Abstract: The present disclosure provides systems and methods for image filtering. The method may include obtaining an initial image block from a reconstructed image; determining at least one candidate image block by performing a filtering operation on the initial image block using at least one trained machine learning model; and determining a target image block based on the at least one candidate image block.

Abstract: A method for screening pesticide residues in medicine food homology (MFH) characteristic agricultural products includes step 1, a liquid chromatography-mass spectrometry database of pesticides is constructed, standard solution are prepared for the pesticides and liquid chromatography and mass spectrometry are used to detect the standard solutions; step 2, a sample is processed and then is detected to obtain a liquid chromatography-mass spectrogram; step 3, the liquid chromatography-mass spectrogram obtained from step 2 and liquid chromatography-mass spectrograms in the liquid chromatography-mass spectrometry database obtained from step 1 are compared and analyzed individually.

Abstract: Provided are Enterococcus lactis, a drug for preventing or treating tumor, and use thereof, relating to the technical field of biomedicines. Novel Enterococcus lactis is separated and chosen, and the Enterococcus lactis has an important influence on the effect of tumor immunotherapy for cancer patients. Gut microorganisms and immune cells can interact with each other to jointly regulate the human immune system. The provided Enterococcus lactis can induce the differentiation and maturation of iDC (immature DC) cells, and mature DC can effectively activate T cells and increase the infiltration of tumor infiltrating lymphocytes having anti-tumor activity, thereby killing tumor cells. In addition, the provided Enterococcus lactis can induce the differentiation of M0 type macrophages into M1 or M2 type cells.

Abstract: It is related to a process for purifying organic amines. The process comprises (a) providing an organic amine to a first vessel (5), the organic amine having a normal boiling point at one bar; (b) filling the first vessel (5) with inert gas; (c) heating the organic amine in the first vessel (5) to a sub-boiling temperature, wherein the sub-boiling temperature is at least 15° C. less than the normal boiling point; (d) cooling the vapor from the first vessel (5) in a second vessel (20) to provide a liquid; and (e) contacting the organic amine with a resin polymer matrix, wherein the resin polymer matrix is embedded with an amino compound selected from the group consisting of iminodiacetic acid, aminomethylphosphonic acid or a combination thereof.

Abstract: The present application provides anti-CD137 constructs that bind to CD137 (e.g., anti-CD137 monoclonal antibodies and multispecific antibodies), nucleic acid molecules encoding an amino acid sequence of the anti-CD137 constructs, vectors comprising the nucleic acid molecules, host cells containing the vectors, methods of preparing the anti-CD137 constructs, pharmaceutical compositions containing the anti-CD137 constructs, and methods of using the anti-CD137 constructs or compositions.

Abstract: A power transformer fault diagnosis method based on a stacked time series network, includes: collecting gas-in-oil data of a transformer in each substation; performing z-score normalization on the collected data to obtain a normalized matrix; dividing the normalized matrix into a training set and a test set in proportion; constructing a stacked time series network based on Xgboost and a bidirectional gated neural network, and inputting the training set and the test set to perform network training; and normalizing real-time collected data to obtain trainable data to predict a fault and update network parameters. The gas-in-oil data is predicted by using Xgboost and a gated neural network, obtains prediction data of a power transformer from two time series networks by using a meta learner, and obtains a fault diagnosis result of the transformer by using a Softmax layer. The neural network has accurate fault diagnosis performance and stable robustness.

Abstract: The present disclosure provides a display panel, a manufacturing method thereof, and a display device. The display panel includes: a first medium and a first spacer wall between the first substrate and the second substrate, wherein the first sub-panel has filter pixels arranged at intervals, the first spacer wall is black and arranged along spaces between filter pixels, and a dielectric coefficient of the first spacer wall is greater than that of the first medium; and a second sub-panel on a light emergent side of the first sub-panel and including a second medium and a second spacer wall between the third substrate and the fourth substrate, wherein the second sub-panel has display pixels arranged at intervals, the second spacer wall is black and arranged along spaces between display pixels, and a dielectric coefficient of the second spacer wall is greater than that of the second medium.

Abstract: The present disclosure provides a display method. The display method includes: receiving a first operation, the first operation being used to adjust a proportion of a first region in a first display interface of an electronic device; and displaying, in the first region in response to the first operation, at least one launch entry of at least one application, when the proportion of the first region is smaller than or equal to a first threshold; and displaying, in the first region in response to the first operation, a display interface of a first application in the first region, when the proportion of the first region is greater than the first threshold. Embodiments of the present disclosure further provide a display apparatus and a storage medium.

Abstract: Disclosed of the present application is relation to deep learning based voice data processing. The voice data to be detected is converted into target text data based on a voice recognition model so that the keyword text corresponding to the predetermined target voice keyword can be converted. Then, the data is matched with the target text data to determine whether the voice data to be detected includes the target voice keyword based on the matching result. Thus, because the voice recognition model is obtained by deep learning based on the obtained voice recognition data training set, it can obtain high-precision target text data, thereby improving the accuracy of subsequent matching. The problem of low accuracy of detecting voice data for keyword detection can therefor be solved.

Abstract: The present disclosure relates to a method of manufacturing an array substrate. The method of manufacturing an array substrate may include forming a main via hole in a substrate, filling a first conductive material in the main via hole, and forming a pixel circuit layer on a first surface of the substrate. The pixel circuit layer may include a first via hole. An orthographic projection of the first via hole on the substrate may at least partially overlap the corresponding main via hole.

Abstract: A display substrate includes a base, a plurality of light-emitting devices disposed on the base, an encapsulation layer disposed on a light-emitting side of the plurality of light-emitting devices away from the base, and at least one photosensitive sensor disposed on a surface of the encapsulation layer away from the base. Each of the at least one photosensitive sensor is configured to collect optical signals for texture recognition.

Abstract: A method for increasing bulkiness of reconstituted tobacco by adding tobacco stem particles includes (1) pulverizing a first portion of tobacco stems to obtain tobacco stem particles; (2) classifying the tobacco stem particles with mesh sieves and selecting the tobacco stem particles with a predetermined mesh size; (3) extracting a second portion of the tobacco stems with water and grinding to form a tobacco stem slurry that has a beating degree of 12-14° SR, and mixing the tobacco stem slurry with tobacco leaves in a weight ratio of 6:4 and grinding to obtaining a tobacco slurry that has a beating degree of 18-20° SR; (4) cutting plant fiber pulp boards and dispersing in water to form a plant fiber pulp; (5) preparing a filler solution that contains 10 wt % of a mineral filler; (6) mixing, rolling and drying to obtain the reconstituted tobacco.

Abstract: A method of preparing an aramid paper coated with aramid nanofibers includes the following steps: (1) mixing a meta-aramid fibrid slurry and a chopped meta-aramid fiber slurry, filtering, pressing and drying to obtain a meta-aramid paper; (2) mixing potassium hydroxide, deionized water, dimethyl sulfoxide, and para-aramid nanofibers in a container, and stirring to obtain a para-aramid nanofiber coating solution; and (3) applying the para-aramid nanofiber coating solution to a first side of the meta-aramid paper, washing with deionized water, and drying; applying the para-aramid nanofiber coating solution to a second side of the meta-aramid paper, washing with deionized water, and drying; and hot pressing to obtain the aramid paper coated with aramid nanofibers.

Abstract: Disclosed of the present application is relation to deep learning based voice data processing. The voice data to be detected is converted into target text data based on a voice recognition model so that the keyword text corresponding to the predetermined target voice keyword can be converted. Then, the data is matched with the target text data to determine whether the voice data to be detected includes the target voice keyword based on the matching result. Thus, because the voice recognition model is obtained by deep learning based on the obtained voice recognition data training set, it can obtain high-precision target text data, thereby improving the accuracy of subsequent matching. The problem of low accuracy of detecting voice data for keyword detection can therefor be solved.

Abstract: The present disclosure provides a display panel, a method for manufacturing the same, and a display device. The insulation layer is provided above the first conductive electrodes in the bonding area of the display panel, the insulation layer covers the first conductive electrodes, and the insulation layer is capable of being pierced by ACF particles. When the display panel is bound to an FPC by an ACF, second conductive electrodes on the FPC can be electrically coupled to the first conductive electrodes on the display panel through the ACF particles, thereby achieving the bonding connection between the display panel and the FPC, even if a conductive foreign object falls into the area where the first conductive electrodes are located, short circuit cannot be caused, thereby improving the product yield.

Abstract: Disclosed are a display panel and a display device. The display panel includes a substrate and multiple organic light-emitting units located in the display region on a first side of the substrate, where an area of at least one of organic light-emitting units in the second display region is smaller than an area of each of organic light-emitting units with a same light-emitting color in the first display region, and/or, density of the organic light-emitting units in the first display region is greater than density of the organic light-emitting units in the second display region. The second display region includes at least one quantum dot light-emitting unit, which does not overlap the organic light-emitting units; and each of the at least one quantum dot light-emitting unit emits light of a same color as at least one of the organic light-emitting units located in the second display region.

Abstract: An optical adjustment method and an optical adjustment device for a display panel, and a display device are provided. The optical adjustment method includes: displaying N groups of testing images sequentially on the display panel, each group of testing images including M images distributed at different display regions of the display panel, each image corresponding to one to-be-adjusted reference color, N being an integer greater than or equal to 1, M being an integer greater than or equal to 1; and when each group of testing images are displayed on the display panel, detecting, by an optical detection unit, optical parameters of the M images in the group of testing images simultaneously, and performing optical adjustment on the display panel in accordance with the optical parameters.

Abstract: A display substrate includes a base, a plurality of light-emitting devices disposed on the base, an encapsulation layer disposed on a light-emitting side of the plurality of light-emitting devices away from the base, and at least one photosensitive sensor disposed on a surface of the encapsulation layer away from the base. Each of the at least one photosensitive sensor is configured to collect optical signals for texture recognition.

Abstract: A display substrate and a manufacturing method thereof, a display panel, a display motherboard and a testing method thereof, and a display device are provided in the present disclosure. The display substrate is provided with a display region and at least a wiring region outside the display region. The display substrate includes a base substrate and a plurality of connection structures on the base substrate. The plurality of connection structures are connected to a plurality of edge signal terminals through a control module. All of the plurality of connection structures, the plurality of edge signal terminals and the control module are located in the wiring region. The plurality of edge signal terminals are located on an edge of the wiring region. The control circuit is configured to control the connection or disconnection between the plurality of signal terminals and the plurality of connection structures.

Abstract: A display module Gamma correction method includes: obtaining corrected Gamma register values corresponding to binding points of a grayscale by correcting register values of s binding points selected from a set of m binding points of the grayscale based on a group of initial Gamma register values that correspond to the m binding points and a target Gamma curve; selecting, from x sets of alternate Gamma register values wherein each set corresponds to m binding points and the initial Gamma register values, a set of Gamma register values used for Gamma correction of the display module(s) as reference register values; and; and correcting register values of remaining m?s binding points based on the reference Gamma register values and the target Gamma curve to obtain a set of target Gamma register values corresponding to the m binding points, wherein s, m and x are all integers greater than one.