54results about How to "Get good results" patented technology

The invention relates to a red-tide living thing image automatic identifying device and method, provided with a LED, a light source lens, a mobile room, an objective, a depth of focus lens and a camera; the LED is at the focus of the light source lens, the mobile room is between the light source and the objective, and the depth of focus lens is between the objective and the camera. The identifying method: obtaining image, preprocessing, detecting object, separating image, obtaining image feature, classifying and identifying object, and outputting result. It adopts a far-focus objective, adds an operating space between the mobile room and the objective, giving a development space to the mobile room, and increases the application range of the apparatus. It can automatically analyze images, heightening automation degree of the apparatus, and meets the request for field monitoring red-tide living things, hopeful to implement real-time monitoring the generating and disappearing course of the red tide, and on this basis, provides a precaution to the danger degree of the red tide.

The invention discloses a polarized image defogging method combined with a dark channel prior principle. The method comprises the following steps of: firstly, obtaining a polarized image by a polarization imaging technology, thereby obtaining linear polarization stokes vectors of a scene; secondly, selecting a region from a total light intensity image of the scene, and estimating an atmosphere diffusion light polarization degree, a polarization angle and an atmosphere diffusion light intensity of the infinite area based on the region; then calculating the atmosphere diffusion light intensity of each pixel point according to the estimated atmosphere diffusion light polarization degree and the polarization angle, and based on a physical model of polarized defogging imaging, combining the atmosphere diffusion light intensity of each pixel point with the atmosphere diffusion light intensity of the infinite area to calculate the target light intensity after the defogging; and finally, carrying out image quality evaluation by utilizing a gray scale entropy function, automatically optimizing the bias coefficient of the atmosphere diffusion light intensity of the infinite area, and obtaining the defogged image. The method provided by the invention can accurately and automatically estimate the information of atmosphere diffusion light, the reliability is higher, the calculation speed is high, and the contrast of the defogged image is high.

The invention discloses frozen and cooked noodles capable of being heated through microwave, which belong to the technical field of frozen instant foods. The frozen and cooked noodles are prepared mainly by the following raw materials in part by weight: 40-50 parts of water, 115-124 parts of flour, 12-17 parts of cassava modified starch, 4-7 parts of cassava starch, 1-3 parts of duck egg pulp, 2-4 parts of table salt, 0.3-0.7 part of dietary alkali and 0.5-0.8 part of composite phosphate. The invention improves the quality of the frozen and cooked noodles by adopting a reasonable frozen and cooked noodle prescription, contains 70+/-5 percent of water and can be cooked by adopting microwave heating or cooked by heating with a common stove. The cooked noodles cooked by adopting microwave heating have higher quality, smooth edible taste, even color and complete shapes and can not generate the phenomenon of breakage or soup mixture.

The invention discloses an image style migration method, system and device and a storage medium. The method comprises the following steps: obtaining a content picture; inputting the content picture into a pre-trained image style migration model for style migration processing, and outputting a target picture which has a specific style and retains the original content; enabling the image conversionnetwork and the discrimination network to form a generative adversarial network (GAN), and carrying out alternate updating in the model training process. According to the method, the network is continuously updated and optimized by sensing the adversarial loss function until the loss is minimized, and the image style migration model with a better effect is obtained, so that the output picture closer to the content picture and the style picture can be obtained, the problem of picture background distortion is effectively avoided, and the method, system and device can be widely applied to the field of data image processing.

The invention provides environment-friendly anticorrosive paint and a preparation method thereof. The environment-friendly anticorrosive paint comprises a main agent and a curing agent. The main agent is prepared from, by weight of 100 parts, 39-50 parts of low-viscosity solvent-free epoxy resin, 4-8 parts of diluents, 5-15 parts of toughening resin, 0.5-15 parts of toner, 10-40 parts of filler, 5-15 parts of low-viscosity epoxy ester resin and 0.5-1.5 parts of a thixotropic agent. The curing agent is modified amine resin.

The invention discloses a wear-resisting concrete terrace overall molding construction method, and the method comprises the steps as follows: casting and compacting the base layer concrete, purifying syrup for the surface, spreading the wear-resisting material for the first time, grinding via machine, spreading the wear-resisting material for the second time, grinding via machine, polishing, curing the wear-resisting concrete and cutting. The wear-resisting concrete terrace prepared by the method has integrated surface layer and concrete base layer for eliminating the crack and empty drum quality defects caused by bad combination.

The invention relates to a shallow sea underwater image enhancement method based on relative global histogram extension. The shallow sea underwater image enhancement method comprises input image module, an image RGB histogram distribution analysis module, a relative global histogram extension module, a global extension module and an output image module. The advantages are that the method is low in image enhancement computation complexity, good in effect, good in robustness and suitable for various shallow sea underwater images of underwater plants, shallow sea personnel and seabed rocks and the like; the method can calculate parameters of a relative extension range and automatically select the extension range, and can accurately position a range to be extended; through a color correction method, color loss of global extension of S and V is carried out through an HSV color space; and the method can enhance contrast, saturation and visibility of each underwater image.

The invention discloses thermostability transformation of maleic acid cis-trans isomerase and application thereof, and belongs to the field of enzyme engineering. The maleic acid cis-trans isomerase gene maiA from serratia marcescens is subjected to thermostability transformation by a semi-rational design method. Compared with the wild maleic acid cis-trans isomerase, the obtained mutant has the advantages that the half-life period of the mutant G27A-G171A at 55 DEG C is 2.9 times of that of the wild type, and the enzyme activity is 1.96 times of that of the wild type; the half-life period of the mutant G27A-K104R is 4.18 times of the wild type at 55 DEG C, and the enzyme activity is 1.59 times of that of the wild type.

The invention belongs to the field of image classification, and discloses an image multi-label classification method, system and device, and a readable storage medium. The method comprises the steps:S1, obtaining original images corresponding to multiple labels to be classified, and generating a sample set; s2, obtaining an initial deep convolutional neural network for image multi-label classification, and adding a gate activation function layer to obtain a deep convolutional neural network; s3, iteratively training a deep convolutional neural network through the sample set to obtain a saliency map corresponding to each label; s4, selecting a saliency map corresponding to a label with a preset requirement, and extracting a target from the corresponding original image through the saliencymap; s5, setting a to-be-generated image, and migrating the target area to the to-be-generated image to obtain a target image; s6, adding the target image into the sample set, iterating S3 to S5 for preset times, and performing image multi-label classification of the input image through the final deep convolutional neural network. According to the method, the precision and the accuracy of image multi-label classification are effectively improved.

The invention relates to a sand-in-water coating and a preparation method thereof. Raw materials of the sand-in-water coating comprise a dispersion phase and a continuous phase; the dispersion phase involves, by weight, 0.5%-1% of first protection glue powder, 20%-30% of a pure acrylic emulsion, 0.4%-0.7% of hollow microspheres, 0.1%-0.5% of a first alkali swelling thickening agent, 25%-35% of quartz sand, 3%-10% of a first assistant and 25%-40% of water; the continuous phase involves, by weight, 2%-4% of second protection glue powder, 30%-50% of a silicone acrylic emulsion, 0.3%-0.6% of a second alkali swelling thickening agent, 3%-10% of a second assistant and 50%-65% of water; the proportion of the dispersion phase is 1 g/ml-1.06 g/ml; the viscosity of the continuous phase is 40 ku-55 ku. The sand-in-water coating prepared by adopting the two-phase system is low in proportion of the dispersion phase, in the processes of prilling production, mixing and storage of finished products, color points are in a suspension state, and the sedimentation phenomenon cannot easily occur. The viscosity of the coating is low, the color point separation effect is good, the production efficiency is high, the water-whitening resistance performance of the coating is good, and the viscosity is stable.

The invention discloses an image gesture identification method and device, an image color white balance adjustment method and device, an exposure adjustment method and device. The image gesture identification method comprises the steps of detecting a face area in an image; determining a white balance parameter for adjusting a color of the face area into a skin color; carrying out white balance adjustment of the color on the image according to the determined white balance parameter; and identifying a gesture in the image after the white balance adjustment of the color. Through application of the methods and the devices, the problem that the image color distortion is serious is solved without manually carrying out the color correction, so the accuracy of identifying the gesture in the imageis improved.

The invention discloses a video fine structuring method based on multi-feature fusion, and belongs to the field of multimedia content processing. The video fine structuring method comprises the following steps of firstly, carrying out video shot boundary initial detection through a color feature HSV; then, establishing a fusion feature vector under the premise of equal-size rectangular partitioning, and introducing an adaptive threshold value to carry out the candidate lens boundary reinspection; clustering the shot sets through a clustering algorithm to carry out scene classification; extracting a key frame in the scene by comparing the difference between the frame and the average feature of the scene; and finally, storing the structured information, such as the shot set, the scene set, the key frame set, etc., into a content server, so that a user can conveniently retrieve information in a database. According to the method, the process time consumption is reduced at design, the selected characteristics conform to the human eye attention characteristics, the local information and the global information are considered, and the accuracy is improved.

The invention discloses a primer group and method for identifying dendrobium huoshanense and a dendrobium officinale product based on Real-time ARMS-qPCR. The primer group comprises two pairs of pre-amplification primers and two pairs of ARMS specific primers. The identifying method comprises the steps of DNA extraction of a to-be-detected sample, pre-amplification and amplification of the Real-time ARMS-qPCR. The primer group for identifying the dendrobium huoshanense based on the Real-time ARMS-qPCR can be amplified effectively, and according to the primer group, the method that Real-time Quantitative PCR is combined based on the ARMS technology is adopted so that the dendrobium huoshanense and the dendrobium officinale product of the dendrobium huoshanense can be identified rapidly and precisely, and the built method for identifying the dendrobium huoshanense and the dendrobium officinale product of the dendrobium huoshanense is accurate, sensitive, rapid, stable and repeatable.

The invention relates to an ibuprofen suspension and a preparation method thereof. The ibuprofen suspension is a Chinese and western compound preparation, a primary medicine of the ibuprofen suspension comprises ibuprofen, the primary medicine is added to auxiliary medicines of the ibuprofen suspension to make the primary medicine suspend in order to prepare a suspension, and the suspension is disinfected and packaged to obtain the ibuprofen suspension. The ibuprofen suspension can simultaneously treat the principle and subordinate symptoms of infantile recurrent respiratory tract infection and infantile fevers, and has the advantages of non-toxicity and innocuousness to human bodies, no side effects, low price, simple preparation and wide application prospect.

The invention belongs to the field of monitoring and control of water environments, and discloses a method for optimizing layout of a lake, reservoir and river basin water quality monitoring cross section based on comprehensive layering clustering. The method comprises the following steps of firstly, according to the originally collected water quality monitoring data, establishing a standard matrix, and calculating the similarity between the data variables by five distance algorithms, so as to obtain a close and distant relationship of the variables; automatically clustering a data set by four connecting algorithms; using a weighted Pearson correlation coefficient as a judging criterion to select the optimal combination for layering clustering; finally, generating a corresponding optimal clustering tree matrix to optimize and arrange a cross section monitoring station, so as to obtain the monitoring data for reflecting the whole water quality of the water domain, and improve the monitoring quality of the water environment. The method has the advantages that the convenience, simplicity and rapidity are realized, the result is visual, and the feasibility is higher.

The invention discloses a wave blocking device for hydraulic engineering construction, and relates to the field of hydraulic engineering construction.The wave blocking device comprises a floating plate; a back plate is fixedly connected with one side of the top of the floating plate; an underwater floating mounting mechanism is arranged at the bottom end of the floating plate; a wave blocking mainbaffle is arranged above the floating plate; a driving plate penetrates through the back plate; a power generation driving device matched with the driving plate is arranged on the back part of the back plate; the two ends of the power generation driving device are connected with rotating columns; the rotating columns are connected with a reverse water drainage impact device; the top end of the back plate is fixedly connected with an ejector rod; and the top end of the ejector rod is fixedly connected with a warning lamp. According to the wave blocking device for hydraulic engineering construction, by setting the underwater floating mounting mechanism, the device can be fixed underwater and the position of the floating plate can be adjusted adaptively according to the level of the water level; water waves can be effectively blocked and eliminated by arranging the wave blocking main baffle and a wave blocking branch baffle, so that water wave power generation and reverse water deliverycan be achieved to counteract the sea waves; and the impact force of the water waves can be observed through the brightness of the warning lamp, water wave information can be conveniently obtained, and the use effect is good.

The invention relates to a medicine composition for treating recurrent respiratory tract infection of children. The medicine composition is prepared from the following materials in parts by weight: 5 parts of ibuprofen, 20 parts of radix bupleuri, 20 parts of coptis chinensis, 18 parts of Chinese yam, 15 parts of bamboo leaf, 15 parts of radix zanthoxyli, 15 parts of astragalus membranaceus, 12 parts of pinellia ternate, 10 parts of trichosanthes kirilowii maxim, 10 parts of radix angelicae, 5 parts of red flower, 5 parts of radix curcumae, 3 parts of pomegranate bark and 3 parts of liquorice. By virtue of a regular medicine preparing method such as adding starch or edible cellulose into the materials, capsules, tablets or suspensoid can be prepared. The medicine composition can be used for treating both manifestation and root cause of recurrent respiratory tract infection and fever of children, and is non-toxic and harmless to a human body, low in cost, simple to prepare and wide in application prospect.

The invention belongs to the field of medical technology and discloses a benorilate vitamin B1 particle for children. The benorilate vitamin B1 particle for children is prepared from 0.3kg of benorilate, 3g of vitamin B1, 2.69g of sucrose, 3g of essence and 0.25g of tartrazine. The invention also discloses a preparation method of the benorilate vitamin B1 particle for children. According to the benorilate vitamin B1 particle for children, a good curative effect is brought to cold fever, the cold fever is not easy to relapse, less toxic and side effects are caused and the preparation process is simple and feasible.