6014results about How to "The detection process is fast" patented technology

The invention belongs to the technical field of the pattern identification and image processing, and relates to a license plate detection and identification method based on the maximum stable extremal region, which comprises the following steps: extracting the maximum stable extremal region (MSER) to obtain text regions of candidate license plates; adopting one effective feature for description of each extremal region, classifying the extremal regions into 'text' regions and 'non-text' regions with a classifier obtained in prior training, and extracting a license plate from original images based on the characteristics of the structure of the license plate; and describing with the characteristics of the shape context, and completing the character identification through template matching. Since the maximum stable extremal region has the affine invariance, high stability and multi-scale features, and the regions are determined only according to the gray value and are not light-sensitive,the license plate detection and identification method with the maximum stable extremal region as the base is applicable to the complex background, and has the advantages of good stability and high identification rate.

The invention provides a lane line detection method based on a full-convolution network and relates to the field of traffic information detection. According to the method, probability operation is performed on an output feature map of the full-convolution lane line detection network, the probability that a lane line appears in each region in an input picture is obtained, a prediction probability threshold is set, and extraction and detection of lane lines are realized. Through the method, detection of straight lane lines and curved lane lines can be realized simultaneously; a lane line detection loss function is used to train the full-convolution lane line detection network, so that the detection effect of the lane lines is improved; the convolutional neural network learns abstract features of the lane lines from a lane line classification dataset instead of simply extracting external features of the lane lines; detection of a newly input image can be realized just by storing a lane line detection network model, so that storage space is saved, and the method is suitable for vehicle-mounted embedded equipment; and the small and shallow full-convolution lane line detection network isadopted to perform detection acceleration, so that detection speed is high.

The invention discloses a method and system for quick lossless evaluation on multiple indexes of freshness of fresh beef by applying a visible / near infrared spectrum technology, belonging to the technical field of lossless detection of food. According to the invention, a mathematical prediction model is established between spectral information capable of reflecting fresh beef ingredients and physical state information, and multiple indexes of the freshness of the fresh beef, and the whole detection system is constructed on the basis of the mathematical prediction model, so that the system can be used for accurately and reliably detecting multiple unknown indexes of the freshness of the fresh beef (volatile basic nitrogen (TVB-N), pH value, total bacterial count and beef color (CIE color parameters L* and a*) and the like simultaneously, and predicating the storage time of the beef; furthermore, the freshness of the beef is graded comprehensively by combining multiple indexes, so that the quick lossless detection and evaluation on the freshness of the fresh beef are realized.

A real-time visual detection and identification method for high speed rail surface defects comprises the following steps of: (1) image acquisition; (2) image preprocessing; (3) defect preliminary detection, that is, performing logic or operation combination of detection results based on gray scale compensation with detection results based on top-hat operation, and detecting whether an abnormal area exists in an image, if not, finishing the detection, or else continuing the processing; (4) defect accurate positioning, that is, accurately positioning the defect by an algorithm of bonding single defect, an algorithm of filling holes in the defect area, and an algorithm of selecting the main defect, and extracting the defect area through marks; (5) defect classification, that is, extracting and selecting characteristics of the defect area, designing and training a BP neural network, and classifying the defects by the BP neural network. The invention has the advantages of simple principle, high automation degree, high detection speed, and high detection precision.

The invention relates to a nondestructive detection method for tea quality based on a near infrared spectrum and a machine vision technology and a device thereof. The detection method comprises the following steps: spectrum and image acquisition to tea are synchronously performed, obtained initial data are subjected to preprocessing, feature extraction and fusion, and then an expert evaluation result is combined to establish a quality grade evaluation model of the tea to quickly and nondestructively detect the comprehensive internal and external quality of the tea. The detection device comprise a spectrum acquisition device, an image acquisition device and a computer, wherein the spectrum acquisition device and the image acquisition device finish the signal acquisition of a tea sample and then transfer a signal to the computer; and the computer finishes the extraction and fusion of feature information, and then substitutes in the pre-established model to judge a final result of the quality of the tea. The method applies a fused nondestructive detection means based on the two technologies to the detection process of the comprehensive quality of the tea, has the advantages of high detection accuracy, good result consistency, strong automation degree and so on, and creates conditions for standardized classification of the quality of the tea.

The embodiment of the invention discloses a training and detecting method of a malicious code family. The method includes the steps that malicious codes are mapped into images, and image features of the images are extracted; the malicious code images are clustered according to the image features, and malicious code family labeling is conducted on different types of malicious code images obtained after clustering; a convolutional neural network model is established; the convolutional neural network model is trained with malicious code image sets obtained after malicious code family labeling, and the trained convolutional neural network model is used as a detection model; family detection is conducted on a malicious code sample to be detected and the variation of the malicious code sample with the detection model. The embodiment of the invention further discloses a training and detecting device of a malicious code family.

The present invention provides a neural network-based face detection model training method, a neural network-based face detection method, a neural network-based face detection model training system and a neural network-based face detection system. The training method includes the following steps that: the loss function of the bias network layer of a prediction face frame is calculated according to the bias information of the prediction face frame relative to a default face frame and the bias information of a real face frame relative to the default face frame; the loss function of the confidence network layer of the prediction face frame is calculated according to the confidence of the default face frame; the error of the two loss functions is calculated, and is fed back to a neural network, so that the weight of the neural network is adjusted; iterative training is repeated until convergence appears, so that a face detection model can be obtained, and therefore, the prediction face frame can contain a face more accurately. The detection method includes the following steps that: a face image to be detected is inputted to a trained face detection model, bias information and confidence are outputted; corresponding prediction face frames are calculated according to the bias information; and a prediction face frame corresponding to confidence greater than a preset confidence threshold or the highest confidence is selected as a face detection result.

The invention provides a testing machine used for testing the appearance of a printing circuit board, comprising a frame which is provided with a transmission mechanism; the front end of the transmission mechanism is provided with an upper plate mechanism; the transmission mechanism is internally provided with two scanning mechanisms; a panel turnover mechanism is arranged between the two scanning mechanisms; the tail part of the transmission mechanism is butt-jointed with a sorting mechanism; and all mechanisms are controlled by control host machines cooperatively. The testing machine adopts a whole-line digital camera photographing technique to detect the defects on the printing circuit such as foreign matter, exposed copper, supplemented oil, scraping, poor gold-plating, character error, non-uniform green oil, non-uniform welding disc, non-uniform depth of channel, residual copper, omitted printing, developing and the like, automatically feeds and sorts the nondefective and waster, improves the simpleness and reliability of the operation, can observe the defects of the board in real-time by adopting a powerful image processing and analysis software system to provide a plurality of image processing and analysis systems, has quick detection speed, high detection efficiency, safe and reliable equipment, simple operation, is convenient and easy to be learned and meets the man-machine interaction.

The invention discloses a pedestrian detection method based on a deep learning technology. The method comprises the steps that firstly, a binary classification model is trained through a step-by-step migration strategy on the basis of transfer learning to initialize final model parameters; secondly, pedestrian detection work is completed by adopting and modifying a currently popular and efficient Faster RCNN frame, and on the basis of the CNN characteristics of the frame, not only can images with any scales be processed, but also the detection speed is high. Compared with the prior art, the method has the advantages that the network does not need to be specially designed, existing available data is fully utilized, a good experiment result still can be achieved by adopting a general network structure, the advantages of a deep convolution network are fully achieved, and the advantages of being simple in design, good in robustness, high in detection accuracy and low in omission ratio are achieved.

The invention discloses a method for detecting vehicle door outer plate shape and size based on single-projection encoding structured light and aims to break through the condition that structured light visual detecting technology is rarely applied to online detection on the vehicle door outer plate shape and size. The method comprises the following steps of 1, calibrating a structured light detecting system, i.e. calibrating a camera (2) and a projector (4), and establishing corresponding non-linear relation between a space position and an image coordinate; 2, designing an encoding projection template; 3, decoding and identifying a detecting image, i.e. projecting the encoding projection template to the surface of a vehicle door outer plate by using the projector (4), shooting a surface image of the vehicle door outer plate by using the camera (2), and transmitting the shot image to a computer to be processed and identified; and 4, calculating a space point cloud coordinate and evaluating an error, i.e. 1) calculating a three-dimensional space coordinate of surface projection image characteristic points of the vehicle door outer plate, and 2) matching and rectifying a detecting point cloud obtained through calculation and a point cloud model measured by a three-coordinate measuring machine.

The invention discloses a multi-vision defect detecting equipment and a multi-vision defect detecting method for a large-size LCD glass substrate in a production line. For the equipment, a linear scanning imaging detection system is adopted for acquiring the high-definition gray image of an LCD glass substrate to be detected, and the equipment is simple in structure and convenient to operate; according to the method disclosed by the invention, the acquired image is processed, a kmeans clustering method is adopted to carry out defect existence judgment on the processed image of the LCD glass substrate, and by making the defect regions, the defect category is judged by a classification method of a support vector machine (SVM), and the defect quantity is counted. Processed image and detection results are transmitted to an office in real time by a filed bus control system for operators on duty to check, and the field production parameters are transmitted, so that remote monitoring of the system is realized.

The present invention discloses a deep learning-based SAR remote sensing image water surface target detection method, which mainly solves the problems of the slow detection speed and low positioning accuracy in the existing SAR image water surface detection method. The specific implementation is: designing and pre-training an image classification model; then designing and training a target detection model based on the model, wherein the target detection model comprises an RPN region suggestion network and a Fast R-CNN target detector; and finally using the trained model to detect the SAR imagewater level. The method disclosed by the present invention has the advantages of a high detection speed and high detection accuracy, and can be used for water surface target detection of large-area SAR images.

The invention discloses a TFT-LCD Mura defect machine vision detecting method based on B spline surface fitting and belongs to the field of LCD display defect detecting. The method includes: using a CCD camera to collect the grayscale image of a lighted to-be-detected LCD; filtering the original image; extracting an interested area; using double three-time B spline surface fitting to fit an image background; using the original image to subtract the background image so as to obtain the image with the background, with uneven brightness, being removed; using a Canny operator to detects Mura defects; determining defect level. The method has the advantages that the method is reliable, high in accuracy and time saving in calculation.

The invention provides a thermal imaging temperature monitoring device and a thermal imaging temperature monitoring system applying the same. The system comprises at least one thermal imaging temperature monitoring device, at least one signal analysis equipment and a data management platform, wherein the device is used for acquiring infrared radiation and visual light video images in a visible monitoring region, and comprises an infrared thermal imager, a visible camera, a holder and a temperature sensor; the infrared thermal imager and the visible camera are arranged on the holder; the temperature sensor is used for correcting drifting of the infrared thermal imager caused by environment temperature changes; the signal analysis equipment is used for analyzing a video signal acquired by the thermal imaging temperature monitoring device to acquire position and temperature information of each target in the monitoring region; and the data management platform is used for providing information of a monitored object for monitor staff and accepting a command of the monitor staff. The thermal imaging temperature monitoring device and the device can monitor the temperature of a human body in a large scale range of a public place, realize non-contact, rapidness and no restriction of flowing of monitored staff.

The invention proposes a dangerous driving behavior real-time detection method based on deep learning, and the method comprises the steps: building a front car image collection system, collecting a training sample, carrying out the manual marking, and building a dangerous driving behavior data set; proposing a spatial pyramid pooling convolution depth confidence normalized classification network (SPP-CDBRNet) model according to the characteristics of the dangerous driving behavior data set based on a deep learning method; carrying out the preprocessing of the built data set, carrying out the training of the SPP-CDBRNet model through a random gradient descent method with a momentum and the data set after preprocessing, and obtaining an SPP-CDBRNet which precisely recognize whether there is a dangerous driving behavior (behaviors of using a cellphone and smoking during driving); carrying out the detection of a front car image through the determined SPP-CDBRNet model, and achieving the real-time detection of the dangerous driving behavior. The method can effectively improve the detection precision of dangerous driving behaviors, is good in instantaneity and mobility, and is good in application prospect.

The invention discloses visual detection equipment of a circuit board. The visual detection equipment of the circuit board comprises a bearing plate and a control system, a conveyor trough arranged on the bearing plate, a conveyor belt clamped in the conveyor trough and used for conveying a supplied to-be-detected circuit board, a first visual detection device, a second visual detection device, a finished product collection device and a defective product collection device, wherein the first visual detection device comprises a first CCD (charge coupled device) imaging system and a first blocking mechanism; the second visual detection device comprises a second CCD imaging system and a second blocking mechanism; the finished product collection device is used for collecting finished circuit boards after detection; and the defective product collection device comprises a defective product rejecting manipulator and a defective product collection conveyor belt used for collecting defective products. Compared with the prior art, only one conveyor belt is adopted for the detection process, so that the cost is saved greatly, and the visual detection equipment of the circuit board can perform detection on a first surface and a second surface of the circuit board respectively, the detection covering range is large, and the detection speed is high.

The invention relates to a multi-channel portable colloidal gold test paper strip quick test device, which mainly comprises a detector, wherein the detector comprises a detecting part, an electrical part and a software detecting part; the detecting part comprises a plurality of mutually independent detecting channels; the electrical part comprises an ARM data processor, a keyboard and an LCD; image data detected by the detecting part are processed and displayed by the electrical part; and the software detecting part controls detecting steps. The device has a multi-channel detecting structure, and four channels can carry out the same detection item as well as different detection items. The device has the advantages of high detection speed, large amount of samples and great practical value.

The invention relates to an on-line detection grading device for a lithium ion battery pole piece and a method thereof. In the device, image data of the lithium ion battery pole piece are acquired by using an embedded visual recognition device and are analyzed and processed to judge whether the appearance quality of the pole piece meets requirements or not, so that the actions of a turnover part and a grading part are controlled to realize real-time on-line double-side detection and grading of the electrode pole piece. The method has high detection speed; generally, the detection of the frontside and the back side of one electrode pole piece can be finished within 1 second, so that the production efficiency is increased; if the pole piece is manually detected, longer time is required to be taken; the detection accuracy is high; the embedded visual detection technology is adopted, so that the accuracy is high and the stability is also realized, and a detection result has high uniformity; the running time is long, namely a detection system can uninterruptedly run for 24 hours, so that the productivity is improved and the labor cost is saved; and on-line detection can be realized, so that the on-line detection of the electrode pole piece is realized, the interference in a manual operation process is reduced, and the productivity is improved.

The invention discloses machine visual on-line detection method and equipment for the tank cover quality in a high-speed production line. The method comprises the following steps of: in an automatic production line, separating tank covers to be detected into single tank covers moving on a delivering device; imaging the tank covers by using an imaging system at a high speed when the tank covers move to a detection station of the delivering device; transmitting an acquired tank cover image to an industrial computer for processing; carrying out quality detection on a plurality of detection areas to generate quality detection signals; eliminating unqualified products by using an eliminating device positioned at the eliminating station of the delivering device; making the qualified products continuously move on the delivering device and pass through a shaping device and outputting the qualified products in sequence. The equipment comprises a cover feeding device, the delivering device, the shaping device, the eliminating device, the imaging system, the industrial computer and an electric control system. The invention has the advantages of simple and compact structure, low cost, simple and convenient operation, high precision, high detection speed and the like.

The invention discloses equipment for detecting circuit board defects based on a visual detection method. The equipment comprises a material loading module, a detection function module, a material discharging module and a control module which is used for controlling a circuit board positioning mechanism to move and performing defect detection and discharge control, wherein the material loading module is connected with an inlet of a feeding conveying belt; the outlet of the feeding conveying belt is connected with the detection function module; the circuit board positioning mechanism is arranged at the outlet of the feeding conveying belt; the detection function module is connected with the material discharging module, and comprises a detection functional module base, a cameral pedestal provided with a charge-coupled device (CCD) camera, an electromagnet baffle plate and a detection conveying belt used for conveying the circuit board to be detected; the detection conveying belt is arranged on the detection function module base; the camera pedestal is positioned above the detection conveying belt; and the electromagnet baffle plate is arranged at the outlet of the detection conveying belt. By the equipment, the accuracy is improved, the detection speed is high, and the working efficiency is high.

The invention provides an SAR image target detection method based on the convolutional neural network and mainly solves the problem of low detection speed and low detection accuracy rate in the conventional SAR image target detection technology. The method comprises the steps of firstly training a convolutional neural network classification model M0 and designing two to-be-trained convolutional neural network models M1 and M2 based on the model M0; secondly, using an expanded MiniSAR training set as the input for the to-be-trained convolutional neural network models M1 and M2 and acquiring an optimal slice extracting model M12 and an optimal detection model M22 through training; thirdly, detecting the MiniSAR training set by using the optimal slice extracting model M12 and the optimal detection model M22; fourthly, reserving candidate areas with output probability values of the optimal detection model M22 being greater than a threshold value and acquiring a detection result. The method has the advantages of high detection speed and high detection accuracy rate and can be used for vehicle target detection.

The invention relates to a microfluidic control chip which is formed by bonding and sealing PDMS (Polydimethylsiloxane) material integrated with a micro valve, and comprises an upper layer of air control passage, a middle layer of PDMS membrane and a lower layer of microfluidic passage. A real-time PCR virus quick detection device comprises the microfluidic control chip, a temperature control unit, a signal detection unit and a data processing and control unit. The invention designs the microfluidic control chip which is integrated with the micro valve, is in a three-layer PDMS structure, andis used for virus detection and the real-time fluorescent PCR detection device. Compared with the traditional method, the invention has the advantages of short time, less sample usage, quick detection speed, simpleness and convenience in operation, integration and the like.

The invention relates to a direct power-spectral method-based all-weather long drainage basin water quality monitoring and early-warning system, belonging to the technical field of water quality monitoring. The system comprises a plurality of water quality monitoring devices, a cloud data processing platform and a remote monitoring platform; the water quality monitoring devices are placed at a shoreside of a water body, a water surface buoy or other water surface platforms according to needs, used for extracting water quality data of the long drainage basin, use a direct power-spectral method to acquire an absorption spectrum of a monitored water body and sends the monitored data and water quality information to the cloud data processing platform in a wireless transmission manner; the cloud data processing platform summarizes and analyzes the received data and stores into a database; the remote monitoring platform inquires and displays the data stored in the database and realizes the real-time monitoring and early-warning for an all-weather long drainage basin water quality environment. The system can realize the online measuring, real-time monitoring, emergent early-warning, drainage basin analyzing and water environment monitoring of all weathers of the water body in the drainage basin, thus reducing the loss caused by pollution to the water body and ensuring drinking water safety.

The invention discloses a method for measuring and locating leakage of a gas pipeline network based on a geographic information system (GIS) and supervisory control and data acquisition technology (SCADA). The specific method comprises the following steps of establishing a gas pipeline network GIS system; reading and storing a pipe network graph, pipeline network attribute data and the like; collecting parameters of pressure, temperature and flow for each section of gas pipeline in real time by the SCADA system; comparing the collected parameters and the stored data and calculating; calculating the measuring values of the head end and the tail end of the pipeline and calculating the difference among parameter values; representing the abnormal pipeline information; searching extreme points and determining the pressure catastrophe points; calculating the position of the leakage point according to the parameters of gas consumption, temperature and pressure by using a leakage locating formula. By adopting SCADA, GIS and simulation calculations, the invention can realize self monitoring and locating of the gas pipelines with the advantages of quick measuring speed, high measuring precision, low missing report rate and false report rate and the like.

The invention discloses a rapid sub pixel edge detection and locating method based on machine vision. The method includes the following steps that firstly, a detection image is acquired; secondly, denoising pretreatment is conducted on the image; thirdly, the gradient Gx of each pixel point in the horizontal direction and the gradient Gy of each pixel point in the vertical direction are calculated; fourthly, the gradient magnitude G0 and the gradient direction Gtheta of each pixel points under polar coordinates are calculated; fifthly, neighborhood pixel points of each pixel point are determined; sixthly, pixel-level edge points are determined; seventhly, the distance between a sub pixel edge point of each pixel-level edge point in the eight-gradient direction and the pixel-level edge point is calculated; eighthly, the distance d between each sub pixel edge point in the actual gradient direction Gtheta and the corresponding pixel-level edge point is calculated; ninthly, a cosine lookup table method is adopted for calculating rectangular coordinates of each sub pixel edge point in the actual gradient direction Gtheta, so that the image edge points are detected and sub-pixel-level localization is conducted. The whole method is high in calculation accuracy and speed.

The invention provides a digital image based detection method of a surface flaw of a steel rail. The method comprises the following steps of: extracting steel rail regions from shot steel rail images; simulating a human vision mechanism; and converting gray-scale images into contrast images so as to locate and judge suspicious flaw regions. The detection method provided by the invention has the advantages of high detection speed, high accuracy, self-adaption capability to the steel rail image acquisition environment and can effectively eliminate the influence of unfavorable factors, such as noise, illumination, and the like.

The invention belongs to the technical fields of biochips and diagnostic reagents, and discloses a method for detecting various respiratory viruses, and primers and probes thereof. In the invention, nucleotide sequences of 14 respiratory viruses, namely adenovirus, human metapneumovirus, influenza virus A, influenza virus B, respiratory syncytial virus, bocavirus, rhinovirus, coronavirus (HKU1, NL63 and SARS), and parainfluenza virus (type I, type II, type III and type IV) are analyzed, and corresponding reverse transcription primers, PCR primers and specific probes are designed. Specific gene segments are amplified by reverse transcription and multiple asymmetric PCR methods; a fluorescence-coded microsphere group coupled with the virus specific probes and the PCR amplification product are incubated and hybridized by liquid phase chip technology; and finally the Bio-PlexTM200 is used for detection. The detection method has the advantages of high flux, high specificity and sensitivity, stable results and good repeatability, the detection method is easy to operate, and the detection speed is high.

The invention discloses a pedestrian detection method based on video monitoring, which comprises the following steps of: rapidly detecting a pedestrian by utilizing an expanded gradient histogram characteristic and an Adaboost algorithm, and then further identifying and verifying the detected pedestrian by utilizing the gradient histogram characteristic and a support vector machine. In the invention, experiments based on a standard pedestrian database indicate that the method not only can greatly reduce the detection time, but also markedly improves the detection rate. As one of key technologies of target detection, pedestrian detection has broad application prospect in the fields of video monitoring safety, intelligent traffic management and the like.

The invention relates to a primer pair and a probe used for noninvasive polygene methylation combination detection for early stage colorectal cancer, and includes the primer pair and the probe used for detecting methylation of genes Spetin9, NDRG4, BMP3, THBD and SDC2 and the primer pair and the probe for internal reference ACTB; the sequences of the primer pair and the probe are represented as the SEQ ID No.1 to the SEQ ID No.18. The invention also provides a kit containing the primer pair and the probe and applications thereof. The application method includes free DNA extraction from a plasma specimen, sulfite conversion, PCR amplification reaction, fluorescent signal detection and result determination. The kit and the method are suitable for methylation detection of the five genes Spetin9, THBD, SDC2, NDRG4 and BMP3 in human peripheral blood; compared with a conventional colorectal cancer diagnosis method, the application method fully utilizes the free DNA extraction from a plasma specimen, the DNA methylation and QPCR associated technologies, thus developing the kit having high sensitivity and specificity. The primer pair, probe and kit are used for performing early stage noninvasive screening to human colorectal cancer.

The invention provides a novel full-automatic blood type analyzer, which comprises a sample disk subsystem, a reagent sampling subsystem, a reagent distribution subsystem, a reaction disk subsystem, a blowing mixing subsystem, a digital image recording and analyzing subsystem, a reaction cup cleaning subsystem, a computer control subsystem and a software management system, wherein the subsystems are arranged on an analyzer main frame and can perform automatic sample acquisition, and blood type analysis, recording and management respectively; the sample disk subsystem, the reagent sampling subsystem, the reagent distribution subsystem, the reaction disk subsystem, the blowing mixing subsystem, the digital image recording and analyzing subsystem and the reaction cup cleaning subsystem are connected with the computer control subsystem respectively; and a software management system and a digital image recording and analyzing program are stored in a master control computer of the computer control subsystem. The novel full-automatic blood type analyzer has the characteristics of high degree of automation, high detection speed, stable performance, high analysis accuracy, low cost and application to medical establishment.