Abstract: A nanodelivery system based on globulin-1 S allele, a preparation method therefor and use thereof are provided. The preparation method includes the following steps of: dissolving globulin-1 S allele in a NaCl solution, and performing magnetical stirring until the globulin-1 S allele is completely dissolved to obtain a mixed solution; self-assembly: centrifuging the mixed solution, taking the supernatant to be dispersed into a dispersion solution for self-assembly to obtain a globulin-1 S allele dispersion; and adding hydrophobic plant polyphenol to the globulin-1 S allele dispersion, performing well mixing and centrifuging, and removing the supernatant to obtain the nanodelivery system based on the globulin-1 S allele.

Abstract: A progressive localization method for text-to-video clip localization. The method comprises: first, respectively extracting features of two modes, namely a video mode and a text mode by using different feature extraction methods; then progressively selecting different step sizes, and learning the correlation between the video and the text in multiple stages; and finally, training a model in an end-to-end manner based on the correlation loss of each stage. Moreover, the fine time granularity stage is fused with information of the coarse time granularity stage by means of a condition feature update module and up-sampling connection, such that different stages are mutually promoted. Different stages can pay attention to clips with different time granularities, and the model can cope with the situation that the length of a target clip is obviously changed based on the interrelation between the stages.

Abstract: The present disclosure discloses a method and a system for wavelet domain-based normalizing flow super-resolution image reconstruction. The method includes constructing a training set and a normalizing flow model, wherein the normalizing flow model includes a plurality of levels, each of the plurality of levels including a squeeze layer, two types of conditional mapping layers, a split layer, an activation standard layer, and a quick response (QR) layer; determining a stable normalizing flow model through a wavelet transform, a reconstructed QR layer, and a T-distribution based on the normalizing flow model; determining a wavelet domain-based normalizing flow super-resolution model by adding a refinement layer based on the stable normalizing flow model; training the wavelet domain-based normalizing flow super-resolution model based on the training set; and reconstructing a super-resolution image based on a trained normalizing flow super-resolution model.

Abstract: A method and a system for filtering ill corpus are provided. The method includes following steps: acquiring a text corpus to be recognized, and preprocessing the text corpus to be recognized to obtain a basic text corpus; extracting entities in the basic text corpus, and performing matching search on the entities of the basic text corpus according to an ill-text knowledge graph to obtain a first recognition result; detecting and recognizing the basic text corpus according to a corpus recognition model to obtain a second recognition result; and filtering the text corpus to be recognized according to the first or/and the second recognition result, and updating the ill-text knowledge graph according to the second recognition result. With semantic network essence and strong correlation ability of knowledge graph technology, candidate ill entities can be obtained, thus facilitating filtering of obscure ill information in forms of pinyin, homophonic words and split words.

Abstract: A method for determining a degree of doneness of a sturgeon meat based on flavor fingerprinting and a partial least squares (PLS) regression is provided. The method determines the degree of doneness of a sturgeon meat sample and predicts a flavor characteristic based on gas chromatography-ion mobility spectrometry (GC-IMS) data and a linear regression equation of the sturgeon meat sample.

Abstract: Disclosed is a method for performing fine-grained text sentiment analysis based on the users' harshness. According to the method, product attributes are automatically extracted according to comments in a data set, and the comments are split according to the product attributes, so as to achieve the effect of fine-grained analysis. Then, an inference model, which takes users' harshness into consideration, is used, and the inference model takes the users' harshness as hidden variables and solves same by means of an expectation-maximization (EM) algorithm, so as to obtain evaluation results of several product attributes of each product, thereby further improving the accuracy of sentiment analysis. According to the method, the problems in sentiment analysis of analysis granularity being too large and the users' harshness differing are solved, such that high-quality sentiment analysis can be performed on product comments.

Abstract: The present invention discloses Lactobacillus sakei for alleviating allergy and a preparation method thereof, and particularly Lactobacillus sakei for alleviating wheat-dependent, exercise-induced anaphylaxis (WDEIA). The Lactobacillus sakei is classified and named as Lactobacillus sakei. The Lactobacillus sakei has been preserved in China Center for Type Culture Collection since Apr. 12, 2003, with a preservation number of CCTCC DB 20082413. The Lactobacillus sakei can regulate the immune function of a human body to prevent/treat WDEIA by reducing the contents of specific IgE and histamine in serum of WDEIA patients.

Abstract: A method for establishing an epicutaneously sensitized food allergy animal model and belongs to the technical field of medical evaluation and detection. The method includes an allergen sensitization stage and an allergen challenge stage, wherein an MC903 ethanol solution containing OVA is smeared on the skin of experimental animals in the allergen sensitization stage, and the OVA dissolved in a buffer solution is used for an oral challenge in the allergen challenge stage.

Abstract: The present disclosure provides a method for detecting a defect of a film. The method includes obtaining a film image, determining one or more pieces of scratch information corresponding to the film image through processing the film image using a recognition model, the recognition model includes a convolution layer, a regression layer, and a classification layer, determining whether each piece of scratch information in the one or more pieces of scratch information meets a preset condition, each piece of scratch information includes position information, angle information, and size information, in response to a determination that each piece of scratch information meets the preset condition, adding one or more pieces of annotation information to the one or more pieces of scratch information that meets the preset condition, and generating prompt information based on the one or more pieces of annotation information.

Abstract: The present disclosure provides a method for recognizing a small target based on a deep learning network. The method comprises: determining, based on a collected image, spot defect information through a recognition model including a first feature determination layer, a second feature determination layer, and a spot defect determination layer, determining, based on the collected image, a first feature map, determining, based on the first feature map, a second feature map by fusing with positional encoding, determining, based on the second feature map, a third feature map through the second feature determination layer, and obtaining, based on the third feature map, positional information of the spot defect through a first determination layer, and determining, based on the third feature map, classification information of the spot defect through a second determination layer.

Abstract: The present disclosure provides a method for recognizing a small target based on a deep learning network. The method comprises: determining, based on a collected image, spot defect information through a recognition model including a first feature determination layer, a second feature determination layer, and a spot defect determination layer, determining, based on the collected image, a first feature map, determining, based on the first feature map, a second feature map by fusing with positional encoding, determining, based on the second feature map, a third feature map through the second feature determination layer, and obtaining, based on the third feature map, positional information of the spot defect through a first determination layer, and determining, based on the third feature map, classification information of the spot defect through a second determination layer.

Abstract: An end-to-end multimodal gait recognition method based on deep learning includes: first extracting gait appearance features (color, texture and the like) through RGB video frames, and obtaining a mask by semantic segmentation of the RGB video frames; then extracting gait mask features (contour and the like) through the mask; and finally performing fusion and recognition on the two kinds of features. The method is configured for extracting gait appearance feature and mask feature by improving GaitSet, improving semantic segmentation speed on the premise of ensuring accuracy through simplified FCN, and fusing the gait appearance feature and the mask feature to obtain a more complete information representation.

Abstract: A preparation method of eicosapentaenoic acid (EPA) ethyl ester is provided and relates to the field of EPA ethyl ester processing technologies. The preparation method includes: performing degumming and deacidification on a sardine crude oil to obtain a semi-refined sardine oil, performing esterification on the semi-refined sardine oil to obtain an esterified sardine oil and then performing winterization on the esterified sardine oil to thereby obtain an semi-refined esterified sardine oil, performing bleaching and deodorization on the semi-refined esterified sardine oil, and performing multi-stage distillation treatment. Finally, the sardine oil is purified by liquid chromatography to obtain the high purity EPA ethyl ester. The preparation method can improve a utilization rate of the sardine oil and obtain the high-purity EPA ethyl ester.

Abstract: The present disclosure provides a method for detecting a defect of a film. The method includes obtaining a film image, determining one or more pieces of scratch information corresponding to the film image through processing the film image using a recognition model, the recognition model includes a convolution layer, a regression layer, and a classification layer, determining whether each piece of scratch information in the one or more pieces of scratch information meets a preset condition, each piece of scratch information includes position information, angle information, and size information, in response to a determination that each piece of scratch information meets the preset condition, adding one or more pieces of annotation information to the one or more pieces of scratch information that meets the preset condition, and generating prompt information based on the one or more pieces of annotation information.

Abstract: The present disclosure relates to a searchable encrypted data sharing method and system based on blockchain and homomorphic encryption, which protects security of sensitive data on the blockchain and realizes searchable and homomorphic calculation of data ciphertext. According to the present disclosure, a data owner encrypts the generated sensitive data and the keywords extracted according to the data with his own key, and then sends the encrypted transaction information to the cloud server. The cloud server verifies the identity of the data owner. If the verification succeeds, the uploaded ciphertext data is stored on a local server, and a ciphertext index, keyword ciphertext and related evidences of the data storage are uploaded to an alliance chain. The alliance chain node verifies the consistency of the uploaded transaction information, and if the verification succeeds, the transaction information is recorded.

Abstract: A three-dimensional electrode-ozone oxidation-electrocatalytic membrane coupled wastewater treatment device, including a circulating fluidized bed reactor. The circulating fluidized bed reactor includes a funnel-shaped internal, a truncated cone, a fiber ball filter, a gas-liquid distribution plate, an inner cylinder, an intermediate cylinder and an outer cylinder. The inner cylinder, the intermediate cylinder and the outer cylinder are coaxial. The inner cylinder is an electrocatalytic membrane assembly; the intermediate cylinder is a gas diffusion electrode; and the outer cylinder is a stainless-steel mesh. A particle electrode is filled between the intermediate cylinder and the outer cylinder, and between the intermediate cylinder and the inner cylinder. The intermediate cylinder is connected to a negative electrode. The inner cylinder and the outer cylinder are connected to a positive electrode. A wastewater treatment method using the device is also provided herein.

Abstract: An end-to-end multimodal gait recognition method based on deep learning includes: first extracting gait appearance features (color, texture and the like) through RGB video frames, and obtaining a mask by semantic segmentation of the RGB video frames; then extracting gait mask features (contour and the like) through the mask; and finally performing fusion and recognition on the two kinds of features. The method is configured for extracting gait appearance feature and mask feature by improving GaitSet, improving semantic segmentation speed on the premise of ensuring accuracy through simplified FCN, and fusing the gait appearance feature and the mask feature to obtain a more complete information representation.

Abstract: The disclosure provides an MQTT-based distribution Internet of Things system and a method thereof. The system includes: a sensing ring including a plurality of data acquisition and monitoring devices and a sensing control module provided in a distribution network; an application ring including a plurality of application servers and an application control module; a communication ring including at least one communication module, each communication module being connected with the application control module and a corresponding sensing control module respectively.

Abstract: The invention provides a dynamic membrane reactor with function of nitrogen and phosphorus removal and an operation method thereof, and comprises a biological treatment system, a dynamic membrane loading system and an automatic system. The operation method comprises the following steps. (1) Before the formation of dynamic membrane, a porous filter for phosphorus removal is used as a cathode, a conductive precision filter screen is used as an anode, and aerobic denitrifying bacteria are inoculated into the dynamic membrane reactor under certain constant current density, hydraulic retention time and flux. (2) After the dynamic membrane is formed, the porous filter for phosphorus removal is used as the anode, the conductive precision filter screen is used as the cathode. And intermittent aeration is started at the anode under certain constant current density. (3) When the transmembrane pressure difference exceeds a certain range, hydraulic backwashing is performed under certain constant current density.

Abstract: The disclosure provides an MQTT-based distribution Internet of Things system and a method thereof. The system includes: a sensing ring including a plurality of data acquisition and monitoring devices and a sensing control module provided in a distribution network; an application ring including a plurality of application servers and an application control module; a communication ring including at least one communication module, each communication module being connected with the application control module and a corresponding sensing control module respectively.