33results about How to "Unsatisfactory solution" patented technology

The invention provides a straw returning cellulose-decomposing biological control fungus, a microbe and application of the microbe, wherein the strain F7JX993849 is classified and named as trichoderma virens which is preserved in China General Microbiological Culture Collection Center with preservation number of CGMCC No. 3.17613. The trichoderma virens F7JX993849 is relatively strong in decomposing capacity simultaneously on cellulose, hemicellulose and pectin components in crop straw. In addition, the trichoderma virens F7JX993849 is capable of antagonizing the invasion of phytopathogenic fungi, and has a biological control function. Furthermore, the microbe prepared from the straw returning cellulose-decomposing biological control fungus disclosed by the invention can be used for solving the problems of a general straw decomposition agent which is un-ideal in effect and is instable in the process of application; and the microbe has a good application value.

The invention discloses a Chinese word segmentation method based on a bi-directional long-short time memory network model. Firstly, the data set of any field is obtained, and the data set is divided into training set and test set. Then, the training set is preprocessed, the preprocessed training set and the open data set msr of Microsoft Research Asia are embedded separately, and the processed training set and data set msr are input into Bi-LSTM neural network model is trained to get the training set model X_Bi-LSTM Model and msr_Bi-LSTM model, and then using X_Bi-LSTM model, msr_Bi-LSTM modelpredicts the label of the test suite, and weights the forecasting probabilities of the two models, the probability of each Chinese character label after combination is obtained, then, the Viterbe algorithm is used to calculate the probability of each label after the combination to obtain the final probability of each Chinese character belonging to each label, and the label with the maximum probability value belongs to each Chinese character as the label of each Chinese character, thereby completing Chinese word segmentation. The invention can obtain better word segmentation result, and improves the accuracy of word segmentation.

The invention particularly relates to an ultrasonic thyroid nodule benign and malignant prediction method based on deep learning. Comprising the following steps: S1, collecting an ultrasonic nodule image sample, and preprocessing the ultrasonic nodule image sample, wherein the preprocessing comprises image sample quality improvement and image sample data labeling; s2, training a deep neural network based on the image sample annotation data; s3, applying the trained deep neural network to a real-time ultrasonic image to obtain a segmented nodule region image; s4, based on the segmented nodule region image, marking nodule types through a nodule judgment database, and taking the nodule types as training samples of a nodule prediction model; s5, adopting a feedforward neural network, using labeled sample data as a training set network, calculating accuracy and recall rate through feature selection, carrying out learning training through repeated iteration, and selecting a network model with the best evaluation result for output; and S6, predicting the type of the thyroid nodule based on the trained network model.

The invention relates to a roadbed compaction quality detection device and method. The method comprises the following steps: when a road roller rolls a roadbed, obtaining a vertical acting force applied by a vibrating wheel of the road roller to the roadbed, and obtaining a vertical displacement amount of the vibrating wheel at the same time; and according to the obtained vertical acting force applied to the roadbed by the vibrating wheel and the vertical displacement of the vibrating wheel, judging a roadbed compaction index and outputting. According to the detection method, the vertical acting force and the corresponding vertical displacement borne by the roadbed during roadbed rolling are directly obtained, so that the roadbed compaction index capable of more accurately reflecting the roadbed compaction degree is obtained, and the roadbed compaction quality detection efficiency and precision are improved; and compared with an existing vibration wheel center vertical acceleration calculation mode, the detection method can reflect the roadbed compactness more accurately.

The invention discloses a method for preventing and treating tobacco virus diseases in the technical field of tobacco disease prevention. The method comprises the following steps of 001, medicament selection, specifically, selecting Jin weiguo No.5, 8% ningnanmycin water aqua and 10% diethyl aminoethyl hexanoate soluble granules; 002, medicament blending, specifically, (1) preparing four stirringrods, a mixing container and three dissolving containers, (2) respectively putting one of three medicaments including the Jin weiguo No.5, the 8% ningnanmycin water aqua and the 10% diethyl aminoethylhexanoate soluble granules into one dissolving container, fully dissolving the medicaments to prepare mother liquor, and (3) sequentially pouring the mother liquor into the mixing container; and 003,pesticide spraying, specifically, containing pesticides into a sprayer, and spraying the pesticides in the morning and evening of sunny weather. Compared with the prior art, the method has the beneficial effects that the three medicaments for inhibiting viruses, promoting growth and supplementing nutrition are mixed for use, the tobacco mosaic virus disease is prevented and treated by combining the three manners, and the pesticides are applied in the morning and evening of a sunny day, so that the prevention and treatment effect of the pesticides is further ensured.

The invention relates to a special effect model training method and device and a special effect generation method and device. The special effect model training method comprises the steps that a training sample set is acquired, wherein the training sample set comprises a plurality of training images; the training images and a first mask graph are input into a special effect model to obtain an output image and a second mask graph after a special effect is added to a preset object in the training images, wherein the first mask graph is the mask graph of the preset object in the training images, and the second mask graph is the mask graph of the preset object after the special effect is added to the output image; the value of a target loss function is determined based on the training images, the output image, the first mask graph and the second mask graph; and the parameters of the special effect model are adjusted based on the value of the target loss function, and the special effect model is trained. According to the invention, the problem that the GAN special effect is not ideal in color, texture and other attributes in the prior art can be solved.

The invention is applicable to the technical field of biomedicine, and provides a gene-loaded exosome as well as a preparation method and application thereof, and the gene-loaded exosome is obtained by loading an SMAD7 gene into an exosome carrier; wherein the exosome carrier is formed by conjugating a CD34 antibody and phospholipid polyethylene glycol succinimide ester and then binding the conjugated antibody and phospholipid polyethylene glycol succinimide ester with the surface of a secretory exosome. A CD34 antibody is conjugated with phospholipid polyethylene glycol succinimide ester and then is combined with the surface of a secretory exosome to synthesize an exosome carrier with a high endothelial cell targeting characteristic, and then an SMAD7 gene is loaded into the exosome carrier, so that a novel gene-loaded exosome is prepared; a new preclinical research strategy is provided for targeted therapy of HO, and a certain theoretical basis is provided for further research and elaboration of pathogenesis of HO.

The present invention relates to the field of electronic circuits, especially to a bypass parallel constant current source circuit. The present invention includes a controllable load connected in parallel to the load and the controllable load includes an operational amplifier IC1 and a feedback unit , the current on the bypass circuit flows to the operational amplifier IC1 through the feedback unit, and the operational amplifier IC1 outputs a current corresponding to the main circuit according to the feedback current, so as to achieve the effect of constant output current, and has the characteristics of stable output current. It solves the problem of unsatisfactory performance and easy instability of the traditional series constant current source and improves the power efficiency.

The invention relates to the technical field of natural language understanding, discloses a long text-oriented semantic matching method and system, and the method is used for solving the problem of unsatisfactory effect of a text semantic understanding method in the prior art. The method comprises the following steps: performing data processing on an input text, wherein the data processing comprises removing special characters, segmenting words and segmenting characters; mapping the input text subjected to data processing into a numerical sequence; inputting the numerical sequence of the inputtext into a feature extraction model to obtain a feature vector of the input text; clustering based on the feature vectors; based on the clustered database, selecting TOP-N types of candidate data most similar to the input text from the database; and performing similarity measurement on the feature vector of the input text and the feature vector of the candidate data, and selecting TOP-K data most similar to the input text from the candidate data. The method is suitable for semantic matching of the long text.

The invention belongs to a selective wave absorber applied to radiation cooling, and particularly relates to a glass photonic crystal selective wave absorber based on a DBS algorithm, so as to achievethe purpose of radiation cooling. The wave absorber is composed of periodic structures, and one periodic structure unit comprises an optimized selective absorption area, an absorption layer and a reflection layer. According to the selective absorber designed on the basis of the DBS algorithm, the problem of multilayer loss in the process of manufacturing the multilayer radiation cooling selectiveabsorber is solved, the problem that the effect of optimizing the radiation cooling selective absorber through the algorithm is not ideal is solved, and the selective absorber which is excellent in performance, single in material and capable of being produced in batches is achieved.

The invention provides a processing method of fruit and vegetable chips through pulsating vacuum in-situ steam explosion. The processing method comprises the working procedures of selecting fruit andvegetable raw materials, performing cleaning, performing cutting to obtain slices, performing color protection, performing pulsating vacuum in-situ steam explosion puffing and performing vacuum packing. In the working procedure of the pulsating vacuum in-situ steam explosion, the characteristics of high temperature and high permeation of saturated steam are used, the raw materials are quickly anduniformly aged, in-situ steam explosion (namely first puffing) is realized, and further a pulsating vacuum puffing method and a quick shaping method are used, so that dainty, fragrant, crisp and healthy fruit and vegetable chips are made. Through adoption of the processing method, the drying and puffing time is greatly shortened, the drying and puffing temperature is reduced, and the drying processing of the fruit and vegetable chips can be completed at 40 DEG C or below within a few minutes; and the fruit and vegetable chips processed by the processing method disclosed by the invention are crispy in mouth feel and high in nutrient components, original color, original fragrance and original taste of fruits and vegetables can be well reserved, and the fruit and vegetable chips are bran-newpurely natural non-deep-fried fruit and vegetable foods.

The invention relates to a copolymer, particularly a copolymer with electric storage characteristics. The copolymer is formed by copolymerizing a monomer X and a monomer Y, wherein the monomer X is a carbazolyl group, and the monomer Y is an azo group; the mass percent of the monomer X is 30-90%, and the mass percent of the monomer Y is 10-70%; and the number-average molecular weight of the copolymer is 5000-15000. The invention also relates to a method for preparing the copolymer. The invention also relates to an electric storage device prepared from the copolymer and a preparation method of the electric storage device. The copolymer with electric storage characteristics and the electric storage device prepared from the copolymer can implement binary storage, have excellent stability, provide more options for large-scale application of organic electric storage materials and devices, and solve the problems of fewer varieties of electric storage materials and poor effects at present to some extent.

The invention discloses an intelligent LOW-E glass film removing machine in the technical field of LOW-E glass film removing, wherein the intelligent LOW-E glass film removing machine comprises a motor and an inclined base frame; the motor is fixedly arranged on the side wall of the base frame, a mounting plate is fixedly arranged on the base frame, two symmetrical vertical sliding through grooves are vertically formed in the mounting plate, and a synchronous rod is vertically and slidably arranged on the inner wall of each vertical sliding through groove; and one clamping plate is fixedly arranged at the ends of the two synchronous rods located on the front side of the vertical sliding through groove, an equipment plate is fixedly arranged at the ends of the two synchronous rods located on the rear side of the vertical sliding through groove, a supporting plate is fixedly arranged at the lower end of the clamping plate, and a supporting rod is rotationally arranged on the side wall of the supporting plate. According to the invention, the problems that an existing Low-E glass edge film is easy to cause residues, the film removing depths are inconsistent and the edges are irregular due to the fact that a manual film removing mode is usually adopted in the existing Low-E glass edge film removing method are effectively solved; and secondly, many pieces of irregular glass appear, and an existing film removing machine adopts an operation table as a glass carrying table, so that the film removing machine is difficult to remove films on the glass with irregular edges.

The invention relates to a liquid crystal copolymer with electric storage characteristics and a preparation method thereof, and application of the copolymer in electric storage. The liquid crystal copolymer with electric storage characteristics is formed by copolymerizing a monomer X and a monomer Y, wherein the monomer X is a carbazolyl group, and the monomer Y is an azo group; the mass percent of the monomer X is 30-90%, and the mass percent of the monomer Y is 10-70%; and the number-average molecular weight of the copolymer is 5000-15000. The invention also relates to a method for preparing the copolymer. The invention also relates to an electric storage device prepared from the copolymer and a preparation method of the electric storage device. The liquid crystal copolymer with electric storage characteristics and the electric storage device prepared from the copolymer implement ternary storage by utilizing the liquid crystal property, have excellent stability, provide more options for large-scale application of organic electric storage materials and devices, and solve the problems of fewer varieties of electric storage materials (especially fewer varieties of electric storage materials capable of implementing ternary storage) and poor effects at present to some extent.

The invention relates to a machine learning method based on difficult sample learning and neural network fusion, and belongs to the technical field of artificial intelligence. According to the method,learning is carried out on all training sample sets and difficult training sample sets to obtain two neural networks, and then the two neural networks are fused to obtain a new neural network with better performance; the difficult sample learning and network fusion process can be repeatedly carried out for many times until the performance of the neural network cannot be improved, so that the difficult sample calculation precision is effectively improved on the premise of keeping the easy sample calculation precision; the method is also suitable for networks obtained in any other modes, for example, networks obtained by learning on newly added training samples are fused with an original network in incremental learning, or networks obtained by two different learning methods are fused. According to the method, two or more different neural networks are fused into a network with better performance, the overall calculation is more reliable, and the precision is higher.