavior states of the plurality of mobile terminals at the target position; and determining whether the target position is a public transportation station or not according to the change information of the user behavior states of the plurality of mobile terminals at the target position. According to the invention, automatic judgment of the public transport station can be realized, so that the updating timeliness of the public transport station is improved, and the labor cost is reduced.

When a property for verifying a logic system is generated, a list of corresponding events is generated from specifications that the logic system should meet, an event of an undefined state in a first property is extracted from the list of events, and a property representing that the extracted event of the undefined state is improbable is generated as a second property that makes up for an event lacking in the first property.

The invention relates to a tailing pond spillway well monitoring method based on multi-sensor fusion. The method comprises the following steps of firstly, monitoring a tailing pond spillway well through a plurality of monitoring devices, transmitting monitored data to a computer in real time, carrying out fusion calculation on all the monitored data by the computer, specifically, setting different weight values for all detection indexes, performing weighted calculation on the states of all the monitoring indexes according to different weight values, and obtaining a final safety coefficient of a safe operation state of the spillway well; and finally, dividing the obtained final safety coefficient into a low risk, a medium risk and a high risk according to different intervals and the operation state, and making different early warning prompts and processing according to different levels of conditions. The monitoring method is optimized, fusion analysis is carried out on data transmitted by multiple sensors, and early warning is carried out according to the obtained parameters for evaluating the operation state of the spillway well, so that a more comprehensive and accurate result is obtained under various conditions in monitoring of the spillway well, and safe operation of a tailing pond is ensured.