2169results about How to "Reduce misjudgment" patented technology

The invention relates to a fall monitoring and warning method which comprises the following steps of: continuously acquiring at least two kinds of data in acceleration data, body angle data and plantar pressure data of a human body simultaneously; judging whether the acquired at least two kinds of data simultaneously meet the following corresponding conditions that: the acceleration is greater than a set acceleration threshold, the body angle is greater than a tilting angle threshold, and the plantar pressure is smaller than a pressure threshold; and if the conditions are met, starting to warm. The information acquired by a three-axis accelerometer, an attitude sensor and a pressure acquisition module is merged, the fall state can be judged more comprehensively, the accuracy degree of judgment is increased, and false judgment is reduced.

The invention discloses an automatic charging system of an electric car and a control method of the automatic charging system. The automatic charging system and the control method thereof have the advantages that monitoring is started from parking of the car, a charging connection reset device is disconnected after charging of the car, a user's bill is generated, charging plug butt joint and real-time charging process monitoring are included, all related data in the whole charging process are monitored by a software system portion, a hardware mechanical portion is controlled to complete a next action, the hardware portion sends new semaphore to the software portion for new command calculation after completing a command, the software portion is combined with the hardware portion, and accordingly, systematicness and intellectuality of the whole charging system are improved; a charging interface butt joint scheme based on machine vision and a machine learning algorithm is applied, and feasibility and accuracy in butt joint are guaranteed; a three-dimensional lead screw guide rail mechanism with a long stroke is used, the range of activity of a charging connector is widened, and the charging connector is suitable for multiple car models; the automatic charging system is provided with a payment code scanner, payment of related charges can be completed through a payment bar code scanning function of a cellphone-side payment application, and accordingly, intelligent payment is achieved.

The invention discloses a real time filtrating method for large-scale garbage message based on the content, including the steps as following: 1, pre-filtrating by using the black list and the white list; 2, carrying out the online filtrating by using the filtrating module based on the frequency; 3, carrying out the fast filtrating for the message content by using the method of twice hashing; 4, carrying out the pretreating of the message text for suspicion message, and converting the same into the phase vector; 5, judging the suspicion message by using the method of combination of Naive Bayesian classifier and support vector classifier. The invention can greatly improve the filtrating speed of garbage message, and efficiently reduce the produced erroneous judgement rate in the conventional key word filtrating method; can efficiently solve the problem of group sending garbage messages with malicious intent in the short time; can efficiently avoid to mistake the common message as the garbage message so as to reduce the erroneous judgement, and efficiently improve the filtrating accuracy of whole system by analyzing the message content on the semantics.

The invention discloses a video-based detection and recognition system and a method of vehicles. The system comprises a target characteristic database module, a moving target detection module and a vehicle recognition module. A vehicle sample target characteristic database is built to store data by virtue of the target characteristic database module. After a moving target zone is detected by the moving target detection module, the vehicle recognition module builds a search window and recognizes vehicles. The system and the method resolve the technical problems of poor vehicle distinguishing capability and low distinguishing accuracy of the video-based vehicle detection under the condition of obvious background movement, vehicle adhesion, vehicle occlusion and the like in the prior art. The video-based detection and recognition system and the method of the vehicles have the advantages of excellent vehicle distinguishing capability and high accuracy.

The invention relates to a method for detecting the residual quantity of pesticides in a vegetable sample by utilizing a liquid chromatogram-high resolution mass spectrometer, which belongs to the technical field of a food safety detecting. The method provided by the invention comprises the following steps: 1) extracting the residual acetonitrile in the sample; 2) carrying out the dispersion solid-phase extraction and purification on a sample extraction solution; 3) detecting by utilizing the liquid chromatogram-high resolution mass spectrometer; and 4) analyzing the result. In the method provided by the invention, the quality accuracy of 27 pesticides is less than 1ppm; the detection limit is 0.01-0.05mg/kg; the addition recovery rate is 80-100%; and the relative standard deviation is less than 15%.

The invention discloses a method for identifying abnormal motion in a video. The method comprises the steps that an area of which the pixels have certain change is extracted from a video sequence through a foreground extraction module; a movement target detected by the foreground extraction module is judged through a pedestrian judgment module, and a rectangular box judged to be a pedestrian is kept and transmitted to a target tracking module; each identified target in the scene is continuously tracked in a multi-target way through the target tracking module; the posture of each target tracked by the target tracking module within each frame of time is judged through an abnormal behavior trigger module and behavior analysis is performed on the abnormal behavior; and the abnormal behavior is processed and computed into a video segment through a behavior identification module, and the abnormal behavior is alarmed and the behavior type of the abnormal behavior is identified. The invention also discloses a system for identifying the abnormal motion in the video. The problem that abnormal behavior (falling down) detection of massive video data cannot be performed in real time can be solved and the detection accuracy can be greatly enhanced.

A method and device for position detection are disclosed. A self-capacitance detection can be performed by a sensing device. According to the result of the self-capacitance detection, a first mutual-capacitance detection can be performed for determining one or more first 1-D positions. According to the result of the first mutual-capacitance detection, a second mutual-capacitance detection can be performed for determining one or more second 1-D positions corresponding to each first 1-D position. One or more 2-D positions can be provided according to the one or more second 1-D positions corresponding to each first 1-D position. Besides, during the self-capacitance detection, the first mutual-capacitance detection, and the second mutual-capacitance detection, a touch related sensing information corresponding to a touch that covers a wide area can be neglected for palm rejection.

An automatic snapshot system for highway traffic offence comprises a front snapshot unit, a network transmission unit, a detection unit and a management server, wherein the front snapshot unit obtains real-time video in a detected scene, the network transmission unit transmits the real-time video to the detection unit; the detection unit after obtaining the video uses object segmentation and trace tracking and analysis to find vehicles having abnormal behaviors such as reversing, going in a direction not allowed by traffic regulations and illegal parking, providing the position of the target vehicle, controlling a holder to rotate so that the license plate of the target vehicle is in the image capture center of the front snapshot unit, controlling the front snapshot unit to zoom a lens and adjust the size the target in the image to facilitate license plate identification and analysis; and after the license plate is determined, the front snapshot unit controlling an electric zoom lens to restore to a wide angle monitoring mode and snapshot distant view, close view and license plates, and uploading the snapshot results to the management server unit.

The invention provides a method and device for recognizing malicious users, and belongs to the field of network technology. The method includes the steps that historical behavior characteristics of a user to be recognized are acquired; a user classification model is acquired, wherein the user classification model is generated through classifier model training based on historical behavior characteristics of malicious users and historical behavior characteristics of normal users; the historical behavior characteristics of the user to be recognized are input into the user classification model to determine whether the user to be recognized is a malicious user or not. Through the user classification model obtained through classifier model training, the behavior characteristics of malicious users can be automatically analyzed according to the characteristics of a machine learning algorithm; compared with the prior art, the method and device have the advantages that the recognition process is independent of the subjective cognition of the human, and the coverage scope of behavior characteristics is wider, so that misjudgment and missed judgment are effectively reduced, and the accuracy of recognizing malicious users is improved.

The invention relates to a howling detection method which comprises the following steps: an input voice signal frame is converted from a time domain to a frequency domain, and a plurality of peak value frequency points that are maximum in energy in the voice signal frame are chosen; according to probability of occurrence of each peak value frequency point in a preset time period and a peak to average value ratio of each peak value frequency point, whether each peak value frequency point is a suspected howling point is determined; continuous occurrence frequency of each peak value frequency point being determined as the suspected howling point is subjected to confidence counting operation; whether the suspected howling point is a real howling point is determined according to a result of comparison between a confidence count and a confidence count threshold.

Provided are a man-machine interaction method and man-machine interaction equipment based an acceleration sensor and motion recognition. The method includes the following steps: (1) a three-shaft acceleration sensor located on the man-machine interaction equipment collects X-axis acceleration data, Y-axis acceleration data and Z-axis acceleration data, and conveys the X-axis acceleration data, the Y-axis acceleration data and the Z-axis acceleration data to a controller of the man-machine interaction equipment; (2) the controller pre-processes the acceleration data, and removes noise and gravitational acceleration influences to obtain a precise X-axis acceleration sequence, a precise Y-axis acceleration sequence and a Z-axis acceleration sequence and to cache the acceleration sequences; (3) an acceleration threshold value is set, when the maximum of the X-axis acceleration sequence, the Y-axis acceleration sequence and the Z-axis acceleration sequence in cache is larger than the acceleration threshold value, the step (4) is carried out, and otherwise the step (1) is carried out; (4) motion mode recognition is carried out on the obtained X-axis acceleration sequence, the Y-axis acceleration sequence and the Z-axis acceleration sequence by means of a pre-trained classifier; (5) each motion mode corresponds to a trigger command, and the controller executes the corresponding trigger command according to a result after the motion mode recognition and provides corresponding feedback for a user; and (6) the steps from (1) to (5) are repeated.

The invention provides an automatic detection device suitable for multi-diameter pipeline welding lines. A carrier dolly module (3) of the invention is arranged on a flexible track module (1); a scanner module (2) is fixed on the upper part of the carrier dolly module (3) through a vertical plate (8) and a horizontal plate (9); a probe combination module (4) is fixed on the lateral side of the carrier dolly module (3) through a tracking module (5); an encoder (6) is fixedly arranged on an angle adjusting block (3-4) of the carrier dolly module (3); and a motor (7) is fixedly arranged on a reduction gearbox (3-7) of the carrier dolly module (3). The invention can be applied to the diameter welding line detection of a series of pipelines; the efficiency of the detection technique of the invention in the pipeline girth welding line non-destructive detection is much higher than that of manual ultrasonic detection; and the invention is obviously superior in detection speed, environmental radiation pollution reduction and labor intensity decrease.

The invention provides a static analysis method and a static analysis device for an Android application program and relates to the technical field of security detection. The static analysis method includes: S1, unzipping an application program to be detected to obtain a Smali file; S2, traversing the Smali file to acquire a source code information and structuring a control flow diagram and a data flow diagram of the source code information; S3, traversing the control flow diagram and the data flow diagram according to a malicious behavior judging engine to respectively match an API (application program interface) of the application program with another API in a predefined malicious behavior library and marking the successfully matched API of the application program as the malicious behavior API; S4, calculating the malicious measuring value of the application program to be detected; S5, matching the malicious measuring value with a predefined malicious degree index to acquire risk level of the application program to be detected. Through behavior analysis and comprehensive judgment of combined rules, misjudgment rate of virus detection upon the Android application program can be reduced.

The invention is suitable for the technical field of household appliances, and provides a method for judging leakage of a refrigerant in an inverter air conditioner. The method comprises the following steps: obtaining temperature T of a coil pipe of an indoor unit before the compressor of the inverter air conditioner is operated; obtaining temperature T1 of the coil pipe of the indoor unit and indoor temperature TC1; judging whether the absolute value of the difference of T and T1 is smaller than a first temperature threshold value d1 and moreover the absolute value of the difference of TC1 and T1 is smaller than d2, if so, adjusting the inverter air conditioner from the topmost wind gear to the lowest wind gear; operating the inverter air conditioner at the lowest wind gear for a second set time t2; obtaining the temperature T2 of the coil pipe of the indoor unit and the indoor temperature TC2; and when the absolute value of the difference of T and T2 is smaller than d1 and moreover the absolute value of the difference of TC2 and T2 is smaller than d2, determining the leakage of the refrigerant in the inverter air conditioner, and stopping the operation of the compressor. The technical scheme provided by the invention has the advantage of low misjudgment rate.

The invention discloses an intelligent interactive multimedia display system and an implementation method thereof. The intelligent interactive multimedia display system is composed of a commodity display medium, an interactive display unit, a face detection and identification unit, an interaction triggering unit, a signal processing unit and a cloud server. Face detection is conducted in real time by adopting a built-in camera of an all-in-one touch screen to achieve active interaction, and therefore the possibility that commodities are concerned can be greatly increased; compared with the traditional mode that people counting or personnel identification is conducted by an infrared sensor, a WIFI probe and the like, the system is more precise and effective, and various invalid interferences can be eliminated; and differentiation interaction is conducted through a face identification technology, advertising copies can be automatically changed according to the age and the sex, the purpose of enabling advertising to be targeted is achieved, the interest of consumers is increased, and the sale success rate is increased.

The invention discloses a timely and high-efficiency crawling method for internet information and belongs to the technical field of information. The method comprises the following steps: (1) setting a seed address, crawling and storing webpage information, and ensuring navigation pages; (2) carrying out more than once crawling on each navigation page, and analyzing and labeling the crawling webpage; (4) building a theme judgment model and a navigation page change time series prediction model of each website; (5) predicting next time change time of each website navigation page, ensuring next crawling time, crawling the navigation page and extracting a subpage address and an anchor text which are not crawled; (6) adopting the built theme judgment model to judge the extracted subpage address and the anchor text in the last step, and respectively processing the extracted subpage address and the anchor text according to a judgment result; (7) based on a new related page of the crawled theme, forming or updating a present change time series of each website navigation page, and ensuring next crawling time to carry out webpage crawling. The timely and high-efficiency crawling method for the internet information guarantees novelty and topicality of collected information under a small load.

The invention discloses an automatic snapshot device and method for illegal parking. The automatic snapshot device comprises a front-end snapshot unit, a transmission network and a management server. The front-end snapshot unit comprises an electric zoom lens, an intelligent camera and a tripod head. The electric zoom lens is arranged at the front end of the intelligent camera and connected with the intelligent camera. The intelligent camera is arranged on the tripod head and further connected with the management server through the transmission network. The automatic snapshot method comprises the steps that the front-end snapshot unit monitors the whole monitoring area through a wide-angle lens and tracks the motion trail of all vehicles within the monitoring area; when a parking event is found through the trail tracking mode, the barycenter and area of the parking vehicle are determined; the rotation angle and pitch angle of the tripod head are determined through geometric operations; the tripod head is controlled so that the target can be located in the center of an image; the size, in the image, of the target is adjusted through lens zoom control so as to be suitable for image analysis, namely license number recognition; after the license number recognition is completed, the intelligent camera controls the electric zoom lens to restore to a wide-angle monitoring mode.

A pattern shape evaluation method and semiconductor inspection system having a unit for extracting contour data of a pattern from an image obtained by photographing a semiconductor pattern, a unit for generating pattern direction data from design data of the semiconductor pattern, and a unit for detecting a defect of a pattern, through comparison between pattern direction data obtained from the contour data and pattern direction data generated from the design data corresponding to a pattern position of the contour data.

The invention relates to the communication field and discloses a method and device for feeding back ACK/NACK (Acknowledgement/Negative Acknowledgement) information, which is used for improving the system performance of an LTE-A (Long Term Evolution-Advanced) system. By combining the characteristic that the LTE-A system can use the carrier aggregation technology, the method redesigns a mapping relationship between ACK/NACK information and ACK/NACK feedback bits, and the ACK/NACK feedback bits are used for indicating the number of downlink subframes correctly received by UE (User Equipment) rather than the number of data packets correctly received. Thus, the invention can accurately reflect the receiving situations of multiple downlink subframes in multiple downlink carriers based on the limited K-bit ACK/NACK feedback bits when the ACK/NACK feedback information and other uplink feedback information are simultaneously transmitted through a PUCCH (Physical Uplink Control Channel), and can minimize the mistaken judgments of a base station on data receiving situations of user terminals caused by the many-to-one mapping of the ACK/NACK feedback bits so as to avoid unnecessary retransmission, thereby improving the transmission performance of the system.

The invention provides an engine tail gas after-treatment control system and control method. The engine tail gas after-treatment control system and control method are applied to after-treatment of tail gas. The control system comprises a tail gas post-processor, an engine electric control unit, a diesel injection device, a urea spraying device and a post-treatment electric control single element.The control method is based on the measured values of different sensors and the working condition parameters of an engine, and is based on a pre-calibrated data model to complete the four functions ofDPF active regeneration controlling, DPF accumulated dust cleaning, SCR urea injection controlling and system fault diagnosing. The system is composed of a software system and a hardware system, wherein a transient segmented linear interpolation algorithm is adopted for MAP data, and the calculation precision and the execution efficiency are improved to a great extent. A DPF differential pressuresensor adopts a differential pressure dynamic integral and time-sharing comparison method, so that the condition that the differential pressure value is unstable and the fluctuation is large due to the fact that the differential pressure sensor is influenced by various factors is solved, the robustness of a control strategy of the tail gas after-treatment system is enhanced, and the misjudgment of regeneration is reduced to a great extent, so that the improper active regeneration operation is reduced, and the oil consumption is reduced.

The invention provides a method and system for identifying a scalping user in a taxi service. The method comprises the following steps: extracting a cell phone number of a user in a taxi order, and start and finish time of the order; and then acquiring position data of the cell phone number at a base station cell between the start and finish time of the order from a telecom operation server that the cell phone number of the user belongs to, and judging whether the position data of the base station cell are changed between the start and finish time of the order, if not, then the user is the scalping user, wherein the user is a driver or a passenger. The method and system for identifying the scalping user in the taxi service provided by the invention belong to the technical field of mobile Internet, can effectively recognize scalping users in the taxis service, reduce the enterprise loss and system misjudgment, and improve the recognition efficiency.

A cotton developmental phase automatic identification method based on image classification and target detection belongs to the field of agricultural meteorological observation. With development of image processing and deep learning technologies, transformation of agricultural meteorological observation mode from manual observation to automatic observation becomes possible. In order to realize automatic observation for a cotton developmental phase, the invention provides a cotton developmental phase automatic identification method based on image classification and target detection. In method provided by the scheme, different features existing in each developmental phase image are observed and analyzed, then, automatic identification of a three true leaves phase, a five true leaves phase anda squaring phase is realized through image classification based on deep learning, further, flowers and cotton fiber in the image are detected automatically through deep target detection, and finally,results of two algorithms are integrated, and automatic identification for a complete cotton developmental phase is realized. The method provided by the scheme can realize rapid and accurate automatic identification for the cotton developmental phase and has very important application value.

The embodiment of the invention discloses a voice/music identifying method and voice/music identifying equipment, and relates to the technical field of audio; the voice/music identifying method and equipment are capable of enhancing the accuracy for identification of music. The voice/music identifying method comprises steps of: judging whether the energy change rate of the appointed amount of frames before the current frame and the energy of the current frame meet the condition of percussion music, and if so, determining an encoding mode of the current frame as a music mode. The embodiment ofthe invention is mainly applied to the voice/music identification process of audios, and the accuracy for the identification of the music can be enhanced.

The invention discloses a digital image noise reduction method and device. The noise reduction method includes: performing noise analysis on an input image, calculating the difference frame/field image between input image frames or between two fields separated by one field, and based on the difference frame Image analysis of the frame/field noise statistical characteristics; perform noise analysis, calculate noise power, determine noise signs and determine noise classification; adaptively select frame/intra-field noise reduction or/and frame/inter-field noise reduction methods. The device for digital image noise reduction includes: frame memory, noise analysis unit and frame/field noise reduction unit. Since the present invention selects and adopts frame/intra-field and/or frame/inter-field noise reduction methods by analyzing the noise characteristics of difference frame/field images, and selects appropriate weights according to the characteristics of noise and image information, the noise reduction Finally, images with better visual effects are obtained.

The invention relates to food mechanical equipment, in particular to a small-scale movable computer vision automatic navel orange separator, which consists of a staggered arrangement and feeding mechanism, a differential pair twist automatic arrangement mechanism, a conveying line, a classified air blast actuating unit, a mechanical, optical and electric automatic control system, a computer visionsystem and a camera photoelectric box part, wherein the staggered arrangement and feeding mechanism is positioned aside one end of the automatic separator; the differential pair twist automatic arrangement mechanism is angulated at 90 DEG with the staggered arrangement and feeding mechanism and arranged at one end of the conveying line which is connected to the right of the differential pair twist automatic arrangement mechanism; the classified air blast actuating unit is arranged on one side of the conveying line; the mechanical, optical and electric automatic control system is arranged on one side of the front part of the conveying line; and the computer vision system and the camera photoelectric box part are arranged above the middle of the conveying line. The invention replaces the conventional manual separation of the machine with automatic separation, thereby improving the production efficiency and the separation precision and reducing the cost.