Product
Eureka
For R&D, Eureka makes reading and utilizing patents & technical documents easy.
Eureka AIR
Designed for self-driven R&D workflows. Generate viable solutions, solve complex R&D challenges, empower your innovation with AI.
Eureka Materials
Designed for material experts only. Revolutionize your material R&D, from search, analyze, to developing new materials.
Feature
TechResearch
Generate reliable direction feasibility study reports for your R&D in just a few steps.
TechSeek
Discover and master advanced knowledge NOW. Basics, ideas, possibilities, all at once.
TechMind
As an expert in R&D Theories, TechMind can generates customized viable solutions instantly.
TechRisk
Analyze your overall solution with one click, know your potential R&D risks in advance.
TechMonitor
Get weekly tech updates, stay abreast of the latest tech innovations and key insights.
Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
69results about How to "Accurate detection and identification" patented technology
Efficacy Topic
Property
Owner
Technical Advancement
Application Domain
Technology Topic
Technology Field Word
Patent Country/Region
Patent Type
Patent Status
Application Year
Inventor
Video-based detection and recognition system and method of vehicles
InactiveCN102637257AImprove detection accuracyHigh precisionRoad vehicles traffic controlCharacter and pattern recognitionVehicle detectionRecognition system
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.
Owner:北京尚易德科技有限公司
Orchard bactrocera dorsalis insect damage recognition system based on digital signal processing (DSP) and internet of things
ActiveCN103125457AImprove execution efficiencyRun fastInsect catchers and killersGeneral Packet Radio ServiceScanning tunneling microscope
The invention discloses an orchard bactrocera dorsalis insect damage recognition system based on digital signal processing (DSP) and the internet of things. The orchard bactrocera dorsalis insect damage recognition system comprises an installation support, a novel insect trap fixed on the installation support, an industrial camera arranged above the novel insect trap and a casing fixed on the installation support. A photoelectric detection module is arranged in the novel insect trap; a DSP digital image processing module, a scanning tunneling microscope (STM) 32 functional circuit, a global system for mobile communications (GSM) / general packet radio service (GPRS) wireless communication module and a charging control module which are sequentially connected are arranged in the casing; a displayer and a touch screen are arranged on the surface of the casing; and the charging control module is further connected with a solar panel. According to the orchard bactrocera dorsalis insect damage recognition system, the DSP technology, the machine vision technology and the internet of things technology are combined, the machine vision technology is achieved on a DSP processor platform and applied to detection and recognition of orchard bactrocera dorsalis, and accordingly fast and accurate detection and recognition of orchard bactrocera dorsalis are achieved. Simultaneously, automatic real-time monitoring of orchard insect damage is achieved and cost of manpower and material resources in orchard insect damage monitoring is reduced.
Owner:SOUTH CHINA AGRI UNIV
Method for synchronous detection of sweet potato feathery mottle virus (SPFMV), sweet potato virus C (SPVC), sweet potato virus G (SPVG) and sweet potato virus 2 (SPV2)
InactiveCN102230027AClear detection effectLow cost of reagentsMicrobiological testing/measurementSweet potato virus GDiseased plant
The invention relates to method for a synchronous detection of sweet potato feathery mottle virus (SPFMV), sweet potato virus C (SPVC), sweet potato virus G (SPVG) and sweet potato virus 2 (SPV2), and belongs to the field of plant protection. The method comprises the following steps of respectively designing and synthesizing specific forward primers and a universal reverse primer of SPFMV, SPVC, SPVG and SPV2, extracting total RNA of tissue of a diseased plant through a cetyltrimethyl ammonium bromide (CTAB) method, and carrying out one step quadruplex reverse transcription-polymerase chain reaction (RT-PCR) magnification process to realize a synchronous detection of SPFMV, SPVC, SPVG and SPV2. Primers designed by the method have strong singularities and SPFMV, SPVC, SPVG and SPV2 share areverse primer thus an interaction of primers is reduced. In the invention, a CTAB method is utilized for extracting RNA from tissue of an RNA virus infected plant, thus a quality of RNA is guaranteed and a detection cost is reduced effectively; and the inverse transcription and the multiple PCR are completed in one step thus a detection time is saved. The method has the advantages of rapid detection speed, high efficiency, strong singularity, high sensitivity and low cost and can realize a synchronous detection of four kinds of sweet potato virus thus has wide application prospects.
Owner:YUNNAN AGRICULTURAL UNIVERSITY
Power terminal vulnerability attack detection method based on message features
ActiveCN110719275AStrong targetingAccurate detection and identificationTransmissionNeural learning methodsNerve networkAttack
The invention discloses a power terminal loophole attack detection method based on message features, and belongs to the technical field of intelligent power grid terminal equipment safety. The methodcomprises the following steps: S01, acquiring communication message data between power terminal equipment and a master station, and classifying the communication message data of the power terminal equipment in a normal working state and an attacked state into a positive sample and a negative sample; s02, performing feature extraction on the positive sample and the negative sample, and forming a sample feature vector; s03, based on the sample feature vector, selecting a classifier to perform deep neural network training, and generating a vulnerability attack detection model; s04, collecting real-time communication message data between the power terminal equipment and the main station in work; s05, performing feature extraction on the real-time communication message data, and forming a detection feature vector; and S06, inputting the detection feature vector into a vulnerability attack detection model to detect whether the power terminal is attacked or not and the attack type. Accordingto the method, the power terminal equipment is subjected to safety monitoring from a network layer, and the power grid safety is improved.
Owner:ELECTRIC POWER RESEARCH INSTITUTE, CHINA SOUTHERN POWER GRID CO LTD +1
Traffic sign detection and recognition method based on improved YOLOv3-tiny
InactiveCN112464911AImprove accuracySmall footprint for storageImage enhancementImage analysisVideo memoryTraffic sign detection
The invention provides a traffic sign detection and recognition method based on improved YOLOv3tiny. The method comprises the following steps: collecting traffic sign image data, carrying out the enhancement and amplification of the image data through geometric transformation and color transformation, and carrying out the image labeling, thereby obtaining a traffic sign training set; constructingan improved YOLOv3tiny network model, and training the improved YOLOv3tiny network model by adopting the traffic sign training set; and constructing a traffic sign test set according to the traffic sign image data, and detecting and recognizing the traffic sign test set by using the trained improved YOLOv3tiny network model. The improved YOLOv3tiny provided by the invention has relatively strong generalization capability, occupies relatively small storage space and video memory space, improves the detection and recognition accuracy, can also ensure the real-time performance, and can realize accurate and quick traffic sign detection and recognition in vehicle-mounted embedded equipment with limited computing power.
Owner:QINGDAO UNIV OF SCI & TECH
Human body target identification method and apparatus
ActiveCN106127733ASimplify the feature extraction processGood uniqueness and space invarianceImage enhancementImage analysisHuman bodyDeep learning
The invention provides a human body target identification method and apparatus. The method comprises the steps of obtaining a depth image; extracting image pixel characteristics in the depth image; inputting the image pixel characteristics into a human body deep learning model to perform identification and classification; judging whether categories of the image pixel characteristics are matched with existing human body part tags in the human body deep learning model or not; and if the categories of the image pixel characteristics are matched with the existing tags in the human body deep learning model, outputting the tags corresponding to the pixel characteristics. According to the method and the apparatus, the image pixel characteristics are identified by adopting the deep learning model, and human body target detection and identification are finished, so that the detection and identification process is simplified and the detection and identification efficiency is improved.
Owner:HUNAN VISUALTOURING INFORMATION TECH CO LTD
CAPs (C1eaved Amplified Polymorphic Sequences) molecular marker for identifying amaranthus tuberculatus based on SNP (Single Nucleotide Polymorphism) site and application thereof
InactiveCN106498088ARapid detection and identificationAccurate detection and identificationMicrobiological testing/measurementInvasive species monitoringNucleotideAmaranthaceae
The invention provides a CAPs (C1eaved Amplified Polymorphic Sequences) molecular marker for identifying amaranthus tuberculatus based on an SNP (Single Nucleotide Polymorphism) site and application thereof. The molecular marker is located on ITS (Internal Transcribed Spacers) and 26S sequences of an amaranthaceous plant nuclear gene, and a nucleotide sequence of the molecular marker is as shown in SEQ ID NO.1. Primer pair sequences for amplifying the molecular marker are as shown in SEQ ID NO.2 and SEQ ID NO.3. The specific molecular marker provided by the invention can realize rapid and accurate detection of the amaranthus tuberculatus and a reliable detection method is provided for rapid distinguishing and identification of alien invasive species.
Owner:CHINESE RES ACAD OF ENVIRONMENTAL SCI
Environmental perception method based on machine vision and millimeter wave radar data fusion
ActiveCN111505624AAccurate detectionAccurate detection and identificationCharacter and pattern recognitionRadio wave reradiation/reflectionRadar systemsEnvironmental perception
The invention relates to the technical field of sensor data processing, in particular to an environmental perception method based on machine vision and millimeter wave radar data fusion. The method comprises a data acquisition step of acquiring the data of a radar system and a vision system; a data association matching step of obtaining the target object data lists detected by the two sensing systems, and performing matching association on the target object data detected by the two sensing systems; a target object tracking step of performing matching tracking on the target object, and updatingthe life cycle state of the target object; and a target object data fusion step of fusing the data signals of the target object outputted by the radar system and the visual system. According to the environment sensing method based on machine vision and millimeter wave radar data fusion, the advantages and disadvantages of two sensing systems of a vision system and a radar system can be fused to achieve the purpose of accurately sensing environment information, and the problem that a single-characteristic sensor is difficult to meet the precision and reliability requirements of a sensing system is solved.
Owner:CHINA AUTOMOTIVE ENG RES INST +1
Method for testing and identifying cepaea hortensis through polymerase chain reaction (PCR)
ActiveCN102952885AReliable resultsStrong specificityMicrobiological testing/measurementForward primerReverse primer
The invention discloses a method for testing and identifying cepaea hortensis through polymerase chain reaction (PCR), and the method comprises the following steps that a specific primer is applied to identify the molecular features of cepaea hortensis; a forward primer of the PCR is CN-(P1):5'-ACCTCCTTCCTTTCTACT-3', and a reverse primer is CN-(P2):5'-GTCAACATCTATCCCAAC-3'. The total volume of a PCR system is 25mu.L, including 12.5mu.L of 2*TaqPCRMasterMix, respectively 0.5mu.L of the forward and reverse primers, 2mu.L of a deoxyribonucleic acid (DNA) template, and the rest is sterilizing ddH2O; and PCR procedures are as follows: predegeneration is carried out for 5 minutes at 5DEG C, 50 seconds at 95DEG C, 30 seconds at 52DEG C and 50 seconds at 72DEG C, and the process is repeated for 35 times; and extension is carried out for 10 minutes at 72DEG C, and the reaction is over. According to the method for testing and identifying cepaea hortensis through PCR, the cepaea hortensis can be quickly and accurately tested and identified, a novel technical means is provided for the testing and the identification of the dangerous and harmful creature cepaea hortensis, and important significance in preventing the cepaea hortensis from being spread and diffused is realized.
Owner:INSPECTION & QUARANTINE TECH CENT OF FUJIAN ENTRY EXIT INSPECTION & QUARANTINE BUREAU
Illegal website identification system and method based on critical path
InactiveCN106776958ATimely identificationAccurate identificationWeb data indexingCharacter and pattern recognitionUniform resource locatorData store
The invention relates to an illegal website identification system and method based on a critical path. A system framework is divided into four layers, namely a user layer, an application service layer, a technical support layer and a data storage layer. The user layer provides a main account of the system; the application service layer provides a main function module of the system; the technical support layer comprises a relevant tool used in the system development process and a core algorithm program; the data storage layer provides data used in the system. According to the detailed functional division, the function module of the illegal website identification system based on the critical path comprises data preprocessing, website similarity calculation, website clustering, illegal website critical path extraction and illegal website identification. The system develops a similarity calculation program based on the Path through taking URL self characteristics as research start points, can accurately calculate the similarity among websites and obtain the effective URL critical path based on the website similarity and a Fast Unfolding clustering algorithm, and finally can discover illegal websites in unknown websites through the URL critical path.
Owner:THE THIRD RES INST OF MIN OF PUBLIC SECURITY
Method of detecting and identifying cernuella virgata da costa through fluorescent quantitation PCR
InactiveCN107058592ARapid detection and identificationAccurate detection and identificationMicrobiological testing/measurementDNA/RNA fragmentationFluorescent quenchingBiology
The invention provides a method of detecting and identifying cernuella virgata da costa through a fluorescent quantitation PCR. Molecular characteristics of cernuella virgata da costa can be identified by using a specific primer and a probe, an upstream primer of the fluorescent quantitation PCR is CVF:5'-CGTTTCGGGTGGGAGGGTTA-3', a downstream primer of the fluorescent quantitation PCR is CVR:5'-AAGTCGTCCGAGCTGTTCCG-3', the fluorescent probe is CVP:5'-FAM-AACCCAGGGCGCCTCCGCGG-BHQ-3', wherein a fluorescent reporting group which is marked by a 5' end is FAM, and a fluorescent quenching reporting group which is marked by a 3' end is BHQ. The method can detect and identify cernuella virgata da costa quickly and precisely, the method has the advantages of high specificity and sensitivity, short consuming time, and precise and convenient result determination, PCR post processing is not needed, and a false positive and cross pollution can be avoided effectively.
Owner:INSPECTION & QUARANTINE TECH CENT OF FUJIAN ENTRY EXIT INSPECTION & QUARANTINE BUREAU
Sparse representation-based underground hidden crack filler identification method
InactiveCN105590093AImprove accuracyReduce collectionCharacter and pattern recognitionDecompositionCorrelation analysis
The invention discloses a sparse representation-based underground hidden crack filler identification method. In combination with the feature extraction advantages of empirical mode decomposition (EMD) and wavelet decomposition, firstly, the EMD on the echo signals of different underground crack fillers is conducted to obtain an optimal empirical mode component. Secondly, the original echo signals are subjected to wavelet decomposition to obtain an optimal wavelet decomposition coefficient. Finally, the optimal empirical mode component and the optimal wavelet decomposition coefficient are subjected to correlation analysis, and then the optimal wavelet decomposition coefficient largest in correlation with the optimal empirical mode is adopted as a base of a novel over-complete dictionary. After the over-complete dictionary is obtained, a to-be-identified signal is processed based on the same method described as above. In this way, the optimal feature component of the signal can be obtained. At last, the over-complete dictionary and the to-be-identified optimal component are substituted into the compressed-sensing reconstructing algorithm, so that the signal is reconstructed and identified. The method has the advantages of smaller sampled data volume and higher precision.
Owner:GUILIN UNIV OF ELECTRONIC TECH
Method for detecting and identifying Cernuella virgata Da Costa by PCR
ActiveCN103952491AReliable resultsResult Reliability GuaranteeMicrobiological testing/measurementGeneticsBioinformatics
The invention discloses a method for detecting and identifying Cernuella virgata Da Costa by PCR. The method comprises identifying molecular characteristics of the Cernuella virgata Da Costa by specific primers, wherein the upstream primer of PCR reaction is HP- (P1): 5'-GTTATTGTTACTGCTCACGC -3', the downstream primer is HP- (P2): 5'-CTGCTAAGACAGGGAGAGAT-3'. The total volume of the PCR reaction system is 25 muL, wherein the volume of 2*TaqPCRMasterMix is 12.5 muL, the volume of the upstream and downstream primers are respectively 0.5 muL, the volume of DNA template is 3 muL and the rest volume is supplied with sterile ddH2O; PCR reaction program comprises carrying out a denaturation at 94 DEG C for 5 minutes at 94 DEG C for 30 seconds, at 46 DEG C for 30 seconds, at 72 DEG C for 1 minute, circulating for 25 times, extending at 72 DEG C for 10 minutes and completing the reaction. The Cernuella virgata Da Costa can quickly and accurately detected and identified by the method disclosed by the invention. The method provides a new technical means for the detection and identification of the Cernuella virgata Da Costa in the 'List of Quarantine Pests of the People's Republic of China of the Entry of Plants' and is of great significance for preventing the Cernuella virgata Da Costa from spreading and diffusing.
Owner:INSPECTION & QUARANTINE TECH CENT OF FUJIAN ENTRY EXIT INSPECTION & QUARANTINE BUREAU
Traffic sign identification method and device
InactiveCN109902610AAccurate detection and identificationRobust detection and recognitionCharacter and pattern recognitionPattern recognitionTraffic sign recognition
The invention discloses a traffic sign recognition method and device. The method comprises the following steps: acquiring image data from a camera; Detecting a region of interest of a current frame ofthe image data by using a deep learning technology to obtain an initial traffic sign detection result; Tracking the traffic sign detection result to obtain a final traffic sign detection result whichcan be a tracking result; And predicting the region of interest of the next frame according to the tracking result.
Owner:HANGZHOU FABU TECH CO LTD
Method for detecting and identifying theba pisana through fluorescent quantitative PCR (polymerase chain reaction)
ActiveCN106755524AAccurate and convenient result judgmentStrong specificityMicrobiological testing/measurementFluorescenceQuenching
The invention provides a method for detecting and identifying theba pisana through fluorescent quantitative PCR (polymerase chain reaction). Specific primers and a probe are used for identifying the molecular features of the theba pisana; an upstream primer of the fluorescent quantitative PCR is TPF: 5'-GCTGGAAAGTGGAGGCCGTA-3'; a downstream primer is TPR: 5'-GCGGCGTCGGTTAAGAGAGA-3'; a fluorescence probe is TPP: 5'-FAM-AGCTGTTCCGGGCACCAAGACGTCACG-BHQ-3'; 5' end marked fluorescence reporter gene is FAM; 3' end marked fluorescence quenching gene is BHQ. The method has the advantages that the theba pisana can be fast and accurately detected and identified; the specificity is high; the sensitivity is high; the consumed time is short; the result judgment is accurate and convenient; the PCR post treatment is not needed; the false positive and the cross contamination can be effectively avoided.
Owner:INSPECTION & QUARANTINE TECH CENT OF FUJIAN ENTRY EXIT INSPECTION & QUARANTINE BUREAU
Method for detecting and identifying Helix aspersa MUller through PCR (polymerase chain reaction)
ActiveCN103993092AReliable resultsResult Reliability GuaranteeMicrobiological testing/measurementMolecular biologyPlant quarantine
The invention discloses a method for detecting and identifying Helix aspersa MUller through PCR (polymerase chain reaction). The method comprises the step of identifying the molecular feature of the Helix aspersa MUller by using specific primers, wherein the upstream primer of PCR is HP-(P1): 5'-ACTCTTTGGTATTTGATGCG-3', and the downstream primer is HP-(P2): 5'-ACCTGCTAAAACGGGTAATG-3'. The method can be used for quickly and accurately detecting and identifying the Helix aspersa MUller. The method provides a new technical means for detection and identification of the Helix aspersa MUller in Imported Plant Quarantine Pest Directory of the People's Republic of China, and has significance for preventing the propagation and diffusion of the Helix aspersa MUller.
Owner:INSPECTION & QUARANTINE TECH CENT OF FUJIAN ENTRY EXIT INSPECTION & QUARANTINE BUREAU
Embedded part detection method, device, equipment and system of component production line
The embodiment of the invention discloses an embedded part detection method, device, equipment and system of a component production line, wherein the method comprises the steps of: obtaining the position information of an embedded part to be placed, an initial mould platform image that an embedded part is not placed, and a mould platform image that the embedded part is placed; according to the position information, extracting the initial projection contour line in the initial mould platform image, the laser contour line in the mould platform image and the image characteristics of the embeddedpart; according to a comparison result of the initial projection contour line, the laser contour line and the standard laser contour line, generating an embedded part position detection result, wherein the standard laser contour line is a laser contour line when the placement position of the embedded part to be placed on the mould platform accords with the standard requirement; and, according to acomparison result of the image characteristics of the embedded part and the image characteristics of the embedded part to be placed, generating an embedded part type detection result; thereby, rapidand accurate detection on the placement position and the type of the embedded part can be realized; the automation degree and the production efficiency of a prefabricated part are improved; and the manual labour workload is reduced.
Owner:CHINA MINGSHENG DRAWIN TECH INVESTMENT CO LTD
Semantic segmentation-based text detection method for scene of any shape
PendingCN111553351AImprove detection accuracyAccurate detectionNeural architecturesCharacter recognitionStochastic gradient descentText detection
The invention discloses a semantic segmentation-based text detection method for a scene of any shape, wherein the method comprises the following steps of S1, constructing a semantic segmentation-basedtext detection network model for a scene of any shape; S2, according to the overall target loss function, minimizing the overall loss function by using a back propagation algorithm and stochastic gradient descent optimization, and carrying out iterative training on the model designed in S1; S3, using a step-by-step scale expansion method, performing scene text detection and recognition accordingto the model trained in the S2. For the text detection problem of text instances which are in any shape and are close to each other in a natural scene, the invention provides a semantic segmentation-based method for creatively utilizing a multi-kernel step-by-step scale expansion method to detect the text, so that the positions of the text blocks are detected more accurately.
Owner:FOSHAN NANHAI GUANGDONG TECH UNIV CNC EQUIP COOP INNOVATION INST +1
Target tracking method and device
InactiveCN107103268AAccurate detection and identificationEfficient detection and identificationCharacter and pattern recognitionHistogramGradient direction
The embodiment of the invention discloses a target tracking method. The method comprises the steps of obtaining a histogram characteristic vector of a current frame in a first gradient direction; according to the histogram characteristic vector in the first gradient direction, determining the position of a first target window of the current frame, wherein the first target window is a target position determining window in the current frame; acquiring the position of a second target window in a previous frame; according to the position of the first target window and the position of the second target window, tracing the motion track of a target. The embodiment of the invention also discloses a target tracking device.
Owner:CHINA MOBILE GROUP ZHEJIANG
Obstacle identification method and system based on 3D laser point cloud
PendingCN113640826AImprove processing efficiencyNarrow searchElectromagnetic wave reradiationICT adaptationPoint cloudCloud data
The invention provides an obstacle identification method and system based on 3D laser point cloud. The method comprises the following steps: acquiring three-dimensional environment laser point cloud data; performing ground segmentation processing on the three-dimensional environment laser point cloud data to obtain non-ground laser point cloud data; partitioning the non-ground laser point cloud data, and marking obstacles in each partition by using two-dimensional Euclidean clustering; and performing packaging processing on the clustering result by using a cuboid model to obtain the position and the size of an obstacle. According to the invention, the real-time performance of ground segmentation and point cloud clustering is effectively improved, and the identification precision of obstacles at a relatively long distance is improved.
Owner:SHANDONG UNIV +1
On-Line Signal Event Detection and Identification Method and Apparatus
InactiveUS20180048495A1Accurate detectionAccurate identificationTransmitter/receiver shaping networksPattern recognitionDistortion
A method of and apparatus for on-line detection and identification of synchronous and asynchronous piecewise-constant signals corrupted by noise and distortion. The method achieves joint temporal and amplitude detection and identification by following an initial amplitude approximation with a change detection that continually converges to an accurate joint identification while continually correcting intermediate approximation errors via analysis of temporal signatures. In turn, the temporal identifications employ accumulated statistics partly determined by the initial amplitude approximation.
Owner:MAGIEROWSKI SEBASTIAN +1
Rear-end resistance early-warning method for vehicle based on camera and millimeter-wave radar
InactiveCN109435847AAccurate detection and identificationEnsure safetyOptical viewingRadarPerpendicular direction
The invention discloses a rear-end resistance early-warning method for vehicle based on a camera and a millimeter-wave radar. The method comprises the following steps that 1, the camera is calibrated;S2, the vehicle is recognized; S3, the relative distances di between surrounding objects and the vehicle, the relative speeds vi of the surrounding objects and the included angle alpha I in the perpendicular direction of the emitting surface of the radar are detected by the millimeter-wave radar, wherein i is the positive integer; S4, sampling points of null objects and invalid objects are removed; S5, sampling points corresponding to vehicles running from a lane in the opposite direction are removed; S6, sampling points corresponding to effective targets are selected; S7, the effective target with the smallest longitudinal distance si is determined as a potential rear-end target; S8, the equation that th=si / V is calculated, and according to the result, different warning signals are transmitted. According to the method, the vehicle sampling points of the null targets and the invalid targets can be removed, the sampling points corresponding to vehicles running from the lane in the opposite direction are removed, front vehicles can be detected and recognized more accurately, a driver can be timely warned, and safety in the driving process can be ensured more effectively.
Owner:常州智行科技有限公司
Marine engineer abnormal behavior recognition method based on human body joint point detection and D-GRU network
PendingCN113378638AAccurate judgmentImprove accuracyCharacter and pattern recognitionNeural architecturesEnvironmental geologyHuman body
The invention provides a marine engineer abnormal behavior recognition method based on human body joint point detection and a D-GRU network. The method comprises the steps of designing a human body detection model based on human body joint point detection and a GRU network structure; due to the fact that a cabin is dark in light and complex in working environment, detecting and tracking human body joint points by the model through a joint point detection model, and detecting the abnormal behavior of the marine engineer by the human body joint point detection model; and outputting coordinates of the obtained heat maps of the key points, and inputting the coordinates into a D-GRU network structure to carry out human body posture prediction so as to realize recognition of abnormal behaviors of marine engineers. According to the technical scheme, the problem that in the prior art, when working personnel such as marine engineers work in far places such as ship buildings, engine rooms and boiler rooms, the personal safety of the marine engineers cannot be known and reflected in time is solved.
Owner:DALIAN MARITIME UNIVERSITY
Binder clip defect detection method and system based on deep learning
PendingCN113344847AAccurate detection and identificationRealize detection and identificationImage enhancementImage analysisPattern recognitionImaging processing
The invention discloses a binder clip defect detection method and system based on deep learning, and belongs to the technical field of image processing, and the method comprises the following steps: S1, constructing a sample data set; S2, obtaining a pre-training model; S3, constructing a binder clip defect detection and identification model; and S4, carrying out detection and identification. The YOLOv4-Tiny network serves as a basic network, weight parameters in the pre-training network are shared through transfer learning, a global space attention mechanism module is added, the feature representation capacity is improved, and detection and recognition of binder clip defect images can be accurately achieved; and in an image defect detection effect comparison experiment, the method has the advantages that a good detection effect is shown, the mAP value index reaches 91.66%, detection and recognition of the binder clip defect image are achieved, a foundation is laid for follow-up implementation of binder clip detection and a feedback system, and the method is worthy of being popularized and used.
Owner:ANHUI UNIVERSITY OF TECHNOLOGY
Vehicle hiding person detection method and apparatus thereof
InactiveCN107015268AAccurate detection and discrimination resultsEasy to operateAcoustic presence detectionPerson detectionEngineering
The invention discloses a vehicle hiding person detection method and an apparatus thereof. The detection method comprises the following steps of collecting a vehicle vibration signal of a detected vehicle respectively and a ground environment vibration signal of an environment where the detected vehicle is located; calculating an energy value of the vehicle vibration signal, and a correlation value between the vehicle vibration signal and the ground environment vibration signal respectively; and according to the calculated energy value and the correlation value, determining whether there is a hiding person in the detected vehicle. The apparatus comprises a slight vibration detection unit, a data processing unit and a determination unit. In the invention, the vehicle vibration energy value and the correlation value can be fused, and the vehicle hiding person can be accurately detected and distinguished. The method and the apparatus possess advantages that a realization method is simple; detection efficiency and precision are high; environment adaptability is high; and a detection process is safe and reliable and so on.
Owner:HUNAN NOVASKY ELECTRONICS TECH
Orchard bactrocera dorsalis insect damage recognition system based on digital signal processing (DSP) and internet of things
ActiveCN103125457BImprove execution efficiencyRun fastInsect catchers and killersGeneral Packet Radio ServiceVisual technology
Owner:SOUTH CHINA AGRI UNIV
Image target detection method and device, storage medium and sewage pipeline detection device
PendingCN111191593AUniform precisionQuality improvementCharacter and pattern recognitionNeural architecturesEngineeringComputer vision
The invention discloses an image target detection method, and belongs to the technical field of computers. The method comprises the steps of obtaining a to-be-processed target image; inputting the target image into a convolutional neural network model, and extracting at most five feature maps; outputting a prediction box corresponding to each of the at most five feature maps based on the convolutional neural network model; and performing target identification on all the prediction frames through a detection module to obtain the identification result of the target image, wherein the detection module comprises a residual block and a detection convolution layer which are connected in sequence. The invention further discloses an image target detection device, a storage medium and a sewage pipeline detection device. According to the method, accurate image recognition and detection can be performed under the condition of strictly limiting the system memory and the calculation cost.
Owner:成都云尚物联环境科技有限公司
Face detection model training method and device, electronic equipment and storage medium
PendingCN114565961ASolve the problem of not being able to accurately identify facial imagesAccurate detection and identificationCharacter and pattern recognitionFace detectionImaging quality
The invention discloses a face detection model training method and device, electronic equipment and a storage medium. A training sample set is obtained; wherein the training sample set comprises at least three types of training samples: a face correct labeling type, a face abnormal labeling type and a face non-labeling type; performing training processing on the first to-be-trained face detection model and the second to-be-trained face detection model based on the training set sample set to obtain a first initial face detection model and a second initial face detection model; wherein the first to-be-trained face detection model and the second to-be-trained face detection model have the same model structure, and index parameters of all neural network layers in the model structures are different; and carrying out joint training on the first initial face detection model and the second initial face detection model based on the retraining sample set to obtain a target face detection model, and carrying out detection processing based on a face image in the target face detection model image. The method achieves the accurate recognition of the image under the condition that the image quality is poor.
Owner:AGRICULTURAL BANK OF CHINA
Gear defect visual detection method and system based on improved YOLOv5 network
ActiveCN114677597AAccurate detection and identificationEasy to detectCharacter and pattern recognitionNeural architecturesPattern recognitionData set
The invention discloses a gear defect visual detection method and system based on an improved YOLOv5 network. The system comprises a controller with a built-in improved YOLOv5 network model, a conveying system composed of a first conveying belt and a second conveying belt, and an image acquisition module and a rejection mechanism which are built on the conveying system. According to the improvement mode, an unimproved YOLOv5 network is trained through a sample data set to obtain weight parameters, a convolution attention mechanism module and a repeated weighted bidirectional feature pyramid network are added to a YOLOv5 network model, and the weight parameters are migrated to the improved YOLOv5 network model; and training an improved YOLOv5 network model through the data set, completing the construction of a gear defect detection model, collecting an image through an image collection module, inputting the image into the gear defect detection model for recognition, and rejecting a corresponding defective gear according to a recognition result. According to the invention, accurate identification of gear characteristic defects and automatic detection and sorting of gear multi-surface defects are realized, and the detection efficiency is improved.
Owner:WUHAN UNIV OF TECH +1
Detection of encrypted packet streams using feedback probing
ActiveUS20100232313A1Accurate detectionAccurate identificationError preventionFrequency-division multiplex detailsData typeInternet Protocol
Methods, systems, and devices are disclosed for detecting encrypted Internet Protocol packet streams. A probe stream having a known observable parameter is generated. The observable parameter exhibits a known characteristic of a known type of data encrypted within a stream of packets. The probe stream is communicated to a network element via a communications network. When the probe stream is received by the network element, the network element can compare the known observable parameter to an actual value. Any difference between the known observable parameter and the actual value can be used to correct for network-induced variation and other effects, thus ensuring accurate detection and identification of data types within encrypted streams of packets.
Owner:AT&T INTPROP I L P
Who we serve
- R&D Engineer
- R&D Manager
- IP Professional
Why Eureka
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Social media
Eureka Blog
Learn More Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com