54results about How to "Improve detection efficiency and detection accuracy" patented technology

The invention discloses an automatic detecting system and a method for an intelligent grid data collecting terminal. The system comprises a backend database, an industrial personal computer (IPC) and a detecting tooling. The backend database memorizes detecting configuration parameters and detecting conclusion information. The IPC receives the request for detecting a detected terminal and sent by the detecting tooling, reads the detecting configuration parameters from the backend database, and sends detecting item reports to the detecting tooling. The IPC receives detecting results sent back by the detecting tooling and obtains the detecting conclusion information according to the detecting results, memorizes the detecting conclusion information into the backend database and sends the detecting conclusion information to the detecting tooling. The detecting tooling sends the detecting request to the IPC and receives the detecting item reports; if the detecting item reports need to be processed by the detecting tooling, the detecting tooling processes the detecting item reports and sends back the detecting results; if the detecting item reports need to be processed and detected by the detected terminal, the detecting item reports are forwarded to the detected terminal, and the detected terminal sends back the detecting results to the IPC through the detecting tooling; and the IPC receives and displays the detecting conclusion information. The automatic detecting system and the method for the intelligent grid data collecting terminal can improve detecting efficiency and correctness.

The invention discloses a high-resolution remote sensing image change detection method and system. The method comprises the steps of obtaining neighborhood characteristics of each pixel in two time-phase images; inputting the neighborhood features into a branch convolutional neural network model to obtain types of the image, wherein the types is a change type and an unchanged type, the branch convolutional neural network model is obtained by carrying out unsupervised pre-training on an image, the branch convolutional neural network model comprises an input layer, a branch network, a full connection layer and a classification layer, the branch network is used for taking neighborhood features as tags and carrying out convolution dimension reduction on the neighborhood features, and the classification layer is used for classifying the image into different categories. According to the present invention, the technical problem that the image change detection precision is not high in the prior art is solved.

The invention provides an automatic skip-stop detection system capable of detecting multiple lots of products on any detection station. The automatic skip-stop detection system is characterized by comprising a storage module, an acquisition module, a processing module and an execution module; information of various stations is stored by the storage module; the acquisition module is used for scanning each lot of products passing through the station and acquiring corresponding station information of the products through the storage module; the processing module is used for selecting different sampling detection rules according to the station corresponding to the lot of products; the execution module is used for detecting the products. Different sampling detection rules are set to realize sampling detection of different lots of products according to lot difference and importance degree of the corresponding production stations, so that detection precision and detection efficiency are improved.

The invention relates to an intelligent ocean oil spill detection method for a remote sensing large image. The method includes the following steps of (1) remote sensing large image input and processing; (2) AOI detection of suspected oil spill areas; (3) oil spill area extraction based on CFAR. The intelligent ocean oil spill detection method for the remote sensing large image has the advantages of greatly improving detection efficiency and detection accuracy of ocean oil spill of the remote sensing large image.

The invention provides a pipeline defect identification method. A pipeline robot transmits a video in a pipeline in real time, samples the video into a key frame image, effectively identifies and captures pipeline defects in the image, and stores an extracted defect image positioning mark. According to the method, the basic idea of DefectNet is used for solving the target defect detection as a regression problem. Firstly, feature extraction is performed through a convolutional neural network to obtain a feature map of a certain size; the then, feature maps of different scales are divided intoa plurality of grids by multi-scale prediction, and the grid where the center of a defect target is located is responsible for predicting the defect target; and finally, target classification and frame regression are carried out, and each grid judges the category to which the defect target belongs and adjusts the frame position. Compared with the prior art, the method has the advantages that the detection precision and the detection efficiency are improved, and the method has high innovativeness and practical value and is suitable for popularization.

An embodiment of the invention discloses a printed matter color quality detection method and system. A color code and a positioning core which has a position setting relation with a pure color area in a printed matter are arranged in each image-text cycle of the printed matter. A photoelectric sensor in a first position sends a color code emergence signal to a processor when detecting that the conveyed printed matter have a color code. When the printed matter with the color codes moves to a second position, the processor triggers an image acquisition device for image acquisition and searches the positions of the positioning cores. The processor determines the position of a pure color area according to the position of the corresponding positioning core and calculates the position adjustment amount of a fiber optical spectrometer according to the position of the pure color area, and a motion control mechanism drives the fiber optical spectrometer to move to a designated position. When the detected printed matter moves to a third position, the fiber optical spectrometer performs optical spectrum acquisition, and optical spectrum data are converted into LAB data and then are compared with a standard value to obtain a detection result. The printed matter color quality detection method and system have high detection precision and detection efficiency.

The invention provides an optical detection-based E type magnet sorting method, and relates to an optical detection-based magnet sorting method. The problem that the existing E type magnet detection method is low in detection efficiency and bad in detection precision is solved. The method comprises the steps of: photographing an E type magnet to be detected to obtain an image of the E type magnet, extracting and fitting the magnet edge of the image, detecting the length of the magnet edge of the image to obtain the length information of the E type magnet to be detected, and sorting. In order to improve the detection precision, a step of denoising the image is added in the method and comprises the Hough conversion denoising and the least squares fitting denoising; and in order to improve the quality of a fitting straight, the method can be used for obtaining the edge of the E type magnet to be detected by a subpixel solving technology and the fitting straight. The method is suitable for sorting the E type magnet.

The invention relates to a separated-working-position detecting device for the leakage of high precision gas. The detecting device comprises a standard container, a detected container and gas source. A pressure reduction detecting valve is connected to a detecting pipeline and a standard container pipeline by a main connector valve; the other end of the detecting pipeline is connected with the detected container; the other end of the standard container pipeline is connected with the standard container; a differential pressure connector valve and a differential pressure sensor are arranged between the detecting pipeline and the standard container pipeline; a gas filled pressure reduction valve is connected with a gas filled connector valve; the detecting pipeline is connected with an intelligent compensation controller which is used for determining compensation rate and controlling the feed amount of a stepping motor according to a differential pressure signal; the detected container is arranged on a workbench and is characterized in that the workbench comprises a gas filled working position, a balance working position, a detecting working position, a gas exhaust working position and a working position switching mechanism; the gas filled working position is provided with a gas filled connector valve seat; the balance working position is provided with a balance connector valve seat; the detecting working position is provided with a detecting connector valve seat, and the gas exhaust working position is provided with a gas exhaust connector valve seat. The invention can improve detecting precision and detecting efficiency.

The embodiment of the invention discloses a server abnormity detection method, device and equipment. The scheme comprises the following steps: acquiring capacity monitoring data of each server in a first server set in a preset time period; aiming at any server in the first server set, judging whether the capacity monitoring data corresponding to any server is in a normal range or not according tothe capacity monitoring data corresponding to other servers in the first server set, and obtaining a first judgment result; and if the first judgment result indicates that the capacity monitoring dataof any one server is outside the normal range, determining that any one server is an abnormal server.

The invention relates to a fuel gas watchcase leakage detection device and detection method. The fuel gas watchcase leakage detection device and detection method comprise a vacuum air pump, a pump box and transparent casing sealing plates, wherein the vacuum air pump is arranged on the pump box and is communicated with the pump box, the pump box is linked to an external tooth head of the casing in a sealing mode through windpipes and windpipe connectors, rubber plugs are arranged on other corresponding open portions of the casing, and sealing rings are laid on the upper end opening of the casing, and casing sealing plates are arranged on the sealing rings. According to the fuel gas watchcase leakage detection method, moderate soap water is poured into the casing, and the soap water soaks each detection component; the vacuum air pump is opened, and when negative pressure in the casing reaches to the requirement, transparent acrylic plates at the casing opening are used to observe whether bubbles occur in water soap of relevant detection portions.

The invention provides a crane track detecting device. A track distance measuring host machine and a height difference measuring host machine need to be used in cooperation; the track distance measuring host machine and the height difference measuring host machine are respectively provided with a first crawling driving device and a second crawling device; the track distance measuring host machineand the height difference measuring host machine can be automatically controlled to move on tracks to realize measurement; the signal acquisition device of the height difference measuring host machinecan collect a laser beam emitted by the laser distance measuring device of the track distance measuring host machine; during the movement of the track distance measuring host machine and the height difference measuring host machine, the signal acquisition device can measure the longitudinal deviation of the laser beam, namely height difference between the two tracks; the laser distance measuringdevice can measure the distance between the two tracks by the retracing process of the laser beam of the laser distance measuring device; and the first leveling device of the track distance measuringhost machine can keep the laser beam emitted by the laser distance measuring device level; the second leveling device of the height difference measuring host machine can keep the signal acquisition device level; and therefore, detection precision can be improved.

The invention provides a welding spot defect detection system based on machine vision, comprising an image collection device, a control system and a processing system, wherein the welding spot defectdetection system is mainly applied to welding spot defect detection on an electronic product, and welding objects are irregularly arranged welding spots on stainless steel; the image collection deviceis used for collecting an image of the surface of the detected electronic product and transmitting the collected image to the processing system through data; the processing system is connected with the image collection device and is used for receiving the collected image, processing the collected image, and identifying and judging a defect of the welding spot in the image; and the control systemis connected with the image collection device and the processing system and is used for controlling the image collection device to collect the image and controlling the processing system to process the image. Compared with the detection instrument in the prior art, the welding spot defect detection system provided by the invention has higher detection efficiency and detection precision.

The invention is applicable to the technical field of image processing, and provides an optical filter mirror plane defect detection method and device and a terminal device. The optical filter mirrorplane defect detection method comprises the steps that an optical filter image containing a plurality of optical filters is obtained firstly; a single optical filter image corresponding to each optical filter is extracted from the optical filter image according to preset positioning parameters; secondly, the angle of each single optical filter image is adjusted by using an affine transformation method; mirror plane area image corresponding to each optical filter is extracted from the adjusted single optical filter images correspondingly; and finally, according to preset detection parameters, the mirror plane area image corresponding to each optical filter is detected. The optical filter mirror plane defect detection method and device and the terminal device can realize automatic detectionof the mirror planes of the optical filters, improve the detection efficiency and detection accuracy of the optical filter mirror plane defects, and further improve the overall quality of the opticalfilters.

The embodiment of the invention discloses a radio frequency interference automatic test method, system and device and a storage medium. The method comprises the steps of acquiring a test image of a to-be-detected object; determining a detection area on the test image; determining spatial position coordinates of the detection area according to the pixel coordinates of the detection area; automatically testing radio frequency interference of the detection area based on the spatial position coordinates of the detection area to obtain a detection result; and generating a visual interference detection report according to the detection result. According to the embodiment of the invention, automatic detection of radio frequency interference and intelligent analysis and visual presentation of a test result are realized, and the detection precision and the detection efficiency are improved.

The invention provides a method and system for detecting the tunnel state of channel services in a ring network. The method comprises the steps of determining network equipment related to each channelservice and detection items related to each network equipment tunnel state; and for each set of network equipment related to each channel service, according to the detection items related to the network equipment tunnel state, calling a detection control corresponding to the detection items to detect the network equipment tunnel state so as to obtain a detection result. According to the invention, manual intervention operation in the tunnel state detection process of the channel service can be reduced, automatic and rapid detection of the tunnel state of the channel service is realized, the workload of the operation and maintenance personnel is reduced, and the detection efficiency and the detection accuracy are improved.

The invention belongs to the technical field of cylinder cover detection equipment, and provides a cylinder cover dry-type air tightness detection device. The device comprises a rack, a roller way conveying line driven by a driving device is arranged on the rack, and a guide slideway and a sensor for realizing cylinder cover position detection are arranged on the roller way conveying line; a positioning stop block driven by a first cylinder and a lower sealing plate driven by a hydraulic cylinder are arranged at the bottom of the rack; an upper sealing plate is installed at the top of the rack, the two ends of the upper sealing plate are provided with side sealing plates which are driven by a second air cylinder respectively, the clamping faces of the lower sealing plate, the upper sealingplate and the side sealing plates are provided with sealing rings corresponding to cylinder cover holes in position, and the clamping face of the sealing plate on one side is provided with an air tightness detection hole communicated with an air supply device. And an air tightness detector is further mounted at the top of the rack. According to the invention, automatic airtightness detection of the cylinder cover can be realized, the detection efficiency and the detection precision are greatly improved, the leakage amount can be detected, and the tedious process of treatment after immersion of a test piece is omitted.

The invention provides a method for detecting discontinuity of a nonmetal material by virtue of inductance of a series resonant circuit. The method has the beneficial effects that the discontinuity of the nonmetal material is detected by virtue of an inductance coil in a series resonant circuit device based on an electromagnetic principle, and the limitation of a current nondestructive testing theory is broken through by virtue of the characteristic that the resonant frequency of the series resonant circuit is changed due to the changing of a medium in the inductance coil in the series resonant circuit; by applying the electromagnetic detection method to the detection on the discontinuity of the nonmetal material, the automatic detection is easy to implement, and the efficiency and precision of detection on the discontinuity of the nonmetal material are improved.

The invention provides a current detection circuit used for detecting the current through a product to be detected. The current detection circuit comprises an operational amplifier, a sampling unit, an adjusting unit and a measuring unit. The product to be detected is connected in series with the sampling unit and is connected between a voltage source and the ground, the operational amplifier comprises a first input terminal, a second input terminal, a first adjusting terminal and a second adjusting terminal, the first input terminal and the second input terminal are connected with two ends of the sampling unit, two ends of the adjusting unit are respectively connected with the first adjusting terminal and the second adjusting terminal, and therefore the adjusting unit can adjust the amplification factor of the operational amplifier, and the measuring unit is connected with the output terminal of the operational amplifier. The current detection circuit can detect and calculate minimal current signals of precision electronic devices such as watches or integrated circuits conveniently and accurately, and therefore the detection precision and the detection efficiency are improved.

The invention discloses a railway fastener system comprehensive performance detection device and detection method. A sleeper is horizontally and transversely fixed on a cast iron platform, a steel rail is installed on a rail bearing surface of the sleeper, and a fastener system to be detected buckles the steel rail on the sleeper. A portal frame beam is installed on the cast iron platform throughportal frame supporting columns, a vertical loading base of a vertical loading actuator is installed on the lower surface of the portal frame beam, and a vertical loading end vertically extends downwards to the steel rail from the vertical loading base. The fixed end of a reaction frame is installed on the cast iron platform, a horizontal loading base of a horizontal loading actuator is installedat the supporting end of the reaction frame, and a horizontal loading end horizontally extends to the steel rail from the horizontal loading base. The detection device applies a vertical load and a horizontal load to the steel rail to be detected, the detection requirements of buckling pressure, static rigidity and longitudinal resistance of the steel rail are met at the same time, the load precision and the displacement precision of the vertical loading end and the horizontal loading end are high, and the detection precision and the detection efficiency are high.

This invention relates to one light mask pattern position test method, which comprises the following steps: totalizing mask film edit pattern each filling area as reference area; forming fill-in pattern module graph by each fill graph position and drawing software in the pattern; computing fill-in pattern module area in the graph process software and comparing the reference total area with the filled area; if not, then correcting the light mask pattern design and repeating above steps; if yes, then closing the test.

The embodiment of the invention provides an automobile coating surface defect detection method and a system based on deep learning. The method comprises the steps of acquiring a detection image of a to-be-detected area of an automobile coating surface; uniformly segmenting the detection image into a plurality of small images with set sizes so as to form a small image test set; inputting the smallimage test set into a deep learning network model, and obtaining a detection result list corresponding to the small images of the set size; wherein the set sizes are N types, and N is greater than orequal to 2; and performing de-duplication integration processing on the N detection result lists to obtain a defect detection result. According to the method of the invention, the workload of detection personnel can be effectively reduced, the detection efficiency and the detection precision are improved, and the small defects in the complex area of the automobile coating surface can be efficiently and accurately detected under the non-uniform and uncertain complex illumination condition.

The invention discloses a totally-enclosed nucleic acid extraction and detection integrated kit, which is characterized in that a plurality of areas are arranged in one kit, and the kit comprises a shell, a cleaning solution injector, an eluent injector and a PCR (Polymerase Chain Reaction) tube which are respectively filled with a lysis solution, a cleaning solution, an eluent and a reaction solution; when a sample is detected, the external push rod is pushed, a lysis solution mixed with the sample passes through the silicon material membrane, nucleic acid released by lysis of the sample is adsorbed by the silicon material membrane, along with pushing of the external push rod, the cleaning solution and the eluent are sequentially extruded out, and nucleic acid purification and elution arecompleted on the silicon material membrane. During elution, the shell integrally moves downwards with the help of external auxiliary equipment, the needle head penetrates into the closed PCR tube, and the external push rod is pushed to extrude purified nucleic acid to be mixed with reaction liquid. The PCR tube is subjected to temperature control and nucleic acid amplification reaction through the temperature circulating device, the optical detection equipment detects the amplification result, and totally-enclosed nucleic acid extraction detection is completed.

The invention discloses a bridge safety monitoring method based on big data analysis and machine vision and a cloud platform, and the method comprises the steps that the whole bridge floor of a large-span bridge is divided into a plurality of sub-regions, and the image obtaining and image preprocessing of the bridge floor of each sub-region is carried out through combining with high-definition cameras; meanwhile, feature extraction and road damage type matching are conducted on the bridge floor images of all the sub-regions, the depths of cracks and tracks in the bridge floor road damage types of all the sub-regions are detected, and then statistics is conducted on the comprehensive safety influence coefficient of the bridge, so that safety early warning of different levels is conducted and displayed. The method meets the requirement for rapid detection of bridge floor damage of the large-span bridge, improves detection efficiency and detection precision, reduces the cost of manpower and material resources, guarantees the safety of detection personnel, and greatly prolongs the service life of the bridge.

The invention relates to a control system of an automatic hydraulic torque wrench testing device. The control system comprises a detecting system monitoring upper computer, a detecting system PLC (programmable logic controller), a detecting device torque sensor, a hydraulic pump station pressure sensor, a hydraulic pump station servo motor and an emergency stop button. By a hydrodynamic wrench testing device control system, automatic hydraulic wrench testing is realized, influences caused by human factors to hydraulic wrenches are avoided, and accordingly detecting efficiency and detecting precision of the hydraulic wrenches are improved.

The invention discloses an automatic positioning and crimping device for the wiring terminal of an intelligent electric energy meter. The device comprises a base, a test meter holder and a slide plate. One side of the top end of the base is provided with the test meter holder and a detachable wiring pile. Second sliding grooves are formed in the two sides of the top end of the base correspondingly. The sliding plate is slidably mounted at the top end of each second sliding groove through a first sliding block. Second baffles are installed on the two sides of the top end of the sliding plate. Electric telescopic rods and fixed plates are mounted on the opposite sides of the second baffle plates; a first baffle plate is mounted on one side, far away from the test meter holder, of the base; an air cylinder is installed at the end, close to each second sliding groove, of the first baffle, a fixing block is arranged at the bottom end of the sliding plate and connected with the output end ofthe air cylinder, a controller is installed above the first baffle plate, a third baffle plate is installed at the end, away from the detachable wiring pile, of the sliding plate, and an infrared locator is installed at one end of the third baffle plate. Automatic positioning can be realized, different types of electric energy meters can be detected and the structure is simple.

The invention belongs to the technical field of biological detection and particularly discloses a coronavirus and influenza virus detection device and method. The device comprises a detection module, a signal processing circuit, a controller, a displayer, a digital-to-analog conversion circuit and a clock, wherein the detection module is a sample cell and a transistor sensor combination which is integrated in the sample cell and is used for measuring different targets, and the detection module can comprise a sample cell array, so a plurality of to-be-detected samples can be detected at the same time. The detection method comprises the following steps of adding a to-be-detected sample into the sample pool, and reading the electric signal response of each transistor sensor in the sample pool to judge whether the to-be-detected sample contains the to-be-detected virus or not. According to the method, the shortest detection time for the coronavirus and the influenza virus is 1 minute, and the method can be used for detecting a single-person sample, a mixed sample or a mixed sample and has potential social and economic values.

The present invention discloses a spray hole optical batch detecting device for an aerosol inhalation driver. The aerosol inhalation driver comprises a main body and a nozzle at one end of the main body; a spray hole for providing an aerosol is formed in the nozzle. The optical detecting device includes a light source, a camera and an image processing apparatus; the light source illuminates one end of the main body, wherein the one end of the main body is far away from the nozzle; the lens of the camera faces the spray hole; and the image processing apparatus processes a picture taken by the camera. According to the detecting device, the diameter of the spray hole is obtained by means of an optical mode and a photographing mode. The spray hole optical batch detecting device for the aerosolinhalation driver has the advantages of simple and convenient operation, high accuracy of acquired data and the like.

The invention belongs to the technical field of metal material thickness measurement, and particularly relates to an optical path alignment self-adjusting laser ultrasonic metal composite plate thickness measuring device and method.The optical path alignment self-adjusting laser ultrasonic metal composite plate thickness measuring device is characterized in that a camera coaxial mounting block or a sample clamp is mounted in the middle of a sliding rail; the pulse laser CCD camera and the continuous laser CCD camera are installed on the two sides of the camera coaxial installation block respectively, the pulse laser developing screen and the continuous laser developing screen are installed in front of lenses of the pulse laser CCD camera and the continuous laser CCD camera respectively, and the pulse laser diaphragm and the continuous laser diaphragm are symmetrically installed on the left side and the right side of the sliding rail. Through the combination of the pulse laser CCD camera, the continuous laser CCD camera, the pulse laser diaphragm and the continuous laser diaphragm, it can be guaranteed that two laser beams are adjusted to be coaxial within the error range, and it is avoided that only an excitation point and a detection point are adjusted to be collinear in the thickness direction.

The invention provides a home camera imaging detection device comprising a base, wherein a placing table is arranged on the base, a power supply device is arranged on the base, the power supply device is connected with a product on the placing table, an imaging detection device is arranged above the placing table, and a detection result display device is arranged near the imaging detection device. The camera is placed on the placing table and is powered by the power supply device, the working performance of the camera is detected by the imaging detection device arranged at the upper side to improve the detection precision and the detection efficiency, the position of the camera is determined by setting a contact sensor to avoid a detection result error caused by a position error, and the position of the imaging detection device is adjusted by a mobile device, so that the imaging detection device is suitable for detecting cameras with different specifications.