120results about How to "Eliminate background distractions" patented technology

The invention provides a lane line identification method and a lane line identification system. The lane line identification method comprises steps that an interested area is delineated in an acquired gray level image comprising a lane line; a matched filter is used for multidimensional matching filtering; the top hat transformation of the interested area of the image after the multidimensional matching filtering is carried out; by comparing a preset gray level adaptive threshold with a to-be-detected pixel point, a lane line characteristic point is screened out; the back projection transformation of the lane line characteristic is carried out to acquire a two-dimensional raster map; the neighborhood clustering processing of the two-dimensional raster map is carried out, and the curve fitting of the lane line pixel point is carried out, and then the lane line of the lane, on which a vehicle drives, is determined by combining with the current position of the vehicle. The characteristic enhancement of the lane line area in various environments is realized, and influences of changes of external conditions on the lane line identification based on machine vision are greatly reduced, and the accuracy of the subsequent lane line identification is improved.

The present invention provides a detecting method for the unusual point of the high spectrum image based on the selective nuclear main composition analyzing, and relates to the high spectrum image analyzing and defecting technology field. The present invention aims at solving the problem that the prior high spectrum image defecting method cannot conduct character extracting on the unusual point effectively, and produce more false alarm, and cannot detect the unusual point effectively under high background disturbance. The present invention has the procedures that the data is normalized, and the nuclear main composition is analyzed; a sliding window is constructed in the main sub-quantity; the third moment and the fourth moment of the pixel in the sliding window are calculated and compared the moments with a preset value; the value is recorded after reading all over the main sub-quantity; all the main sub-quantity are processed; the maximal main sub-quantity is selected; the unusual point detecting on the selected main sub-quantity is conducted by the RX operator, and the detecting result is output. The present invention can extract and select the character of the unusual point object in the high spectrum image effectively, and can lower the false alarm rate, and can detect the unusual point normally under high background disturbance.

The invention discloses a resonant frequency detection method for magnetoelasticity sensor and a device thereof. The detection device comprises a detection coil, an exciting coil, a DC exciting branch circuit, an AC frequency sweeping branch circuit, and a signal detection and processing device, wherein the DC exciting branch circuit is connected with the AC frequency sweeping branch circuit in series and then connected to the exciting coil, and the signal input end of the signal detection and processing device is connected with the output end of the detection coil. The detection coil structure in the detection device is symmetric reverse-winding uncoupling combined detection coil structure, i.e. two single coils in the combined detection coil have the same turn number and reverse winding direction, and are wrapped with the exciting coil, so that the positions of the two reverse-winded coils in the exciting magnetic field are symmetric. The detection device and method of the invention can accurately measure the resonant frequency of the magnetoelasticity sensor in different media.

The invention provides an automatic segmentation method for the pre-background of a lepidopteron image based on a full convolutional neural network (CNN). A full convolutional network applied to pixel-level classified prediction is built via a fine tuned and pre-trained CNN model. Before the network is training, data enhancement is performed on an insect image data set to meet the requirement of deep neural network training on a sample quantity. Outputs of the different convolutional layers are fused, and a network model applicable to segmenting the pre-background of the lepidopteron image isacquired by exploration. Edge details are refined by further using a condition random field (CRF) based on an initial segmentation result of the CNN, and the maximum outline of the foreground is extracted and filled to remove noise interference existing in an output result of the network model and cavities in the foreground. According to the method, the pretreatment task of the insect image is completely automatic, and the lepidopteron species automatic recognition efficiency can be significantly improved.

The invention discloses a water quality heavy metal multi-parameter online monitoring instrument which comprises an online sample digestion system, an online sample detection system and a measurement and control system. The water quality heavy metal multi-parameter online monitoring instrument disclosed by the invention can realize automatic online monitoring of content of heavy metals such as mercury, lead, cadmium, chromium and arsenic in water, can be widely applied to the fields of surface water quality pollution monitoring, seawater pollution monitoring, up-to-standard release monitoring of industrial wastewater treatment and municipal sewage treatment and the like and has the advantages of high functional expansibility, wide measurement range and low cost.

The invention discloses an application of metalloporphyrin framework encapsulated carbon quantum dots in the detection of copper ions. When the copper ions are detected by fluorometry, the metalloporphyrin framework encapsulated carbon quantum dots are used as a ratio type fluorescent probe. The addition of Cu<2+> causes the weakening of the fluorescence intensity of a metalloporphyrin framework and makes the fluorescence intensity of the carbon quantum dots not change, and the addition of other metal ions does not cause Cu<2+> similar fluorescence change, so the highly-selective fluorescent quantitative detection of the copper ions is realized, and the influences of the concentration and the environment in existing fluorescent probes are overcome.

The invention discloses an anti-vibration machine vision detection method of foreign matters in a liquid medicine. The detection method comprises the following steps of: step 1, obtaining images; step 2, matching the images: carrying out matching pre-processing on adjacent frame gray level images by utilizing a relevant sub-pixel image matching algorithm based on a hill-climbing search phase; step 3: carrying out inter-frame difference: carrying out difference on the matched adjacent frame gray level images to remove static background information and keep movable foreign matter information; step 4: detecting a target: carrying out binary segmentation and connected domain search on the images obtained by the inter-frame difference, calculating area of each foreign matter area and finding out the largest foreign matter area; and step 5, judging a decision: judging whether an existing product is an inferior-quality product by judging whether the largest foreign matter area exceeds a standard, and finishing the detection process. The anti-vibration machine vision detection method of the foreign matters in the liquid medicine, disclosed by the invention, has the advantages of capability of effectively eliminating the vibration interference of an electromechanical system, strong robustness and capability of greatly reducing designing and manufacturing costs of detection equipment ofthe foreign matters in the liquid medicine.

The invention relates to a laser desorption ionization mass spectra, in particular to application of diamond-like carbon film used as matrix in the laser desorption ionization mass spectra; the surface of the diamond-like carbon film is used as an assisting matrix of laser desorption ionization flying time mass spectra, the diamond-like carbon film has high desorption ionization efficiency to compounds with low molecular weight and can not generate background interference of matrix molecules, so as to realize high-efficiency and rapid detection to the compounds with low molecular weight; the laser desorption ionization flying time mass spectra which adopts the diamond-like carbon film as the assisting matrix is a new desorption ionization method without adding any organic matrix, so as toprovide a general, high-efficiency and rapid detection method for the compounds with low molecular weight such as natural products and ecological metabolites, promote the development of analysis of compounds with low molecular weight by the laser desorption ionization flying time mass spectra technology and find new field for the application of the diamond-like carbon film material.

The invention relates to a data enhancement pedestrian re-identification method based on a generative adversarial network model. The method comprises the following steps: segmenting a mask image of apedestrian in an image by using a Mask-RCNN image segmentation algorithm; training an end-to-end improved star-shaped generative adversarial network in combination with the mask image and the manuallylabeled pedestrian attributes, and generating false training images under any number of cameras from the real pedestrian image under one camera; generating false training images of all camera domainscorresponding to all real images by using the trained improved star generative adversarial network; and sending the real image and the false training image into a pedestrian re-identification model,calculating the distance between the pedestrian images, and completing a pedestrian re-identification function.using Mask-to perform pedestrian re-identification; segmenting a mask image of a pedestrian in the image by using an RCNN image segmentation algorithm; training an end-to-end improved star-shaped generative adversarial network in combination with the mask image and manually labeled pedestrian attributes, and generating false training images under any number of cameras from a real pedestrian image under one camera; using the trained improved star-shaped generative adversarial network to generate false training images of all camera domains corresponding to all real images; and sending the real image and the false training image into a pedestrian re-identification model, calculatingthe distance between the pedestrian images, and completing the pedestrian re-identification function. The method is reasonable in design, more training samples are generated through the generative adversarial network, meanwhile, the generated image background can effectively represent the real scene under the corresponding camera, the robustness and the judgment capability of the pedestrian re-identification model are effectively improved, and the accuracy of pedestrian re-identification is effectively improved.

The invention belongs to the technical field of analysis and detection, and discloses a fluorescence probe for detecting beta-galactosidase as well as a preparation method and application of the fluorescence probe. The substrate of the fluorescence probe is 2-(2-(dimethylamino)ethyl-6-(((3R,4S,5R,6R)-3,4,5-trihydroxy-6-(hydroxymethyl)tetrahydro-2H-pyran-2-yl)oxidation)-1H-benzisoquinoline diketone, and has a structural formula shown in the description. The fluorescence probe has the advantages that naphthalic anhydride is taken as a fluorophore of the substrate of the fluorescence probe, and lined with a galactose group which can respond to the specificity of beta-galactosidase; the ratio of double fluorescence signals being I554 / I445 is taken as a detection signal, and thus the detection accuracy is higher; under the condition of 1U / L enzyme concentration, obvious fluorescence alteration can be detected, and the lower detection limit is 0.35 U / L; the fluorescence probe is excited (418 nm) in a visible light region, so that the background interference can be eliminated to achieve a better detection effect.

The present invention relates to medium materials for MALDI, and in particular relates to application of submicron carbon oxide balls as a medium in MALDI-MS. In the present invention, the submicron carbon oxide balls are used as the MALDI-MS medium, a defect of medium interference when a conventional organic medium is used is solved, the accurate analysis of various compounds with molecular weight less than 1000 Dalton, such as amino acid amino acids, lipid compounds, polypeptides, oligosaccharides and drugs can be fast and efficiently achieved. The submicron carbon oxide ball medium related in the present invention can be prepared by using a simple and controllable method, is regular in morphology, uniform in size, low in cost, and good in water solubility, and overcomes a problem that the dimension is controlled difficultly so as to influence stability of mass spectrometric detection when carbon-based materials such as carbon nano tubes and graphene are used as the MALDI medium, thus being an optimized medium capable of assisting MALDI-MS to achieve high-sensitivity and high-throughput analysis on the compounds within a low molecular weight area.

The invention provides a method for determining microcystic toxins concentration and is based on that molybdenum disulfide quantum dots can interact with gold nanoparticles through the internal filtration effect. Microcystic toxins can affect the degree of aggregation of aptamer-modified gold nanoparticles in the high salt solution, and detection of the microcystic toxins concentration in the water environment is achieved. The method is advantaged in that upconversion fluorescence of the molybdenum disulfide used can effectively avoid interference of background matrix fluorescence in the waterenvironment, optical stability is good, the detection result is accurate and reliable, cost is low, operation is simple, environmental protection is realized, detection specificity is good, influenceof other common metal ions or anions on determination of the microcystic toxins is avoided, the anti-interference ability is strong, the result is sensitive and reliable, the detection limit is low,the response speed is fast, the entire reaction process takes only 20 minutes, detection efficiency is greatly improved, good stability is achieved, real-time online fast, specific detection can be realized, and the method can be used for simple and rapid detection of microcystic toxins in the actual water environment.

The invention relates to the fields of biomedical detection and biochemical analysis, and particularly discloses a novel fluorescence probe of surface-functionalized silicon dioxide fluorescence nanoparticles for intracellular / extracellular zinc ion detection. The probe is prepared by covalently connecting the aminoquinoline-grafted dendriform polyethyleneimine derivatives to the surface of the dipyridyl-ruthenium-embedded silicon dioxide fluorescence nanoparticles; and a ratio fluorescence method is utilized to perform intracellular / extracellular zinc ion detection. The method is an improvement on the basis of the zinc ion fluorescence detection method, and can maximally enhance the detection sensitivity.

The invention relates to the product defect automatic online detection technology and in particular relates to a detonator defect automatic optic inspection machine. A two-dimensional direction adjusting mechanism (6) arranged on a rack (1) is connected with an image acquisition system (2), a 0-degree four-side strip light source assembly (17) is connected under the image acquisition system (2) by an adjusting pull rod (9), a detonator mold base (19) provided with a detonator (20) is positioned under the 0-degree four-side strip light source assembly (17), a driving cylinder (11) arranged on the rack (1) is connected with a two-dimensional direction adjusting mechanism (13) and drives a detonator backlight plate (18) to move up and down by a guide rod (16), a locating sleeve hole (22) consistent with the detonator mold base (19) is formed on the detonator backlight plate (18) and sleeves the detonator (20), a communication and cylinder driving control circuit module (23) is arranged on the rack (1). The detonator defect automatic optic inspection machine has high detection efficiency, and stable and accurate detection result, ensures that the detection result is objective and reliable, is suitable for online matched use and detection of the detonator filling production line, and is a brand-new high-efficiency detonator defect foreign matter automatic optical image detection machine.

The invention discloses a white light-emitting up-conversion nanoparticle and a test paper strip based on the white light-emitting up-converting nanoparticle and capable of simultaneously detecting multi-component tumor markers. According to the present invention, the nanoparticle has a three-layer core-shell structure, and emits white light under the excitation of near infrared light; and after the antibodies of multi-component tumor markers to be detected are labeled with the nanoparticle, the obtained conjugates are immobilized on the labeling complex binding zone of the test paper strip, such that the multiple tumor markers can be simultaneously detected; and with the nanoparticle and the test paper strip, the sensitive detection on the multiple tumor markers by the single up-conversion nanoparticle is achieved, and the advantages of simple operation, good specificity, high qualitative and quantitative real-time detection and high sensitivity are provided.

The invention discloses a glass surface defect detection system and method. The detection system comprises a supporting table, a deep ultraviolet light source, and a dedicated camera, wherein the supporting table is used for supporting the glass to be inspected. The deep ultraviolet light source emits an ultraviolet light and obliquely illuminates the glass to be inspected on the supporting table.The dedicated camera is installed right above the supporting table and faces the surface of the supporting table, and is used for collecting the reflected deep ultraviolet light. The detection methodusing the above detection system comprises four steps: placing the glass to be inspected; irradiating the deep ultraviolet light; acquiring an image; and determining whether there is a defect on theglass to be inspected. The detection system and method provided by the invention can solve the shortages that a visible light source for detecting glass defects in the prior art can generate a large background interference, by using the deep ultraviolet light source to detect, the background interference can be eliminated, and the accuracy rate is effectively improved.

The invention relates to the field of video satellite target tracking, and discloses a video satellite target tracking method based on background knowledge enhancement, and the method comprises the steps: intercepting a video satellite target in a first frame image, and obtaining a target template image and a target search region image; constructing a trained double-branch Siamese network, and generating a target template feature map and a target search area feature map; constructing three cascaded RPN networks, and obtaining a screened high-confidence prediction box; and establishing a background identification module, inputting the high-confidence prediction box into the background identification module to obtain a target tracking result of the current frame image, training the background identification module on line, and updating the scene category of the background identification module. Background enhancement is carried out through background offline training, background trainingset updating, background online updating and the like, the method is suitable for different complex backgrounds, background interference is removed to the maximum extent, the problem that a target issubmerged in the background is solved, and the target is tracked more accurately.

The invention discloses a bolt missing detection method, device and equipment and a storage medium. The method comprises the steps of acquiring a to-be-detected image corresponding to a to-be-detectedbolt; identifying a bolt connection node plate in the to-be-detected image by adopting a first preset model as a region-of-interest image; performing image segmentation on a node plate in the region-of-interest image by adopting a second preset model, and obtaining a detection reference point based on an image segmentation result; correcting the region-of-interest image by adopting perspective transformation based on the detection reference point and a benchmark reference point of a preset reference image to obtain a corrected image; identifying bolt position information in the corrected image by adopting a third preset model; and based on the bolt position information in the preset reference image and the bolt position information in the corrected image, determining a bolt missing detection result of the to-be-detected bolt. The effects of accurately extracting the detection reference point, ensuring the correctness of the corrected image, improving the image recognition precision ofthe region of interest and improving the accuracy of bolt missing detection are achieved.

The invention discloses an edge detection-based wall body crack identification method. The method comprises the steps of 1, acquiring a wall body image; 2, preprocessing the image; 3, extracting a crack edge; 4, filtering background noises; and 5, identifying a wall body crack. The technical problems of poor disturbance term adaptability caused by difficult dirt removal and the like due to adoption of a simple set grayscale threshold for performing binarization, or large calculation amount, poor real-time property and the like due to application of complex algorithms such as a neural network algorithm and the like in a substation wall body crack detection method in the prior art are solved.

The invention provides a fluorescence-quenched colloidal gold chromatography test strip which comprises a bottom plate, and a sample pad, a combination pad, a nitrocellulose membrane and absorbent paper which are sequentially arranged on the bottom plate from left to right; the sample pad and the combination pad, the combination pad and the nitrocellulose membrane, and the nitrocellulose membraneand the absorbent paper are partially laminated; gold nanoparticles are arranged on the combination pad, and antibodies of a to-be-detected object are marked on the gold nanoparticles; an antigen anda detection T line of a dual-excitation dual-emission up-conversion nanoparticle labeled detection object are fixed on the nitrocellulose membrane; the up-conversion nanoparticles can emit red light or green light under the excitation of two different near-infrared lights at the same time. The invention also discloses a preparation method and a detection method of the kit. According to the chromatographic test strip and the detection method thereof provided by the invention, the result can be directly observed under the excitation of near-infrared light, the sensitive components are accuratelyand quantitatively detected, the background interference is eliminated, the background value is low, the signal is stable, and the sensitivity is greatly improved.

The invention discloses a pedestrian detection method, system and terminal for multi-modal information fusion in an image. The method comprises the steps of obtaining a corresponding top view according to an original head-up view; predicting a pedestrian target probability and a pedestrian target frame width for each point on the top view; obtaining a corresponding depth map through the head-up image, and estimating the height of a pedestrian target frame based on the depth; integrating the pedestrian target frame width and the pedestrian target frame height into a target candidate frame, anduniformly encoding the target candidate frame to a semantic space through an encoder to obtain a pedestrian target probability under a head-up view; and performing weighted average on the pedestrian target probabilities of the head-up view and the top view to obtain a final pedestrian target prediction result. According to the method, the pedestrian shielding problem under a head-up view angle iseffectively avoided, complex background interference is effectively removed, the method is suitable for variable situations in a real application environment, the detection robustness is enhanced, thefalse detection and missing detection probability is reduced, and the detection capability of pedestrian targets in video images is effectively improved.

The invention relates to a detection device and a detection method for a heavy metal migration amount of a nano filter tip, and belongs to the technical field of physical and chemical inspection of tobacco accessories. The device comprises an air inlet, air heating chambers, a ten-way valve, a filter tip holder, a smoke trapper and a smoke sucking machine; the ten-way valve comprises a plurality of rotatable air connection ports, a fixed air connection port and a numerical control turntable; the number of the rotatable air connection ports is same as the number of the air heating chambers; each rotatable air connection port is connected with one end of one air heating chamber. The other end of each air heating chamber is connected with the air inlet; the numerical control turntable is usedto control connection of each rotatable air connection port and the fixed air connection port; and the fixed air connection port, the filter tip holder, the smoke trapper and the smoke sucking machine are connected sequentially. The detection device has the advantages of simple structure, convenient operation and reducing of the influence of complex smoke, and can be used as a new analytical technique and method for heavy metal migration amount of the nano-filter tip.

The invention relates to an ultrasonic section identification method and system, computer equipment and a storage medium. The method comprises the following steps: acquiring a to-be-identified ultrasonic image; performing feature extraction on the to-be-identified ultrasonic image to obtain corresponding image features; according to the image features, identifying all foreground targets in the to-be-identified ultrasonic image, and obtaining positioning information of each foreground target contained in the to-be-identified ultrasonic image; and according to the positioning information and theimage features, identifying all section targets in the to-be-identified ultrasonic image, and obtaining first category information and position information of each section target contained in the to-be-identified ultrasonic image. By adopting the method, a plurality of section targets contained in a single ultrasonic image can be effectively identified.

The invention discloses a gasket size detection system and method based on machine vision, a processing terminal and a medium, and relates to the technical field of gasket quality detection and visual optical design. In order to solve the problems that in the prior art, unqualified products cannot be removed in real time, random angle detection cannot be carried out, the method cannot adapt to actual production, the labor intensity is increased, and the maintenance cost is high in the prior art, the number of products in an image is judged, affine transformation is carried out on the products at the specific positions, and background interference is removed; secondly, the to-be-detected gasket is positioned, and the measured position of the to-be-detected gasket is subjected to contour extraction; and finally, the actual size of the to-be-detected gasket is calculated according to the proportional relation between the view field range of the image acquisition equipment and the image pixels. The problem of noisy points caused by optical interference is avoided; according to camera size detection, the placement direction of the workpiece is random, detection can be carried out when the workpiece exists in the collected image, position and angle correction is carried out on the detected object through an algorithm, and feeding consistency is achieved.

The invention provides a fluorescence ratio type spectral analysis method for detecting carcinoembryonic antigen, and belongs to the technical field of optical analysis. The method provided by the invention comprises the main contents of the preparation of long afterglow nanorods ZGO: Sr NRs, the modification of surface carcinoembryonic antigen aptamers, and the preparation of ZGC@PDA nanomaterials. By using ZGC@PDA as both a reference signal and a quencher, and using a detection probe ZGO: Sr NRs that can specifically recognize carcinoembryonic antigens, a ratio-type spectroscopic analysis method that can be used to detect carcinoembryonic antigens is constructed. The analysis method has good accuracy and stability.

The invention discloses an image binarization method and device, an electronic device and a storage medium. The method comprises the following steps of firstly obtaining a plurality of color components of each pixel in a target image, wherein each color component of the plurality of color components corresponds to a preset interval; then determining a binarized pixel value of each pixel accordingto the plurality of color components and the preset interval; then converting the target image into a binary image according to the binarized pixel value. By adopting the embodiment of the invention,the background interference can be effectively removed, thereby improving the accuracy of image recognition.

The invention discloses a preparation method and a detection method of a TMAO biological negative sample and a biological standard substance. The method comprises the following steps: taking a biological sample, adding a biological enzyme to act on TMAO so as to reduce or eliminate the content of TMAO in the biological sample, and terminating the reaction between the biological enzyme and TMAO soas to obtain the biological negative sample. The method comprises the following steps: preparing a TMAO biological negative sample by using a biological enzyme, so that the endogenous TMAO backgroundinterference can be effectively eliminated, the reaction specificity is strong, the efficiency is high, the operation is convenient, and the economic benefit is high; by means of the scheme, endogenous TMAO background interference can be effectively eliminated, an enzyme reaction is ingeniously utilized, oxidative TMAO is reduced, metabolized and decomposed or groups are added to change structuralmodification, and therefore background interference caused by TMAO in a blank matrix is eliminated; and excessive bio-enzyme can be removed by subsequently inactivating or filtering the bio-enzyme, so that the bio-enzyme can be further applied to preparation of TMAO standard substances.

The invention provides a new nm zirconium oxide basis acetone sensitized material obtained by the following steps: mixing the 1 part by weight of zirconia nm power and 0.02-1.55 parts by weight of erbium; adding distilled water to stir into thin paste; intensively and uniformly mixing the mixture; drying at 110degree for 2 hours; heating at 550-650 degree at oxygen atmosphere for 2-3 hours; cooling; finally adequately grinding. The new nm zirconium oxide basis acetone sensitized material can be used for producing the sensor for detecting the acetone with excellent selectivity and sensitivity to the acetone gas with no interference of the circumstance coexisting gas such as ethanol, formaldehyde and the like, and the detection limit can be obtain 0.40ppm.

The invention provides a glass bottle defect detection system. A bearing unit, a defect detection unit, a judgment unit, an information processing unit, an adjusting and supporting unit and a controlunit are arranged in an inspection device. According to the defect detection device, optics and camera shooting analysis are combined; a sensor unit and the control unit are arranged, the convenienceof glass bottle defect detection work and the accuracy of defect detection can be effectively improved, and the yield of glass products is further improved; and the judgment unit and the information processing unit are arranged, the defect type of a glass bottle can be effectively analyzed, reference significance is provided for glass bottle production, and the problems that in the prior art, themanual detection effect is slow, the detection effect is poor, the detection cost is high, defect detection in glass bottle production is inaccurate, and production reference significance cannot be provided for a glass production process are solved.