206results about How to "Improve extraction ability" patented technology

The invention discloses a multi-triple extraction method based on an entity-relationship joint extraction model, which is characterized in that the method comprises the following steps: obtaining text, processing target text in clauses, and carrying out position, type and relation mark on each word in a sentence; establishing an entity-relationship joint extraction model; training the entity-relationship joint extraction model; according to entity-relation joint extraction model, carrying out three-tuple extraction. The three-part marking scheme designed by the invention can exclude entities that are not related to the target relationship in the process of entity relationship joint extraction. In addition, the multi-triple extraction method based on the entity-relationship joint extractionmodel can be used for extracting the multi-triple, and the model based on the triple extraction method of the invention has stronger multi-triple extraction ability compared with other models.

The invention relates to a thorium separation and purification method and in particular relates to an application of a neutral phosphamide extraction agent shown as a general formula I in the specification for extracting and separating thorium, as well as a method for extracting and separating thorium by using the neutral phosphamide extraction agent.

The invention relates to a gastric early cancer histological image classification system based on a deep neural network, and the system comprises a WSI pathological section obtaining module which is used for obtaining a gastric early cancer pathological section image; an image preprocessing module which is used for preprocessing the early gastric cancer pathological section image; a data set division module which is used for dividing the data set after image preprocessing into a training set and a verification set, a training set which is used for model training, and the verification set is used for verifying a model effect; a feature extraction module which is used for extracting picture features of the training set or the verification set based on a GoogleNet double-branch network structure and outputting feature vectors of corresponding pictures; and a classifier training module which is used for training based on the neural network model to obtain a gastric early cancer classification model.

The invention relates to a treatment method of oily sludge, which is particularly suitable for oily sludge generated during the sewage treatment process of a refinery plant. The method comprises the following steps: after centrifugal dewatering, the oily sludge is dehydrated through rotary dehydration equipment under the control of a certain negative pressure and a certain temperature, so as to destroy the water, oil and solid stable system in the oily sludge; parts of oil and water in the oily sludge are distilled out; the coke generated after dehydration is extracted with a solvent; after extraction, the material undergoes the solid-liquid separation; the separated liquid phase is delivered to a delayed coking device; the separated solid phase is dehydrated to form the residues that meet the discharge standards for solids; and the wastewater generated during the process is biochemically treated and then discharged after reaching the discharge standards. The invention is mainly characterized in that the sludge treatment process is short, the speeds and the efficiencies of sludge dehydration and extraction are high, the effect is good, the operational condition is moderate, the energy consumption is reduced, the oily sludge treatment effect is good, and the operating maintenance of the equipment is simple.

The invention provides a magnetic mesoporous silica gel material extracting agent used for enriching and / or separating trace organic pollutants in an environment water sample, and a preparation method thereof. A core of the extracting agent provided by the invention is a silicon gel microsphere (i.e. magnetic silicon gel microsphere) wrapped with a plurality of Fe3O4 magnetic nano particles, and a shell of the extracting agent is a mesoporous silicon gel shell layer with inner and outer surfaces modified by C18 silanization. The extracting agent integrates huge special surface area, excellent extracting capacity and ultrahigh extracting volume of a mesoporous material, magnetic separating capacity of a magnetic material, strong extracting capacity of C18 groups and special volume exclusion function of a vertical channel of the mesoporous material. In addition, the extracting agent provided by the invention has the advantages of high extracting efficiency, low preparation cost and simple preparation method, and is very suitable for preprocessing samples of the trace organic pollutants in a bulk-mass water sample.

The invention discloses a method for detecting flaws of printed fabric based on a Gabor filter. The method comprises the steps that the basic Gabor filter is established and Gabor parameters are extracted; the extracted Gabor parameters are selected, intersected and varied, the Gabor parameters are changed, the parameters with the high adaption degree of an objective function are selected, through intersection and variation, the selected parameter with the high adaption degree are transformed, and therefore parameters with the highest adaption degree are generated; according to a Gabor parameter direction theta and the center frequency u0 which are selected through a genetic algorithm, rotation transformation is conducted on the obtained Gabor parameters, and therefore effective flawless printed fabric textural feature information is extracted; Gabor filtering convolution operation is conducted on a printed fabric image to be detected and a flawless printed fabric image, so that texture background information of printed fabric to be detected is extracted; binaryzation is conducted on the printed fabric image to be detected and the flawless printed fabric image, so that a flaw detection result of the printed fabric is obtained. The method for detecting the flaws of the printed fabric based on the Gabor filter can improve detection efficiency and detection accuracy.

The invention relates to a new technology for refining Fischer-Tropsch synthesis light fraction oil. The new technology comprises solid-phase adsorption, extraction refining, separation and extraction agent refining and a recycling unit. In the new technology for refining Fischer-Tropsch synthesis light fraction oil provided by the invention, Fischer-Tropsch synthesis light fraction oil is used as a raw material; metal ion impurities and oxygenated compounds in the Fischer-Tropsch synthesis light fraction oil are removed by the technology unit while the recycled refined extraction agent is circularly utilized. The new technology for refining Fischer-Tropsch synthesis light fraction oil provided by the invention has the advantages of simple process, high recycling efficiency, recycling and cyclic utilization of extraction agent, low cost of extraction agent, small investment, and the like.

The present invention discloses the solvent impregnated resin which uses amide compound as the extraction agent. The extraction agent is embedded in the polymer. The prevent invention is characterized in that the polymer is the styrene - divinylbenzene copolymer resin, and the extraction agent is the amide compound with the content of 20 percent to 60 percent. The solvent impregnated resin is ivory ball particle in appearance, with a particle diameter of 60 to 200 items a heap density of 0.53 to 0.94 gram per milliliter, and a specific surface area of 128 to 260 millimeters 2 / gram. The present invention also discloses the preparation method for the solvent impregnated resin. The solvent impregnated resin provided by the present invention has the advantages of low cost, less wastage of extraction agent, high column load, good mass transfer performance, easiness to be regenerated, long service life, ease use, good repeatability, and high reproducibility and so on. The solvent impregnated resin is of the characteristics of both particle and liquid in terms of extraction and cleaning. The present invention can be widely used in the separation and analysis of 90Sr in the plant and the animal samples, in the removal and callback of 90Sr during the treatment of nuclear fuel high-level radioactive liquid waste, as well as in the separation, purifying and callback of radioactive medical isotopes 90Sr.

The invention discloses a fresh keeping agent containing essential oil, for fruits, and an application. The fresh keeping agent for fruits is characterized by being compounded from lysozyme, tea polyphenols, nisin, essential oil A, essential oil B, Vc and distilled water in a certain proportion. According to the fresh keeping agent for fruits, disclosed by the invention, the essential oil is optimized to be used as a main component of the fresh keeping agent, so that the fresh keeping agent has excellent effects of bacteriostasis and oxidation resistance and is applied to corrosion resistance and fresh keeping of the fruits, the fresh keeping and storage period of the fruits can be greatly prolonged, the fruits are guaranteed to maintain original quality in the fresh keeping period, and the quality of the fruits is effectively guaranteed. The fresh keeping agent for the fruits, disclosed by the invention is simple in compounding method, and convenient to apply; and the fresh keeping agent is completely compounded from natural components, and harmful chemical substances are not added, so that the fresh keeping agent is safe and environmental-friendly, and has an important industrial value.

The invention discloses a shadow detection method based on an attention mechanism. The shadow detection method comprises the steps of obtaining a to-be-processed public shadow data set for a trainingsystem; obtaining a to-be-detected shadow image from camera equipment or a local hard disk; preprocessing the public shadow data set and obtaining a corresponding training set and a corresponding testset; preprocessing the to-be-detected shadow image; constructing and fusing all modules of a system core neural network to form a convolutional neural network based on an attention mechanism; calculating network prediction and label loss, and adjusting network parameters according to the loss; completely training the deep convolutional neural network and inputting a preprocessed to-be-detected shadow image into the deep convolutional neural network; and outputting a shadow detection result, and performing shadow attribute classification on each pixel to complete a shadow detection process ofthe custom data. According to the invention, the shadow feature extraction capability is improved, the semantic relevance is reduced, the generalization and high efficiency of the detection system areimproved, and the shadow detection result is more accurate.

The invention discloses a purification method of a cobalt nickel hydroxide hydrochloric acid leaching solution. The purification method mainly comprises the following processing steps: I. dissolving and leaching a cobalt nickel hydroxide raw material produced by laterite-nickel ore transformation through industrial concentrated hydrochloric acid; II. adding barium chloride to a leaching solution and implementing sulfate radical removal operation; filter-pressing the solution and reserving the solution as an extracting pre-solution for later use; III. preparing an organic phase at ratio of 25% to 75% of P5O7 to sulfonated kerosene; saponifying the prepared organic phase in a soda soap kettle; reacting the saponified organic phase with a pure nickel chloride solution to obtain nickel soap; and IV. extracting the nickel chloride extracting pre-solution obtained from the step II through the P507 nickel soap, wherein the extracting solution obtained is a finished pure nickel chloride solution. The production process disclosed by the invention is safe and the extracting solution produced is better in component; and moreover, the purification method is short in technological process and more effective in process control.

The invention discloses a natural plant extract for hair care and a preparation method and application thereof. The natural plant extract is prepared from the following raw materials in parts by mass: 40-65 parts of polygonum multiflorum, 30-50 parts of Chinese angelica, 20-35 parts of cacumen biotae, 10-25 parts of sapindus, 20-50 parts of tea seed and 25-40 parts of melaleuca bracteata. In the invention, the effective components in the raw materials are extracted by the supercritical fluid technology; the obtained extract has high content of effective components, contains a great quantity of natural surface active components and natural nutrition conditioning components, does not irritate the scalp, and can thoroughly clean the hair to realize the effects of nourishing, oil control, dandruff removal and itching relief; and after washing hair, the hair becomes naturally glossy, soft, smooth and easy to comb. The natural plant extract disclosed by the invention is simple to prepare, has stable quality and high safety, and can be widely used for preparing the hair washing and hair care articles.

The invention relates to a method and a device for extracting an aroma component. The method comprises the following steps of: (1) performing pretreatment on a material, and then performing pressure reduction treatment on the treated material in an extraction container; (2) introducing a solvent into the extraction container to mix the solvent and the material, and extracting the mixture to obtain extract liquor; (3) putting the extract liquor into a vaporization separation device to separate out a crude extract of the aroma component, and simultaneously separating the solvent, compressing and cooling the solvent, and then reclaiming the solvent for cyclic utilization; and (4) putting the crude extract of the aroma component into a refining system to perform refining treatment to prepare a refined extract. The extraction device comprises an extraction separation system, the refining system, a solvent purification system, a vacuum system and a heat exchange system. The aroma component extracted by the method and the device has no solvent residue, and the effective active components of a natural plant raw material can be maintained to the utmost extent.

The invention discloses adsorption and separation regulation type supercritical fluid extraction equipment, belongs to the field of pharmaceutical equipment, and aims to extract effective substances from traditional Chinese herbal medicines through supercritical properties of carbon dioxide. The equipment is mild in extraction adjustment, free of pollution, good in extraction performance, high inextraction rate, short in extraction time and short in production cycle; when effective substances are completely extracted and a separation step is carried out, a novel regulation adsorption component is introduced, effective substances which are not completely released from the carbon dioxide can be actively adsorbed and retained, so that the recycling rate of the effective substances can be increased; meanwhile, the equipment is supplemented by a sampling detection technique of the prior art, the adsorption component is adjusted according to detection on contents of the effective substancesin the carbon dioxide, the carbon dioxide can be rapidly circulated on premise that adsorption retention is met, the overall extraction efficiency and quality can be remarkably improved, and in addition, the effective substances can be conveniently desorbed and recycled and continuously reused through the adsorption component.

The invention discloses a process for preparing caprolactam, which comprises the following steps: after mixing cyclohexanone and cyclohexane evenly, carrying out an oxamidine reaction to obtain a cyclohexane solution of cyclohexanone oxime; enabling the cyclohexane solution of the cyclohexanone oxime and oleum to react to generate a caprolactam sulfate solution, curing the caprolactam sulfate solution, then carrying out a neutralization reaction on the cured caprolactam sulfate solution and ammonia, afterwards separating the two materials to obtain a crude product of caprolactam and then carrying out extraction and water back extraction respectively to prepare a finished product. Because of the existence of a cyclohexane inert solvent, rectification and separation do not need to be carried out after the oxamidine reaction, and only a rearrangement reaction needs to be carried out after the water phase of a catalyst is separated, thereby saving a rectification device and the steam consumption; as a result, a caprolactam enterprise with an annual yield of a hundred thousand tons can save energy worth approximately 60000000 yuan every year and can also save the one-off equipment investment totaling 60000000 yuan, the technological process of caprolactam is simplified, the procedures of water washing, extraction, distillation and the like are reduced, the production period is shortened, and the annual yield is enhanced.

The invention discloses a question and answer method and device based on artificial intelligence, computer equipment and a storage medium, and the method comprises the steps: carrying out the languagemodel training based on first training data, carrying out the NER model training based on second training data and a trained language model, and carrying out the relation matching model training; identifying an entity in a to-be-processed statement based on the trained NER model, and obtaining a relationship corresponding to the to-be-processed statement based on the trained relationship matchingmodel; and determining and outputting an answer corresponding to the to-be-processed statement according to the relationship corresponding to the to-be-processed statement and the entity in the to-be-processed statement. Based on language model transfer learning and graph transfer learning technologies, a training method commonly used by a language model is improved, higher accuracy can be achieved through a small amount of manual marking data, and the method is more suitable for constructing a knowledge graph question-answering system.

The invention relates to a #-shaped LBP and deep learning combination-based human face identification method, and belongs to the technical field of artificial intelligence. For the problems of insufficient feature extraction and classifier fitting of a conventional LBP, a human face identification method based on a #-shaped LBP and deep learning of a local texture feature is proposed. By utilizingan improved LBP algorithm, the local texture feature of a human face image is extracted, and an LBP histogram is established; then a deep belief network-based deep learning framework is established,the LBP histogram is input to a deep belief network, the network is trained by adopting an unsupervised layer-by-layer training method and a supervised BP algorithm, self-learning and self-optimization of the network are realized, and network parameters are obtained; and finally, the human face image is classified and identified by utilizing the DBN. According to the method, the distinctive texture feature of the human face image can be extracted; the superiority of the algorithm in identification rate is verified; and the robustness is relatively strong.

The present invention relates to a vanadium pentoxide preparation method, and provides a method for preparing vanadium pentoxide through composite extraction. The method comprises: (1) leaching; (2) reducing; (3) extracting, wherein a vanadium-containing material liquid is extracted with a composite extractant prepared from a phosphate organic matter and a modifier to prepare a vanadium-containing supported organic phase; (4) carrying out back extraction, wherein the supported organic phase is subjected to back extraction with sulfuric acid or a vanadium-containing sulfuric acid solution to prepare a vanadium-rich liquid; (5) removing acid, wherein the excess acid in the vanadium-rich liquid is extracted; (6) precipitating vanadium; and (7) calcining. According to the present invention, the vanadium is extracted with the composite extractant having strong extraction ability and strong back extraction ability, such that the vanadium extraction rate and the vanadium purity can be improved so as to improve the yield and the product of the vanadium pentoxide product, and the alkali consumption can be reduced so as to reduce the production cost.

The invention discloses a rotary-vane efficient energy-saving environmental-friendly vacuum pump, relates to a rotary-vane vacuum pump for a tile vacuum-extruder, belongs to the technical field of vacuum pumps and is suitable for being used as vacuuming equipment for standby use in building material industry, chemical industry, medicine industry, electronic industry, ceramics industry and the like, especially suitable for the rotary-vane vacuum pump used for the tile vacuum-extruder. The rotary-vane efficient energy-saving vacuum pump comprises a gas-water separator, an air filter, a pump body stator, a rotator, rotary vanes, a vent valve assembly, a side-placed oil tank, an air-cooled oil-pumping cooler, a venting oil mist filter, a centrifugal fan, a housing, a frame and a motor, wherein the gas-water separator comprises a cylinder body, a clapboard, a netting filter net cylinder and a cone river diversion port; the air filter is provided with a filter cylinder body which is internally provided with an exhaust duct; the filter cylinder body is internally provided with a filter core assembly; and the centrifugal fan is arranged on a coupler, the housing is arranged on the outer sides of the pump body and the air-cooled oil-pumping cooler, so as to form an air-cooled oil-pumping cooler and a pump body cooling air duct.

The invention discloses an airport X-ray contraband image detection method based on an attention mechanism. The method comprises the steps of forming a security check image database; obtaining a CNN model; constructing an attention mechanism model, and obtaining a contraband target attention map; suppressing noise and background information interference in the contraband target attention map, andacquiring a suppressed contraband target attention map; and generating a position window, and obtaining a contraband target detection map by using the inhibited contraband target attention map. According to the airport X-ray contraband image detection method based on the attention mechanism provided by the invention, the advantages of high image feature extraction capability, high recognition precision, high robustness and the like of the attention mechanism model are utilized, and automatic detection of various contrabands in an X-ray security inspection image is realized. Meanwhile, the training mechanism of the attention mechanism model is weak supervised training, so that extra manual annotation is not needed in the training process.

The invention relates to a treatment method of oil-containing sludge. The treatment method is especially suitable for the oil-containing sludge produced in sewage treatment processes of an oil refinery. The method includes feeding the oil-containing sludge to a greenhouse desiccation room, desiccating naturally under conditions of full sunshine, mixing mud cake with an extracting agent at room temperature, performing homogeneous extraction, feeding the material mixture to a solid-liquid centrifugal separator after the extraction is finished, and directly recycling separated liquid. Separated solid residue, after being dried, is mixed with coal and the mixture can be used as fuel for coal-fired boiler. According to the method, energy consumption for treatment of the oil-containing sludge is low, the technology process is short, the equipment investment is low and harmless treatment of the oil-containing sludge can be achieved.

The invention discloses a method for detecting benzopyrene in an environmental water sample. According to the method, alkyl carboxylic acid modified magnetic nano Fe3O4 is utilized to extract polycyclic aromatic hydrocarbon in the environmental water sample, and high performance liquid chromatography / a fluorescence detector is combined for content measurement; the modified magnetic nano Fe3O4 has large specific surface area, good dispersibility in aqueous solution, good selectivity for polycyclic aromatic hydrocarbon, and can reach high enrichment factor after elution of an eluent, and the detection sensitivity can be improved greatly; compared with the traditional sample pretreatment method, according to the pretreatment method, the cost is low and the operation is simple; and besides, the use of organic solvent is reduced greatly, the pretreatment time is shortened, and the magnetic nano Fe3O4 can be utilized repeatedly, is environmentally friendly, and has wide application prospect.

The invention relates to a complexation extraction dephenolization method of a semi coke wastewater containing phenol. An extractant and semi coke wastewater containing phenol are arranged in a thermostat water bath to be shaken, and stand still at a normal temperature to be subject to split phase so as to obtain a raffinate water phase and a solvent phase rich in phenols, the obtained solvent phase rich in the phenols is reextracted with back-extractant sodium hydroxide, the obtained regenerated extractant is recycled, and the obtained phenol resources are recovered. A complexation extraction method is adopted for the dephenolization process of the semi coke wastewater containing phenol, tributyl phosphate is used as a complexing agent, toluene is used as a diluent, and due to a special complexation extraction mechanism, a distribution coefficient of the complexation extraction dephenolization method is higher, the extracting power is strong, a complexation extractant has the advantage of reuse after being reextracted with alkali liquor, and valent matters in the wastewater are enriched, separated and recycled with the alkali liquor so as to realize the efficient treatment and resource recycling of phenols.

The invention belongs to the field of non-ferrous metal recovery, and particularly relates to a recovery method of a waste SCR denitration catalyst. The recovery method comprises the steps of pretreatment, TiO2 separation, extraction and purification, titanium recovery, vanadium recovery, tungsten / molybdenum recovery and the like. The recovery method is capable of recycling a several precious recovery products such as V2O5, ammonium tungstate / ammonium molybdate and BaTiO3 from the waste SCR denitration catalyst. The invention provides an effective technical scheme, and efficient resource utilization of the waste SCR denitration catalyst can be realized.

A cavity lining of plasma etching equipment comprises a cylindrical body. A wafer-conveying opening is formed in the body and used for taking and conveying a wafer to be etched. The outer wall of the body is matched with a reaction cavity of the plasma etching equipment. A circle of annular base plate is arranged at the bottom of the body and extends inward. The annular base plate is matched with the wafer to be etched in positions, and flow-evening slotted holes are evenly distributed in the annular base plate. A fixing flange is arranged at the top of the body and connected with the reaction cavity. The cavity lining of the plasma etching equipment has the advantages of being capable of evenly reacting etching gas in the reaction cavity and good in purity in the reaction cavity and service life of the reaction cavity is long.

The invention relates to a method for separating tetravalent cerium, thorium and rare earth. The method comprises following steps: feed liquid containing the tetravalent cerium, thorium and rare earth is provided; an amino-contained neutral phosphine extraction agent in the general formula I is used for performing extraction and separation on the tetravalent cerium from the feed liquid to obtain cerium-contained extraction liquid and cerium extraction tail liquid; the amino-contained neutral phosphine extraction agent in the general formula I is used for performing extraction and separation on the thorium from the cerium extraction tail liquid to obtain thorium-contained extraction liquid and thorium extraction tail liquid; and the rare earth in the thorium extraction tail liquid is recycled, and the chemical formula is shown in the specification, wherein R1 and R2 are each and independently selected from a C1-C12 alkyl group, R3 and R4 are each and independently selected from a C1-16 alkyl group and hydrogen, and n is an integer of one to eight.

The invention discloses a wheat powdery mildew spore segmentation method for a small sample image data set, belongs to the technical field of computer vision, and solves the problem that wheat powdery mildew disease analysis cannot be accurately performed through pictures in the prior art. The invention mainly designs a semantic segmentation method for performing image segmentation on wheat powdery mildew spores. Wherein the main body comprises two networks, one network is an adversarial generative network model for data enhancement, and the other network is a segmentation network model for segmenting target wheat powdery mildew spores. When a segmentation model is trained, in order to make up for the deficiency of training data, corresponding preprocessing needs to be performed on an image, and an adversarial generative network model and affine transformation need to be used for data enhancement. Meanwhile, in order to perform semantic segmentation on target spores, a contour entity of the wheat powdery mildew spores needs to be extracted through image masks for training.

The invention discloses a method for utilizing sorghum bicolor to restore caesium-contaminated soil, and belongs to the technical field of environment plant restoration. The method comprises: planting sorghum bicolor in caesium-contaminated soil, and after sorghum bicolor is mature, harvesting the overground part or integrally moving sorghum bicolor away, so as to reach the purpose of restoring caesium-contaminated soil. Energy-plant sorghum bicolor is high in biomass and strong in adaptability, the growth of sorghum bicolor is not influenced by caesium contamination, and the caesium content accumulated at the overground part of sorghum bicolor is higher than the caesium content accumulated at the roots. The method is friendly to environment, simple in operation and technology, convenient in manage, low in cost and good in restoration effect, provides a new choice for effective treatment of radioactive contaminated soil, and is applicable to the field of treatment on heavy-metal contaminated soil.

The invention provides a method for extracting gold from material liquid with the high gold content. The method comprises the following steps that first, the pH value of a solution is regulated; second, the solution is appropriately heated; third, acid is added into the solution for filtering and separating; fourth, primary coarse gold powder purification is conducted; fifth, the remaining solution is subjected to acidity regulating and extraction; sixth, reverse extraction solid-liquid separation is conducted. Furthermore, a reducing agent adopted in the sixth step is oxalic acid; furthermore, the reducing agent adopted in the sixth step is sodium sulfite; furthermore, zinc powder is added into the remaining tail solution in the sixth step, and platinum group metal can be displaced; furthermore, the method is suitable for waste gold electrolyte and gold chloroazotic acid dissolved solutions.

The invention provides an age determination method and device and an eye health degree determination method and device, and the method for determining the age based on a fundus image comprises the steps: obtaining at least one fundus image of a user; classifying the fundus image by utilizing a classification model to obtain a classification result for indicating ages; and determining the age of the user according to the classification result.