31results about How to "Efficient detection effect" patented technology

The invention provides a learning-based intelligent detection method and device for display screen defects, wherein the method comprises the following steps of: obtaining an image to be detected of ascreen; determining suspicious defect information of the image to be detected according to preset network parameters; extracting a defect characteristic value according to the suspected defect information; determining a final detection result according to the defect characteristic value; collecting all the final detection results as a result set; randomly selecting the final detection result in the result set as a sampling sample, re-detecting the sampling sample, and determining that the detection result is different from the final detection result as a result to be fed back; feeding back theresult to be fed back, learning and generating the next preset network parameter according to the result to be fed back; according to the next preset network parameter, the preset network parameter being updated, and according to the updated preset network parameter, the next image to be detected being detected.

The invention provides an oblique parking stall identification method and device and a storage medium, and the method comprises the following steps: obtaining a sample image which comprises one or more oblique parking stalls, and identifying parking stall vertexes in the sample image through a parking stall vertex positioning network; pairwise pairing the parking space vertexes, intercepting imageblocks of the successfully paired parking space vertexes, and inputting the image blocks into a parking space classification network to identify inclined parking spaces; identifying parking space vertexes in the sample images by the aid of parking space vertex positioning networks, intercepting image blocks by the aid of the parking space vertexes serving as centers, extracting oblique lines in parking space lines, denoising the oblique lines and fitting the oblique lines by the aid of lines to obtain oblique line slopes; and marking an inclined parking space according to the vertex and the inclined line slope of the parking space.

The invention discloses a method for detecting train driver behavior and fatigue state on line and a detection system thereof. The method comprises: adopting a camera to shoot a train driver; carryingout human face detection on pictures collected by the camera; judging whether the driver does not perform duty or nods off; then, using an eye open image detector trained by the AdaBoost algorithm tofind an eye open image from the human face image so as to judge whether the driver opens eyes or not; and calculating the ratio of eye closing time to total testing time in real time, namely the PERCLOS value, for marking fatigue state. The system comprises a camera, an embedded host, an alarming device, a user press key and a memorizer, wherein, the camera, the alarming device, the user press key and the storage are respectively connected with the embedded host. The invention fills the gap of the field of monitoring train driver behavior and detecting fatigue state and can monitor train driver behavior and detect fatigue state under train environment.

The invention belongs to the technical field of deep learning and computer smart city monitoring, and discloses an anti-complex scene interference manhole cover category automatic detection method, which comprises the steps of S1, performing image collection and classification; S2, carrying out image annotation; S3, carrying out data expansion; S4, adopting a ResNet-101 model as a network skeleton, adding a series of convolution layers, pooling layers and full connection layers from large to small on the ResNet-101 model for predicting category information and position information of a manholecover in a picture, and constructing an R-FCN manhole cover detection model; S5, obtaining an optimal R-FCN manhole cover detection model through iterative training; and S6, identifying the picture or the video to obtain the category and position information of the manhole cover in the picture or the video. According to the invention, the damage of the manhole cover can be detected more efficiently by using the method based on deep learning. The invention further provides an anti-complex scene interference manhole cover category automatic detection system.

The invention provides a lithium battery surface defect detection method based on depth domain adaptation, and the method comprises the steps: designing an adaptation layer based on statistics such asmaximum mean value difference MMD and KL divergence in a classifier, and designing a domain discriminator for adversarial discrimination of which domain the extracted features come from after featureextraction. On the one hand, a complementary mechanism of the two modes can enable the extracted public features of the two domains to be more sufficient; on the other hand, the adaptive layer designbased on the statistics can enable the target domain data to participate in the training of the classifier, thereby enabling the model to have better generalization capability on the target domain. According to the model, a simple and effective multi-scale feature fusion strategy is designed in a feature extraction network, and a good recognition effect can be achieved on small defects. Accordingto the method, an efficient detection effect is achieved, the dependence of deep learning on label data is relieved, and the trained model has better generalization ability for target domain data.

The invention discloses a surface plasma resonance method for detecting a peanut allergen. The method specifically comprises the steps: respectively preparing an anti-peanut Ara h1 monoclonal antibody and an anti-peanut Ara h1 polyclonal antibody, fixing the anti-peanut Ara h1 monoclonal antibody on a detection chip by means of a coupling method, facilitating peanut Ara h1 proteins of different concentrations to flow by the detection chip, adding the anti-peanut Ara h1 polyclonal antibody which is used for magnifying a signal, so that a purpose of rapidly and efficiently detecting a main allergen protein of the peanut can be achieved. The method has characteristics of high sensitivity, high accuracy, rapidness in detection and the like and can become an important technical means for detecting the peanut allergen.

The invention discloses an organophosphorus pesticide molecular probe, a preparation and application method thereof and inorganic / organic composite rare earth upconversion nano material. The organophosphorus pesticide molecular probe is loaded to the surface of the rare earth upconversion nano material modified by tween to efficiently detect the organophosphorus pesticide concentration in a to-be-detected sample. The organophosphorus pesticide molecular probe has the advantages that the probe is high in sensitivity, good in stability, affordable, capable of achieving ratio type fluorescent detection of organophosphorus pesticide, and capable of providing early sensitive prewarning on the existing possibility of organophosphorus type nerve poison.

The invention provides an intelligent warranty maintenance system based on big data. The system comprises a server side, a client terminal and a maintenance terminal, the server side is provided witha database, the database comprises a text repair processing module, a photographing repair processing module, a household appliance identification module, an electronic map module and a maintenance service database, and the client terminal comprises a repair reception module, a database management module and an analysis statistics module. According to the intelligent warranty maintenance system based on the big data, verification is carried out through the safety shield, and it is guaranteed that personal information of the client terminal connected to the system has a safer guarantee; by theadoption of the whole warranty maintenance system, a user and maintenance personnel can conduct maintenance more conveniently and rapidly, interactive maintenance experience can be achieved in the maintenance process, and complete reference is provided; an ANN detection mode is adopted for detection, the labor intensity of manual detection is reduced, targeted detection is carried out through manual training and maintenance opinions in a database, and an efficient and reliable detection effect can be achieved.

The invention provides a method and system for detecting thermal anomaly of the earth's surface based on an unmanned aerial vehicle-carried infrared remote sensor. The method comprises the following steps: acquiring a real-time position of an unmanned aerial vehicle; acquiring the real-time height of the unmanned aerial vehicle; acquiring a thermal infrared image of the earth's surface; acquiring a multispectral image of the earth's surface; and fusing the real-time position and the real-time height of the unmanned aerial vehicle, the thermal infrared image and the multispectral image into a thermal infrared image map, inverting the thermal infrared image map into a surface temperature distribution map by using a surface temperature inversion module, and transmitting the surface temperature distribution map back to a control terminal. According to the method and system for detecting thermal anomaly of the earth's surface based on the unmanned aerial vehicle-carried infrared remote sensor in the invention, the method and the system can obtain the data in real time at the control terminal, carry out quick preprocessing and earth temperature interpretation, compare an interpretation result with an actual measurement temperature, check data quality in real time, and adjust a data observation scheme in time for a region with a poor interpretation effect; and therefore, efficient and real-time detection of the thermal anomaly of the earth's surface is realized.

The invention discloses a depth 6D pose estimation network model and a workpiece pose estimation method, the workpiece pose estimation method solves the problem that a large-volume point cloud data set is difficult to acquire through a method of generating a simulation data set by a physical engine, and a complete scene point cloud is directly used as input by a pure geometric point cloud coordinate. Through the semantic and instance segmentation part of the point cloud, the local and global features of the input point cloud can be extracted, the ability of the network to understand the scene is improved, the accurate attitude is output through the multi-layer feature fusion pose estimation network, the problems of object stacking and self-shielding are solved to a certain extent, the robustness is provided for various symmetrical objects, and the method is suitable for large-scale popularization and application. Through experimental verification of simulating a data set of a real scene, the method provided by the invention has obvious advantages in overall precision and stability, and has higher robustness.

The invention relates to the technical field of cable tension detection equipment, in particular to energy-saving tension detection auxiliary equipment for communication cable production, and the equipment comprises a detection frame, a driving rod is rotatably connected in the detection frame, an outer sleeve sliding block is sleeved outside the driving rod, and a crank connecting rod is movably connected to the right side surface of the outer sleeve sliding block, a steel wire rope is fixedly connected to the upper surface of the crank connecting rod, a positioning sliding plate is movably connected to the lower end of the crank connecting rod, and a connecting rod is movably connected to the front surface of the crank connecting rod through a shaft, with the increase of the pull force of the equipment to the cable, the clamping force of a limiting clamp plate on the cable is also increased, therefore, the effect of preventing the cable from falling off is achieved, the detection effect of the equipment is improved, and meanwhile, the equipment is suitable for detecting cables with different thicknesses, so that a power connection plate is in contact with a left electric plate, then the equipment is automatically stopped, the equipment is prevented from continuously working, and the energy-saving effect is achieved.

The invention discloses a mixed-order or fractional-order vortex light beam modal identification device and method. The device comprises a light beam generation device used for processing stray fundamental mode Gaussian light to generate Gaussian light in a horizontal polarization direction, and converting the Gaussian light into a horizontally-polarized mixed-order or fractional-order vortex light beam; a characteristic sample preparation device used for carrying out turbulence phase and two-dimensional fork-shaped vortex grating phase loading on the mixed-order or fractional-order vortex light beam, obtaining a far-field diffraction array after far-field diffraction, extracting a characteristic sample by utilizing a characteristic parameter extraction device, and obtaining a distortion characteristic sample required by full-connection neural network training and testing; anda network training and recognition device used for carrying out iterative training on the full-connection neural network through the distortion feature sample to obtain an optimal model, and then detecting the mixed-order or fractional-order vortex light mode in the test set through the optimal model. The invention aims to realize high-speed, high-precision and large-range vortex light mode detection, thereby improving the performance of OAM communication.

The invention discloses a manufacture method for a long-period fiber bragg grating hydrogen sulfide gas sensor. One layer of citric acid-molybdenum disulfide composite sensitive membrane is attached to the long-period fiber bragg grating so as to use the obtained long-period fiber bragg grating hydrogen sulfide gas sensor to carry out concentration detection on the hydrogen sulfide gas. The long-period fiber bragg grating hydrogen sulfide gas sensor provided by the invention has the advantages of simple structure, stability in working, small volume and light weight. The manufacture method is simple and low in cost, has a good detection effect on the hydrogen sulfide gas and has the advantages of rapid response time and accuracy and high reliability.

The invention discloses a multi-target tracking method and tracking system for aerial images of an unmanned aerial vehicle. The multi-target tracking method comprises the steps that a Faster RCNN detection network carries out target detection on an input image sequence; an IOU Tracker tracker performs multi-target tracking on the detected image sequence to form an initial tracking trajectory; theKCF tracker carries out updating compensation on the initial tracking trajectories, and forward updating and backward updating are carried out on each initial trajectory in sequence; the method comprises the following steps: designing a trace vote (trajectory vote) algorithm, and for a trajectory updated by a KCF, calculating the trace vote by using a greedy algorithm and an IOU (intersection-parallel ratio) to classify and fuse a plurality of trajectories. According to the invention, a high-robustness, high-precision and high-efficiency multi-target tracking effect can be realized under the condition that the number of targets in aerial images of the unmanned aerial vehicle is large and the target scale is small.

The invention discloses a straightening detection device for cast-iron pipes. The straightening detection device comprises a box and is characterized in that a sound insulation layer is arranged on the inner wall of the box, a guide rail is fixedly mounted at the top in the box, a slide seat is slidably mounted on the guide rail, an air cylinder can be moved up and down is mounted on the slide seat, a servo motor is further mounted on the slide seat and can control the slide seat to move leftwards and rightwards so as to perform selective detection on different pipes, three different support seats are arranged at the bottom of the box, an arc-shaped pipe groove is formed in each support seat, and the lower end of the piston rod of the air cylinder is fixedly connected with a pipe rack. The straightening detection device has the advantages that the device is high in detection efficiency, the base plate of the support seat of each to-be-detected pipe is provided with detection rollers, each support seat is controlled to move forwards and backwards through a screw, comprehensive detection of each to-be-detected pipe is achieved, the detection process can be fast regulated by a parameter saving and memorizing module, and the device is high in automation level.

The invention discloses a sensor embedded bedding. The sensor embedded bedding includes a flexible contact member, a micro movement sensor, and a supporting part; the micro movement sensor and the supporting part are arranged in the flexible contact member; the micro movement sensor is arranged above the supporting part in a state where a sensing surface is parallel to a stress surface of the bedding; and the rigidity of the supporting part is greater than the rigidity of the flexible contact member. The hard micro movement sensor is embedded in the flexible contact member, the supporting portion having the rigidity greater than the rigidity of the supporting part is arranged in a projection region so as to support the hard micro movement sensor; when the bedding is pressed, the flexible bedding is supported by the supporting bed, a user can lay any position of the bedding, a vibration signal of a human body can be effectively sent to the hard micro movement sensor, the touch experience and the vision experience of the bedding can be maintained, and a high detection effect can be achieved.

The invention belongs to the technical field of plant science and discloses an efficient extraction method suitable for measurement of endogenous hormone of musa and heliconia plants. The method comprises the following steps: weighing 0.2g of banana leaves, quickly freezing liquid nitrogen, grinding the quickly frozen liquid nitrogen into powder, adding 6mL of pre-cooled 80% methanol solution, andperforming extraction over night in a refrigerator with the temperature of 4 DEG C; adding 2mL of 80% methanol for washing residues, repeating the operation twice, and mixing supernatant liquid; enabling the supernatant liquid to pass through a 0.22-micron micro-porous filtering film, adding 2mL of 80% methanol for washing residues of the micro-porous filtering film, transferring the filtered supernatant liquid into a rotary evaporation bottle, performing evaporation under reduced pressure at 36 DEG C by dim light until the water phase is about 2mL, taking the water phase, and washing the rotary evaporation bottle twice with 2mL of pure water, totally about 6mL; performing refrigeration for 10min at -80 DEG C, and then, performing freeze-drying to obtain dry powder; performing dissolutionwith 4mL of a phosphate buffer of which the pH is 3.0, performing extraction thrice with an equal volume of ethyl acetate to leave an ester phase, performing evaporation to dryness under reduced pressure at 40 DEG C by dim light, performing dissolution with a mobile phase of a pump A, and performing treatment with a Pak-ODS-C18 small column; and before HPLC analysis is performed, enabling the sample to pass through the 0.22-micron micro-porous filtering film to serve as a test solution. The method disclosed by the invention has the advantages that the existing extraction process is adopted, the operation is simple, the consumed time is short, the time and labor are saved, and the recovery rate is increased.

The invention discloses an organophosphorus pesticide molecular probe, a preparation and application method thereof and inorganic / organic composite rare earth upconversion nano material. The organophosphorus pesticide molecular probe is loaded to the surface of the rare earth upconversion nano material modified by tween to efficiently detect the organophosphorus pesticide concentration in a to-be-detected sample. The organophosphorus pesticide molecular probe has the advantages that the probe is high in sensitivity, good in stability, affordable, capable of achieving ratio type fluorescent detection of organophosphorus pesticide, and capable of providing early sensitive prewarning on the existing possibility of organophosphorus type nerve poison.

A novel cigarette filter stick end-surface quality on-line test and rejection device and a test method of cigarette filter sticks. The device includes a cigarette filter stick end-surface test device arranged on the position of a channel. The cigarette filter stick end-surface test device includes a transporting and transmission apparatus, a filter stick end-surface finishing apparatus, a visual test apparatus and a rejection apparatus. One end of the transporting and transmission apparatus is connected to a filter stick tray unloader or a storage and sending channel and the other end of the transporting and transmission apparatus is successively connected to the filter stick end-surface finishing apparatus, the visual test apparatus and the rejection apparatus. The rejection apparatus is connected to a material storage tank of a filter stick emitting machine. The device can be used for on-line test of the filter stick before use, so that defects, including head shrinking, hair generating, hole melting, deformation and unflatness on the end surfaces and the like, can be effectively rejected during manufacturing and transportation of the cigarette filter sticks. Meanwhile, the device can obtain a high-efficient, accurate and stable test result in a situation of a transportation channel, which is free of a fixed operation station, in the filter stick emitting machine.

The invention discloses a wire harness rapid detection system. The system comprises a tester, test heads, a fixing tool and an upper computer. The fixing tool is used for fixing wire harnesses, the fixing tool comprises a bottom plate and a wire harness clamp arranged on the bottom plate, the wire harness clamp can fix a plurality of wire harnesses arranged in parallel, the test heads are locatedat the two sides of the fixing clamp, the test heads can be close to / far away from two end parts of the wire harness under the action of a driving piece and are conducted with the wire harness, the driving piece is fixed on the bottom plate, the test heads positioned on the same side are connected with a wire harness joint, the wire harness joint is connected with a detection port of the tester, the tester is used for connecting an electrical parameter of the wire harness, and the tester is in communication connection with the upper computer. And a detection result is transmitted to the uppercomputer. The system can be used for simultaneously and rapidly detecting the plurality of wire harnesses and improves detection efficiency.

The invention discloses a mixed-order or fractional-order vortex beam mode identification device and method, the device includes: a beam generating device, which is used to process the heteropolar fundamental mode Gaussian light to generate Gaussian light in the horizontal polarization direction and then transform it It is a horizontally polarized mixed-order or fractional-order vortex beam; a characteristic sample preparation device is used to perform turbulent phase and two-dimensional fork-shaped vortex grating phase loading on the mixed-order or fractional-order vortex beam. After far-field diffraction, the far-field field diffraction array, and use the characteristic parameter extraction device to extract the characteristic samples to obtain the distortion characteristic samples required for the training and testing of the fully connected neural network; the network training and identification device is used to iteratively train the fully connected neural network through the distortion characteristic samples to obtain After optimizing the model, the mixed-order or fractional-order vortex optical modes in the test set are detected by the optimal model. The invention aims to realize high-speed, high-precision and wide-range vortex optical mode detection, thereby improving the performance of OAM communication.

The invention discloses a grain quality on-line detector transparent round barrel dust clearing device. The grain quality on-line detector transparent round barrel dust clearing device comprises a transparent round barrel, a fixed base plate, a driving component and a dust clearing component, wherein the driving component is arranged on the fixed base plate, one end of the dust clearing componentis connected with the driving component, the other end of the dust clearing component is arranged inside the transparent round barrel in an inserted manner, and the dust clearing component is in contact with the inner wall of the transparent round barrel in the rotation process. The grain quality on-line detector transparent round barrel dust clearing device is good in dust clearing effect, the cleanliness of the inner surface of the transparent round barrel can be significantly improved, a grain quality on-line detector is effectively guaranteed to have the high-quality, high-stability and high-precision detection performance, and the grain quality on-line detector transparent round barrel dust clearing device has a quite important application prospect.

The invention provides a method and system for detecting thermal anomalies on the surface of an unmanned aerial vehicle equipped with infrared remote sensing. The method comprises: acquiring the real-time position of the drone; acquiring the real-time height of the drone; acquiring a thermal infrared image of the ground surface; acquiring a multispectral image of the ground surface; The spectral image is fused into a thermal infrared image map, and the surface temperature inversion module is used to invert the thermal infrared image map into a surface temperature distribution map, and the surface temperature distribution map is sent back to the control terminal. The ground surface thermal field detection method and system provided by the invention provided by the UAV equipped with infrared remote sensing can obtain data in real time at the control terminal, perform rapid preprocessing and ground temperature interpretation, compare the interpretation result with the actual measured temperature, and check the data quality in real time. Adjust the data observation plan in time for areas with poor interpretation effects, so as to realize efficient real-time detection of surface thermal anomalies.

The invention discloses a machining and producing device for cast iron pipes. The machining and producing device comprises a box body. The machining and producing device is characterized in that a sound insulation layer is installed on the inner wall of the box body, a guiding rail is fixedly installed at the top in the box, a sliding seat is slidingly installed on the guiding rail, an air cylinder capable of being lifted is installed on the sliding seat, a servo motor is further installed on the sliding seat, the servo motor can control the sliding seat to conduct left and right moving so as to conduct selection and detection on different pipes, three different supporting seats are installed at the bottom of the box body, arc pipe grooves are formed in the supporting seats correspondingly, and a pipe frame is fixedly connected to the lower end of a piston of the air cylinder. According to the machining and producing device for the cast iron pipes, detection efficiency is high, using is simple, the detecting process is adjusted quickly through a parameter saving and memorizing module, the automation degree is high, and detection efficiency is high.

The invention belongs to the field of functional fiber membrane preparation, and particularly relates to a preparation method of a salamander skin collagen fiber membrane capable of efficiently detecting and adsorbing nicotine. By exploring a salamander skin enzymolysis technology and fiber membrane preparation conditions are explored, the salamander skin collagen fiber membrane capable of efficiently detecting the nicotine and efficiently adsorbing the nicotine is obtained. The preparation method has the advantages that the salamander skin is effectively used as a collagen preparation source, and the salamander skin collagen fiber membrane is obtained; the fiber membrane has an efficient detecting effect on the nicotine and capable of being used for preparing a corresponding nicotine sensor; the fiber membrane further has a high adsorption effect on the nicotine and capable of being used as a nicotine adsorbing material; a spinning solution in use is low in toxicity, the obtained fiber membrane can be safely used as a cigarette-end filter tip material after being subjected to volatile treatment through a common solvent.

The invention discloses an ultrasonic apparatus for an obstetrics and gynecology department. The ultrasonic apparatus for the obstetrics and gynecology department comprises a mainframe shell, whereina base is arranged at the lower end of the mainframe shell; the base is closely welded to the mainframe shell; universal wheels are arranged at the lower end of the base and are fixedly connected withthe outer surface of the bottom of the base ; a plug is arranged on the outer surface of the mainframe shell and is movably connected with the mainframe shell; an outer wire is arranged at the lowerend of the plug and penetrates through the plug. The left side of the mainframe shell of the ultrasonic apparatus for the obstetrics and gynecology department is closely attached to miniature probes,when the apparatus transmits a high-level electric signal, the electric signal can be effectively converted into ultrasonic wave to enter the body of a patient by vibration of the miniature probes, meanwhile, the received ultrasonic wave can be converted into a high-frequency electric signal which is displayed by a display screen, moreover, the two miniature probes with different structures are arranged, the ultrasonic apparatus can adapt to various working environments, and therefore, an effective detection effect can be achieved.