46results about How to "Accurate identification and detection" patented technology

The invention provides a flash malicious file detection method, which includes the steps: analyzing a flash file, and extracting virtual machine byte codes from the flash file; and matching the virtual machine byte codes with characteristic byte codes, making statistics on occurrence times of the virtual machine byte codes if the virtual machine byte codes can be matched with characteristic byte codes, and determining the flash file as a malicious file if the times exceed a threshold value. The invention further provides a flash malicious file detection device for implementing the method. By the aid of the flash malicious file detection method and the flash malicious file detection device, high detection efficiency and high detection precision are achieved.

The invention discloses a control system of a multifunctional man-machine interaction humanoid teaching robot. The system comprises a remote monitoring module, real-time protective modules, a speech module, a mobile phone APP module, a robot movement module, a wireless communication module, a steering engine integration module and a control module, the modules are connected with an embedded controller through wires for real-time data exchange, the remote monitoring module is located on the head of the robot, the real-time protective modules are located on the shoulders of the robot, the wireless communication module is located on the chest of the robot and composed of three communication module bodies, robot palms are arranged at the tail ends of robot arms, the chest of the robot is supported on robot feet through robot legs, and the robot movement module and the steering engine integration module receive control instructions sent by the control module. The biological humanoid teaching robot is multifunctional, richer in man-machine interaction and high in environment adaptability, and the adaptability range to people of different ages is wide.

The invention discloses a control method suitable for an SMD extensible intelligent warehousing system. The control method can be used for achieving a function of simultaneous storage and takeout of multiple SMD trays and adopts an RFID reader-writer and bar code scanning dual management recognition system to achieve recognition, scanning and information input of the trays. Full-automated management and material storage and takeout are achieved, inventory information is automatically collected, and the inventory management cost is effectively reduced. A storage space with large storage volume, good leakproofness and excellent storage environment is provided. The expansibility is higher, and each unit consists of multiple tray frames and is provided with an expansion interface for facilitating warehouse scale enlargement. By means of the control method, intelligent storage and management of the SMD trays can be efficiently achieved, and manpower and material costs are greatly saved.

The present invention relates to an information entropy-based snowflake noise detection method. The overall distribution state of the gray values of pixels is obtained; a previous image frame and a subsequent image frame are converted into a gray space; the gray values of all the pixels are clustered into n intervals, and the distribution state of the gray values of the pixels is displayed in the form of a grey level histogram; the information entropy of the previous image frame and the subsequent image frame are calculated; and whether snowflake noises exist in a video stream is judged according to the change amplitude of the information entropy of the previous image frame and the subsequent image frame. According to the method of the invention, when an actual calculated value is not smaller than a set threshold value, and the value of the subsequent image frame is larger than value of the previous frame, or when the actual calculated value is smaller than the set threshold value, while, the previous frame is marked as a noise frame, it can be finally determined that snowflake noises actually exist, otherwise, it is considered that the images are normal images. The method of the invention has high accuracy in video snowflake noise detection and also lays a good foundation for snowflake noise elimination for obtaining a better visual effect.

The invention discloses an intelligent garbage classification device, comprising a garbage classification unit and a garbage identification unit mounted on the garbage classification unit. The garbage identification unit comprises an image preprocessing module, an image extreme-point detection module, an image feature-point positioning module, a principal direction determination module and a feature extraction module. From all extreme points, the image feature-point positioning module determines extreme points as feature points by rejecting noise-sensitive low-contrast points and unstable edge points. The principal direction determination module is used to connect any two adjacent peak values in a gradient direction histogram about the feature points so as to form many sub line segments, to merge, in the length direction, the adjacent sub line segments with similar slopes to form a line segment, and to take the direction of the optimal line segment among the many line segments as the principal direction of the feature points. The invention has the advantages of accurate classification and high speed.

The invention relates to the technical field of biology, in particular to a reagent which can simultaneously distinguish and detect four animal insect-borne diseases as well as a preparation method and application of the reagent. The animal insect-borne disease multi-RT-PCR distinguishing and detecting reagent comprises four pairs of specific primers, and the respective amplification target fragment lengths of Bluetongue virus (BTV), epizootic haemorrhagic disease virus of deer (EHDV), Vesicular stomatitis virus (VSV) and Akabane virus (AKV) are 351bp, 536bp, 300bp and 250bp. The VP7 of the BTV, the EHDV, the VSV and the AKV and a conservative fragment of an N gene are respectively selected as targets, and the primer Express software and the primer prere 5.0 software are applied to deign and combine a primer. An optimal matching and screening test and a multi-RT-PCR test are carried out on a plurality of pairs of designed primers, and four pairs of primers which can carry out the distinguishing and detection on the four animal insect-borne diseases and have high amplification efficiency and good specificity can be obtained by a plurality of reaction condition optimization and comparison tests and verification tests. A kit formed by the reagent can obtain a qualitative distinguishing and detecting result in six hours after a sample is received. The multi-RT-PCR distinguishing and detecting reagent is a sensitive and reliable method for detecting the BTV, the EHDV, the VSV and the AKV in the clinical sample.

The invention discloses an intelligent air purifier. The intelligent air purifier comprises an air purifier and a scene recognition device mounted on the air purifier, wherein the scene recognition device comprises an image preprocessing module, an image extreme point detecting module, an image feature point positioning module, a main direction determining module, a feature extracting module and a scene determining module; the image feature point positioning module is used for determining extreme points of feature points by removing low-contrast points sensitive to noise and unstable edge points among the extreme points; the main direction determining module is used for connecting any two adjacent peak values in a gradient direction histogram about the feature points to form a plurality of sub line segments, merging the adjacent sub line segments with similar slope in a length direction to form a line segment and taking a direction of an optimal line segment among a plurality of line segments as a main direction of the feature points. The intelligent air purifier has the advantages of high scene recognition accuracy and fast scene recognition speed.

The invention discloses a metal oxide gas sensor array and a method for identifying the type and concentration of methanol or ethanol or isopropanol by using the gas sensor array. The sensor array comprises sensors coated with n different nickel oxide sensitive films, the nickel oxide sensitive films are prepared through a solution method, and the types and recognition performance of the metal oxide sensitive films are expanded through doping of different metal ions; Furthermore, thermal modulation resistance response signals of the sensor array to methanol, ethanol or isopropanol with different concentrations and types are tested through variable-temperature thermal modulation, and then normalization processing of thermal modulation sensitivity response signals is carried out; mapping (calibration) between methanol or ethanol or isopropanol with different concentrations and a normalized sensitivity signal is established through a supervised learning algorithm, then types and concentrations of the three alcohols are identified, and the method can be widely applied to quantitative identification and detection of types and concentrations of other volatile organic compounds very easily.

The invention relates to an intelligent repair system of a connecting rod mold visual detection machine. The intelligent repair system comprises a supplementary welding robot, a welding power source, adjusting slide rails, a cooling system, a laser tracking system, a 3D intelligent scanning camera, a supplementary welding workbench and a control system. The supplementary welding workbench comprises a workbench body, workpiece locating frames, ejector pins and adjusting drive devices. At least one adjusting slide rail surrounds the supplementary welding workbench to form a closed annular structure. The laser tracking system and the 3D intelligent scanning camera are arranged on the supplementary welding robot. The cooling system is arranged on a tackle and connected with the supplementary welding robot and the supplementary welding workbench. The welding power source and the control system are arranged on the side surface of the supplementary welding workbench. On one hand, connecting rod surface quality can be precisely recognized and detected effectively, and the efficiency for finding surface defects is improved; and on the other hand, the surface defects are precisely analyzed and calculated, then welding equipment is driven for welding operation, and therefore the precision quality of surface supplementary welding operation is improved.

The invention discloses a P-type metal oxide gas sensor and a preparation and application method thereof. The sensor consists of a ceramic base which is attached with test and heating electrodes and coated with P-type metal oxide nano particles. The method is characterized in that bacterial cellulose hydrogel is first immersed in the metal salt solution, is then subjected to standing, rinsed and then lyophilized, the obtained cellulose xerogel attached with metal ions is first calcined, mixture of the obtained flocculent metal oxide particles and volatile reagent is further ground, and lastly, the obtained metal oxide nanoparticle slurry is coated on a ceramic substrate and aged, and a desired product is obtained; The method is advantaged in that calibration processing is performed in advance, according to what kind of volatile organic compound (VOC) envelope the measured signal belong to in the calibration result, or from what kind of volatile organic compound (VOC) envelope the distance is the shortest, what kind of volatile organic compound the gas measured at the time is obtained. The method is advantaged in that the method is extremely easy to be widely commercialized for accurate identification detection of a variety of volatile organic compounds.

The invention discloses a multifunctional smart cleaning robot apparatus which comprises a cleaning robot apparatus and a scenario identifying device mounted on the cleaning robot apparatus. The scenario identifying device comprises a pre-treatment module for images, an image extreme point detection module, an image characteristic point positioning module, a principal direction determining module and a characteristics extracting module and a scenario determining module. The image characteristic point positioning module determines the extreme points of characteristics points by sorting out the low contrast ratio points sensitive to noises and the unstable marginal points. The principal direction determining module connects any two adjacent peak values in the histograms of oriented gradients for the characteristic points to form a plurality of sub-line sections wherein adjacent sub-line sections with similar gradients are consolidated in the length direction to form line sections. The direction of the most optimal one of the line sections is regarded as the principal direction of the characteristic points. According to the invention, high identifying accuracy for scenarios and high speed can be achieved.

The invention provides a process step correctness automatic detection method and system which are applied to a server, and the method comprises the steps: carrying out the semantic segmentation of a to-be-detected image based on a deep learning method, and obtaining the target type of each pixel in the to-be-detected image, wherein the to-be-detected image is an image frame in a video containing aprocess step execution process; determining a human body region and a working region in the to-be-detected image based on the target category; recognizing the human body area and the working area based on a deep learning method to obtain human body posture information and working task information respectively; determining process step information based on the human body posture information and the work task information; extracting a target feature vector of the process step information based on a deep learning method, and judging whether the process step is correctly executed or not based onthe target feature vector. The technical problems that in the prior art, the cost is high, and the process steps cannot be detected in real time are solved.

The invention generally relates to the field of manufacture of semiconductor integrated circuits, particularly to a bearing device and a method adopting the device to transfer wafers. In the bearing device and the method adopting the device to transfer wafers, a movable sensor is arranged on a bearing platform, the position of the sensor can be set as per wafers of different specifications and sizes, so that the movable sensor can be used for recognizing and detecting wafers of different sizes and accurately detecting and recognizing wafers with special shapes, and overlapped wafers in a bearing platform can be effectively avoided, the danger of crushing wafers as wafers cannot be recognized is reduced, the yield of products is improved, the process cost is reduced, little modification is carried out to the conventional equipment, and the compatibility of the device is stronger.

The invention discloses a method for detecting a human voice or a machine sound through a voiceprint engine to perform identity verification. The method comprises the following steps: A, receiving anaudio to be verified; B, preprocessing received audio information, wherein the preprocessing comprises the following steps: a) time length detection; b) silence detection; c) pre-emphasis; and d) framing and windowing; and C, carrying out voiceprint comparison, thereby judging identity information. According to the method, the human voice or the machine sound is detected through the voiceprint engine to carry out the identity verification. The voiceprint engine is capable of efficiently, rapidly and accurately identifying whether the detected sound is expressed by a speaker in real time or a recorded sound played by an audio player, and judging whether the sound is from a detected person or not, so that an application scene of voiceprint identification has higher accuracy and efficiency; and therefore, the method has a wide application prospect.

The invention discloses an intelligent intersection passing early warning system based on target detection. The intelligent intersection passing early warning system comprises a camera acquisition module, a control module, a display module, an alarm module and a power supply module. The system can perform target detection according to pictures shot by the intersection camera, displays moving objects such as vehicles, people and animals about to arrive at the intersection and about distances to the intersection in real time through the display module, and gives different levels of warnings according to different distances. The system can be installed on a curve part with insufficient line sight distance, a position with extremely small curve radius of a road in a mountain ridge area, an intersection (especially an intersection without signals) with poor intervisibility, a railway crossing and a narrow turning position of an underground parking lot, and traffic accidents caused by the fact that vehicles at the turning intersection cannot respond in time are avoided. Compared with an existing road reflective mirror, the problems that a reflective object is distorted and deformed, the mirror surface is easy to dirty and damage, the occupied space is large, and the attention of a driver is distracted are solved. Meanwhile, a target detection algorithm is optimized, and the target detection algorithm is applied to video detection of the intersection, so that a safer, more convenient and more intelligent intersection passing early warning system is provided for automobiles and pedestrians to pass through some special types of intersections.

The invention relates to a living body identification and living body identification model construction method and device. The method and device are also applied to a scene that a model-based computer system or electronic equipment performs true prosthesis identification detection on biological characteristics. The living body recognition method comprises the following steps: acquiring a group of image focus stack sequences focused at different depths of an object to be recognized obtained by shooting the same observation point position; based on a feature extraction model, performing feature extraction on each image in the image focus stack sequence to obtain a feature vector of each image in the same feature space; analyzing differences among the feature vectors of the images to obtain a feature difference distribution sequence of the images related to the image focus stack sequence; and inputting the feature difference distribution sequence into a pre-trained living body identification model, and identifying whether the to-be-identified object is a living body biological feature or a non-living body biological feature. The method has the advantages of high identification accuracy, simplicity in operation and convenience in application.

The invention relates to the technical field of biology, in particular to a reagent which can simultaneously distinguish and detect four animal insect-borne diseases as well as a preparation method and application of the reagent. The animal insect-borne disease multi-RT-PCR distinguishing and detecting reagent comprises four pairs of specific primers, and the respective amplification target fragment lengths of Bluetongue virus (BTV), epizootic haemorrhagic disease virus of deer (EHDV), Vesicular stomatitis virus (VSV) and Akabane virus (AKV) are 351bp, 536bp, 300bp and 250bp. The VP7 of the BTV, the EHDV, the VSV and the AKV and a conservative fragment of an N gene are respectively selected as targets, and the primer Express software and the primer prere 5.0 software are applied to deign and combine a primer. An optimal matching and screening test and a multi-RT-PCR test are carried out on a plurality of pairs of designed primers, and four pairs of primers which can carry out the distinguishing and detection on the four animal insect-borne diseases and have high amplification efficiency and good specificity can be obtained by a plurality of reaction condition optimization and comparison tests and verification tests. A kit formed by the reagent can obtain a qualitative distinguishing and detecting result in six hours after a sample is received. The multi-RT-PCR distinguishing and detecting reagent is a sensitive and reliable method for detecting the BTV, the EHDV, the VSV and the AKV in the clinical sample.

The invention discloses a porcine epidemic diarrhea virus G1 / G2 type RT-PCR identification primer set. The porcine epidemic diarrhea virus G1 / G2 type RT-PCR identification primer set comprises two pairs of specific primers of PEDV-BF and PEDV-BR and PEDV-F and PEDV-R respectively which have the base sequences shown in SEQ. ID. NO. 1-SEQ. ID. NO.4 in the sequence listing. Accordingly, the inventionalso provides a corresponding diagnostic kit. The porcine epidemic diarrhea virus G1 / G2 type RT-PCR identification primer set has the advantages that infection types of PEDV G1and G2 can be identified in the same reaction tube, thereby providing a quick and simple tool for accurate detection of swinery PEDV, laying a solid foundation for the clinical rapid identification of toxic strains of the PEDV G1and G2 types and the laboratorial epidemiology research, helping the pig industry accurately judge the causes of diseases and timely formulate a therapeutic schedule, and reducing the economic losses; experimental studies show that the porcine epidemic diarrhea virus G1 / G2 type RT-PCR identification primer set has good specificity, high sensitivity, good repeatability, short consumed time and low cost.

The invention provides an ultrasonic detecting device with a scene identification function. The ultrasonic detecting device comprises an ultrasonic detecting device and a scene identification device which is mounted on the ultrasonic detecting device. The scene identification device comprises an image pretreatment module, an image extreme point detecting module, an image characteristic point positioning module, a main direction determining module, a characteristic extracting module and a scene determining module, wherein the image characteristic point positioning module determines extreme points as the characteristic points through eliminating low-contrast points which are sensitive to noise and instable edge points. The main direction determining module is used for drawing a line between two random adjacent peak values in a histogram of oriented gradient related with the characteristic points for forming a plurality of sub-line-segments, combining the sub-line-segments which have approximate gradients and are adjacent in a length direction for forming one line segment, and utilizing the direction of the optimal line segment in a plurality of line segments as the main direction of the characteristic points. The ultrasonic detecting device has advantages of high scene identification precision and high speed.

The invention discloses an image scene mode generation device. The device comprises an image preprocessing module, an image extreme point detection module, an image characteristic point positioning module, a main direction determining module, a feature extraction module and a scene determination module, wherein the image characteristic point positioning module determines extreme points as characteristic points by rejecting noise sensitive low-contract points and instable edge points from different extreme points; and the main direction determining module connects two random adjacent peak values, in a gradient direction histogram related to the characteristic points, into a line to form multiple sub line segments, merges the adjacent sub line segments with similar gradients into line segments, and uses the direction of an optimal line segment among the multiple line segments as the main direction of the characteristic points. The generation device has the advantages that scene identification is high in precision and speed.

The invention discloses a method for detecting ovine babesiasis indeterminate species, Babesia motasi LZ subspecies and Babesia motasi NB subspecies as well as a detection kit thereof. The detection method takes ovine babesiasis thrombospondin-related anonymous protein TRAP gene as a detection identifier. The kit comprises six primers shown in SEQ ID No.1, SEQ ID No.2, SEQ ID No.3, SEQ ID No.4, SEQ ID No.5, and SEQ ID No.6, The method takes TRAP as the detection identifier, the difference between species and subspecies is large, the detection sensitivity is better than that of the current technologies. The kit can simultaneously amplify several target DNA molecules in a PCR reaction, and realizes a purpose of simultaneously discriminating and detecting a plurality of microbes in a same sample. Compared with the other molecule detection technologies, the method has the advantages of simpleness, rapidity, high efficiency and saved cost, is suitable for rapid detection of cause of disease, and can be a hotspot technology for scientific research and clinical diagnosis.

The related biological agent designed with target object as virus specific and conservative gene fragment of SVDV, VSV and FMDV, special the conservative fragment of VP1, N gene and 3D gene. Wherein, the RT-PCR detection agent includes three couple specific primers, and the amplification target fragment length is 126bp, 301bp and 189bp respectively. This invention can detect three viruses (SVDV, VSV and FMDV) in one reaction tube simultaneously.

The invention relates to the technical field of image processing, in particular to a water body shaking detection and analysis method and system based on image processing. The method comprises the following steps of acquiring multiple frames of water body initial images, removing a mirror image in the initial water body image to obtain an actual water body image, graying the actual water body image, and endowing the gray value of each pixel point in the actual water body image with the gray value corresponding to the gray level interval according to a preset number of gray level intervals to obtain a water body target image, calculating a difference value between two adjacent frames of water body target images to obtain a bubble image, acquiring the number of bubbles and the dispersion degree of the bubbles, and inputting the bubble number and the bubble dispersion degree into a preset water body shaking degree detection model to obtain the water body shaking degree. According to the method, the water body shaking phenomenon in the equipment air tightness detection process can be simply and quickly identified, so that subsequent workers can accurately detect the air tightness of the equipment.

The invention discloses a distribution network transformer real-time detection method and system based on a neural computing rod. Comprising the following steps: acquiring a to-be-detected distribution network transformer picture by using a yo3 target detection strategy, converting the to-be-detected distribution network transformer picture into image data, transmitting the image data to an image sensor for affine transformation, and outputting a corresponding actual position; acquiring transformer oil characteristic gas in a transformer by using a gas chromatographic analysis method, and carrying out pretreatment by fusing a WSMOTE strategy; constructing a positioning model based on a distributed sequential gas leakage source position evaluation strategy to detect the actual position of the distribution network transformer and the preprocessed characteristic gas, accelerating the positioning judgment process of the positioning model in real time in combination with an AI accelerator, and outputting a positioning result; and importing a positioning result into a Bayesian analysis model for secondary verification and judgment to obtain position information of the distribution network transformer, and displaying the result in real time by using a mobile terminal or a Raspberry Pi display. The neural computing rod of the AI accelerator is adopted to achieve the high-speed AI inference operation capability and the low-power-consumption capability.

The invention discloses a method for identifying and detecting porphyra red rot pythium pathogens, and belongs to the technical field of algae cultivation disease detection, identification and control. According to the method, specific primers are designed on the basis of genomic difference sequences of the porphyra red rot pathogens P.porphyrae and P.chondricola, and a single PCR detection method and a double PCR detection method for the two pythium pathogens are established. By the method, the P.porphyrae and the P.chondricola as well as the possibility of coexistence of the P.porphyrae and the P.chondricola can be identified and detected. The lower limit of single PCR primer P-for / P-rev detection is 103 copies, and the lower limit of C-for / P-rev primer detection is 102 copies. The lower limit of double PCR detection is 104 copies. According to the method, a specific and sensitive porphyra red rot pythium pathogen identification method is established, the P.porphyrae and the P.chondricola can be identified without sequencing and comparison, and a technical support is provided for prevention and control of red rot in porphyra cultivation in China.

The invention relates to an intelligent repair system of a connecting rod mold visual detection machine. The intelligent repair system comprises a supplementary welding robot, a welding power source, adjusting slide rails, a cooling system, a laser tracking system, a 3D intelligent scanning camera, a supplementary welding workbench and a control system. The supplementary welding workbench comprises a workbench body, workpiece locating frames, ejector pins and adjusting drive devices. At least one adjusting slide rail surrounds the supplementary welding workbench to form a closed annular structure. The laser tracking system and the 3D intelligent scanning camera are arranged on the supplementary welding robot. The cooling system is arranged on a tackle and connected with the supplementary welding robot and the supplementary welding workbench. The welding power source and the control system are arranged on the side surface of the supplementary welding workbench. On one hand, connecting rod surface quality can be precisely recognized and detected effectively, and the efficiency for finding surface defects is improved; and on the other hand, the surface defects are precisely analyzed and calculated, then welding equipment is driven for welding operation, and therefore the precision quality of surface supplementary welding operation is improved.

The invention relates to a snowflake noise detection method based on information entropy. The present invention is based on obtaining the overall distribution state of gray value of pixels, by transforming the front and rear image frames into gray space respectively, and then clustering the gray values ​​of all pixels into n intervals, in the form of gray histogram Show its distribution state, and calculate the respective information entropy, and finally judge whether there is snowflake noise in the video stream by comparing the change of the information entropy of the frame before and after. In the present invention, when the actual calculated value is not less than the established threshold and the value of the next frame is greater than the previous frame, or when the actual calculated value is less than the established threshold but the previous frame is marked as a noise frame, we finally determine that there is real snowflake noise , otherwise it is considered a normal image. The present invention has higher accuracy in detecting video snowflake noise, and also makes a good foundation for obtaining better visual effect after removing snowflake noise.

The invention discloses a laser detection apparatus with scenario identifying function, comprising a laser detection apparatus and a scenario identifying device mounted on the laser detection apparatus. The scenario identifying device comprises a pre-treatment module for images, an image extreme point detection module, an image characteristic point positioning module, a principal direction determining module and a characteristics extracting module and a scenario determining module. The image characteristic point positioning module determines the extreme points of characteristics points by sorting out the low contrast ratio points sensitive to noises and the unstable marginal points. The principal direction determining module connects any two adjacent peak values in the histograms of oriented gradients for the characteristic points to form a plurality of sub-line sections wherein adjacent sub-line sections with similar gradients are consolidated in the length direction to form line sections. The direction of the most optimal one of the line sections is regarded as the principal direction of the characteristic points. According to the invention, high identifying accuracy for scenarios and high speed can be achieved.