130results about How to "Make up for the shortcomings" patented technology

The invention discloses an abnormity detection method based on a generative adversarial network. The abnormity detection method comprises the steps that S1: a semantic segmentation image data set is processed so as to obtain a data set meeting the original image and a data set of Mask images; S2: an abnormity detection generative adversarial network structure including a generative network and a discrimination network is established; S3: the data set meeting the original image and the data set of the Mask images are trained so as to obtain a generative adversarial network model of abnormity detection; and S4: the original image is inputted and the image of the specific abnormity type is obtained. Compared with the methods in the prior art, the abnormity detection method based on the generative adversarial network has the following advantages: 1. the abnormity part image acts as the objective of model training, and the abnormity part concrete locating information acts as the training objective in comparison with the conventional abnormity detection method so that the abnormity part of the image is enabled to be further visual; and 2. compared with the single integrated network structure of the conventional abnormity detection method, the generative adversarial network model is established so that the disadvantage that the image of the abnormity part cannot be outputted can be compensated.

The invention pertains to the technical field of gasochromic film, in particular to a method for improving gasochromism speed of the gasochromic film. The invention adopts sol-gel method for preparing the stable solution containing various metal oxides and the compound solution thereof; and various methods are employed for coating transparent and even gasochromic film on common glass or polished silicon wafer, and ultraviolet or high energy ray and high energy-particle mixing with catalyst are employed in the gasochromic film method so as to change the structure of film and enhance the coloring response speed of the gasochromism of the sol-gel technique, shorten coloring saturation time. Compared with the non treatment film, the gasochromism efficiency are greatly improved; and the performance is the same as the gasochromism produced by magnetron sputtering method, while, cost is greatly reduced; and the produced film by the invention also has the same advantages as the electrochromic film. The invention makes up the shortages of gasochromism smart window made by sol-gel preparation method, thus the smart window technique can be more mature.

The invention discloses a semantic mapping and positioning method based on priori laser point cloud and depth map fusion. The method comprises the steps of S1, collecting priori laser point cloud data; S2, acquiring a depth image and an RGB image, generating RGB-D point cloud based on the depth image, and initializing and registering priori laser point cloud and RGB-D point cloud; S3, camera poseconstraints are provided by the registered prior laser point cloud to perform camera pose correction; S4, a three-dimensional geometric point cloud map is created by adopting a front and back window optimization method; S5, geometric increment segmentation is carried out on the three-dimensional geometric point cloud map, object recognition and semantic segmentation are carried out on the RGB image, geometric increment segmentation and semantic segmentation results are fused, and a 3D geometric segmentation map of semantic enhanced geometric segmentation is obtained; and S6, semantic association and segmentation probability allocation updating is carried out on the object to complete construction of a semantic map. Accumulated errors of large-scale indoor mapping and positioning can be effectively eliminated, and precision and real-time performance are high.

The invention discloses a mildew-preventing antibacterial hydrophobic coating and belongs to the technical field of coatings. According to the mildew-preventing antibacterial hydrophobic coating disclosed by the invention, a coarse surface shape can be formed on a substrate by adding polymeric particles, so that the hydrophobic property of the coating is greatly improved; and meanwhile, by taking nano-zinc oxide modified chitosan as an antibacterial agent, the mildew-preventing antibacterial hydrophobic coating disclosed by the invention has a good antibacterial effect and overcomes the defects that silver oxide which serves as the antibacterial agent is active in property, is easily oxidized in air to lead to discolor and is high in cost.

The invention provides a process for preparing substitute natural gas through sulphur-tolerant methanation of coal synthesis gas. The process is characterized in that the synthesis gas enters sulphur-tolerant methanation reactors I and II in sequence, and the mixed gas of the synthesis gas and water vapor enters a rectisol system so as to remove the impurities such as CO2, H2S and the like in the gas after undergoing sulphur-tolerant shift and sulphur-tolerant methanation reaction on a molybdenum-based bifunctional catalyst and then undergoes methanation reaction in methanation reactors I and II in sequence under the action of Ni-based methanation catalysts, thus obtaining the natural gas product. The process has the advantages of simple process flow, small equipment investment, low comprehensive energy consumption and excellent natural gas products.

The invention provides an indoor typical scene matching and positioning method based on neural network. The method comprises the steps of step 1, establishing a standard typical scene positioning image library at a server side; step 2, adopting a Siamese deep neural network model, through training of a large amount of data, enabling the neural network to learn from the data to judge similarity measurement; step 3, outputting a feature vector by the deep neural network, calculating the similarity with a standard typical scene image library by utilizing the feature vector, judging the matching degree of the typical scene according to the similarity, and evaluating the quality of the model; and step 4, loading the trained model into a server, obtaining video data, sending the video data intothe trained deep neural network in the server, calculating similarity, and judging the current position. The method has the advantages of being high in training efficiency, high in convergence, high in modeling precision, good in matching effect, capable of meeting complex environments and the like, and matching and positioning of equipment in online indoor typical scenes can be accurately and efficiently achieved.

The invention discloses a cultivated land change detection method based on depth learning, which is applied to cultivated land census task and has higher convenience and efficiency than manual on-the-spot investigation, at the same time, the remote sensing detection result has higher authenticity compared with the results reported by various levels of departments. Traditional remote sensing interpretation is mainly in the form of artificial visual, is high in human resources cost, but also cannot meet the growing amount of remote sensing data interpretation needs. The invention firstly accurately matches the remote sensing images of the same area taken at different times, then extracts the cultivated land area by adopting the full convolution network, then directly obtains the preliminarycultivated land change result by using the difference method, and finally obtains the final change detection result by adopting the morphological methods such as corrosion and expansion. The inventionalso discloses a method for extracting the cultivated land area from the remote sensing images. The invention adopts the method of depth learning semantic segmentation network to realize automatic extraction and change detection of cultivated land resources, which has higher detection accuracy than the method of the invention in the prior art.

The invention relates to a green and environment-friendly fertility fertilizer capable of thoroughly improving soil, activating and repairing soil, protecting soil from fertilizer damage, completely restoring the soil, resisting fertilizer damage and preventing pollution, enduringly and strongly activating soil, completely getting rid of soil hardening, enriching vegan nutrition and continuously increasing both production and income. By adopting the methods of conditioning of soil and physiological regulation of crops, synthesis of biological nutrition and recombination of (mother) nutrient components, inorganic and organic combination and the addition of (sub-element) microelement supplements, the green and environment-friendly fertility fertilizer is scientific, targeted and flexible and makes up for the shortcomings that the modern fertilizer has nutrient deficiency or excess easily caused by the fixed ratio of nutrients. The green and environment-friendly fertility fertilizer has the advantages that nitrogen is supplemented if nitrogen is deficient, phosphorus is supplemented if phosphorus is deficient, potassium is supplemented if potassium is deficient, elements are supplemented if elements are deficient, the fertilizer damage is effectively inhibited, the properties of soil are improved, the fertility of soil is improved, the deterioration of soil and environmental pollution are avoided, and thus the green and environment-friendly fertility fertilizer is named as'fertility fertilizer and fertilizer in fertilizer' and is extremely perfect.

The invention relates to a technology for preparing natural gas through sulfur resistant methanation by coal-prepared synthesis gases, which is characterized in that coal-prepared synthesis gases is subjected to secondary sulfur resistant transformation and methanation reaction on a molybdenum-based bifunctional catalyst, then is eluted by methanol at low temperature to remove acidic gas, and finally, the Ni-based methanation catalyst is used for performing secondary methanation reaction on unreacted CO and H2 to obtain the excellent natural gas product. The technology for preparing natural gas has the advantages of simple process flow, less equipment investment, low comprehensive energy consumption and excellent natural gas product.

The invention discloses a granulated feed for beef, containing corn straws, dry groundnut stems, corns, lees, bean pulps, common salt and microelements, with a ratio of 40-50:10-20:5-15:10-20:5-10:0.5-1 according to parts by weight in sequence. The granulated feed has the advantages of wide raw material source, high digestion and utilization rate, balanced nutrition and easy storage.

The invention provides a railway barrier detection and alarm radar display and control system and a railway barrier determining and alarm automatic processing method thereof. The system mainly comprises a data communication unit, a database operation unit, a target display unit, a target processing unit, a radar parameter setting unit, a barrier alarm unit and a state management unit. Whether a barrier is on a railway line is detected by comparing radar echo characteristics acquired in real time with radar echo characteristics which are pre-acquired and stored in a barrier-free condition, and alarm information is issued in the case of the barrier. A human-computer interaction system can set the operating parameters of a radar. According to the railway barrier determining and alarm automatic processing method of the railway barrier detection and alarm radar display and control system, barrier determining and alarm of the alarm detection radar are more simple and visual.

The invention relates to the technical field of water-based ink and discloses a preparation method of water-based ink, containing an Ag-carrying sepiolite antibacterial material, for tipping paper. The preparation method comprises the following steps: (1) preparing Ag-carrying sepiolite; (2) compounding silane coupling agent modified Ag-carrying sepiolite with a water-based resin liquid A and a water-based resin liquid B and then compounding a mixed resin liquid with deionized water, ethanol, tea saponin, an organic pigment, a defoaming agent, a levelling agent and a buffer agent, thus preparing the water-based ink, containing the Ag-carrying sepiolite antibacterial material, for tipping paper. The preparation method has the beneficial effects that containing various antibacterial components, such as nano-silver, titanium dioxide and tea saponin, the water-based ink prepared by the preparation method has strong bactericidal broad-spectrum property; the bactericidal components have synergistic effects, so that the bactericidal effects are greatly enhanced; the dispersibility of the bactericidal components in the water-based ink is quite outstanding.

The invention discloses a submerged tubular membrane element, and relates to the technical field of water treatment microfiltration and ultrafiltration. The submerged tubular membrane element comprises a membrane producing water collecting box, a water catchment connector, membrane tube connectors, a tubular membrane array and membrane tubes, wherein the water catchment connector is arranged on one side of the membrane producing water collecting box, multiple membrane tube connectors for connecting the membrane tubes are arranged on the other side of the membrane producing water collecting box, an interface component array is formed by the multiple membrane tube connectors, the membrane producing water collecting box is connected with the tubular membrane array formed by the multiple membrane tubes through the interface component array, an aeration device is arranged below the tubular membrane element, and mixed liquor is driven to form circling-type cross flow by aeration to enhance membrane cleaning. The submerged tubular membrane element disclosed by the invention is simple in structure and reasonable in design, achieves uniform distribution of producing water flux of the membrane tubes, can achieve thorough back washing, and is low in cost, high in flux, low in energy consumption and not easy to cause pollution, thus having a high market popularization value.

The invention discloses a phosphorus nitrogen hybridization synergistic intumescent flame retardant and a preparation method thereof. The invention provides a synergistic intumescent flame retardant which is formed by an in-situ cross-linking/grafting reaction, has a phosphorus nitrogen hybridization structure and integrates an acid source, a gas source and a carbon source into one whole body. Theretardant can achieve an excellent flame-retarding effect by the physical and chemical flame-retarding process, is low in water absorption and high in stability and can be widely applied to plasticssuch as ABS (Acrylonitrile Butadiene StyreneAcrylonitrile Butadiene Styrene) and the like and an organic foam product.

The invention relates to the technical field of machine tool fixtures, in particular to a machine tool fixture for clamping and positioning multiple workpieces with various materials and external physical characteristics. The multifunctional machine tool fixture comprises an upper sliding guide rail, a lower sliding guide rail, a clamping mechanism, a fastening bolt, an exchangeable pressing head, an internal thread pin, a first pin and a second pin. The multifunctional machine tool fixture is characterized in that the clamping mechanism is connected with the upper sliding guide rail through the fastening bolt, the internal thread pin and the first pin, and is connected with the exchangeable pressing head through the second pin; the upper sliding guide rail is connected with the lower sliding guide rail; and an opening groove is formed in one end of the lower sliding guide rail. The exchangeable pressing head comprises a finished planar exchangeable pressing head, a cylindrical surface exchangeable pressing head and a roughed planar exchangeable pressing head. The second pin can be pulled out or inserted in so as to replace the exchangeable pressing head of a corresponding workpiece. The multifunctional machine tool fixture makes up the defects of a general fixture and a special fixture, is conveniently used, is smooth in appearance line, and is affordable.

The invention discloses a local pavement adhesion coefficient estimation method based on image segmentation, and the method comprises the steps: 1, carrying out the offline pre-training of an image segmentation network, and specifically comprises the steps: a, collecting pavement images of different weather conditions through CARLA software, b, carrying out the local marking of the collected pavement images of different weather conditions, and forming a data set of local pavement adhesion coefficient estimation, c, building a deep learning algorithm network model of image segmentation by usinga deep learning means; and d, carrying out end-to-end training on a deep learning algorithm network framework of image segmentation by using a data set estimated by a local pavement adhesion coefficient. Step 2, acquiring a real-time road surface image, and estimating the local adhesion coefficient of the road surface in real time, specifically comprising: a, acquiring the real-time road surfaceimage by using a vehicle-mounted camera, b, classifying the real-time acquired image by using a pre-trained image segmentation network and positioning different categories to form a real-time road condition map, and c, performing local road surface adhesion coefficient estimation on the real-time road condition map according to the road surface type.

The invention relates to a fingerprint identification-based handle lock mechanism. The fingerprint identification-based handle lock mechanism comprises a face cover, a PCB (printed circuit board) arranged on the face cover, a fingerprint collecting module arranged on the PCB, and a movable handle case, wherein one end of the movable handle case is fixedly connected to the face cover, the other endof the movable handle case is provided with a fixing case fixedly connected with a lock hole, and the fixing case is in clearance fit with the movable handle case; the other end of the movable handlecase is fixedly connected with a lock plate; a motor is arranged inside the movable handle case; the inside of the other end of the movable handle case is provided with a locking piece, and the output shaft of the motor is connected with the locking piece. The fingerprint identification-based handle lock mechanism integrates fingerprint collection and fingerprint identification onto the PCB, anda locking clutch mechanism, a driving control device and the PCB inside the movable handle case, thereby being small in size and simple in structure. The fingerprint identification-based handle lock mechanism can replace the lock cylinder of a connecting rod lock, a cam lock, a blocking lock or the like, thereby being simple in installation and operation, increasing convenience of replacement of fingerprint lock and saving the cost. The fingerprint identification-based handle lock mechanism is simple in structure, saves costs and meanwhile achieves high reliability.

The invention relates to a motion mathematical model of a cycloidal propeller driven ship for a marine simulator. The motion mathematical model is realized by the following steps of: obtaining the equivalent pitch ratio of a VSP propeller as well as the percentages of resolutions of the total thrust generated by the VSP propeller in the vertical direction and the lateral direction of a ship body by operating hardware devices (X-Pitch and Y-Pitch), working out the total thrust with the disclosed first-quadrant atlas of the VSP propeller, and obtaining the resolutions (torque) in the vertical direction, lateral direction and yaw by multiplying the percentage of the resolutions by the total thrust; and adding the obtained thrust (torque) as external force to the ship motion equation, obtaining the posture of the VSP ship motion with a forth-ordered Runge-Kutta integration method, and offering the posture to other modules of the marine simulator system. The established VSP-propeller ship motion mathematical model is used in the marine simulator and combined with the hardware devices, so that the human-machine interaction is realized and the ship model databases in the system are diversified.

The invention discloses a method for preparing rare earth-doped quartz glass microstructure optical fibers by a laser melting technology. The method comprises the following steps of powder preparation, melting, doped rod treatment and optical fiber precast rod preparation, and rare earth-doped quartz glass microstructure optical fiber drawing. The invention discloses a use of the uniformly doped quartz glass rod in an all solid state laser and a microstructure optical fiber. The invention discloses a laser melting system. The laser melting system comprises a powder feeding device, a laser source, a mother rod, a clamp, a motor and a guide rail. The prepared rare earth-doped powder and the quartz glass microstructure optical fibers realize uniform doping of a plurality of rare earth ions. The prepared doped microstructure optical fibers have a rare earth ion effective-doping concentration more than 10000ppm.

The invention provides a growth incentive book recommendation method, which comprises the following steps of 1, collecting book data, and performing book label labeling; 2, performing feature extraction on the reading data of the book to form three-dimensional features; wherein the three-dimensional features comprise interest features, mood features and character features; and 3, obtaining book information read by a reader, wherein the book information comprises three-dimensional features and book labels; calculating the distance between the book read by the reader and the book in the database, and obtaining a forward recommended book according to a k-nearest neighbor calculation result; and calculating reverse labels of the books by counting the book labels of the books read by the readers to generate reverse recommended books. The invention further provides a growth incentive book recommendation system.