Abstract: The present application discloses a method for detecting cigarette appearance defects based on variational Bayesian inference. In a cigarette appearance defect detection scenario, in order to solve the problem that a current point estimation-based machine learning algorithm leads to an overconfidence decision in a data scarcity area, the present application proposes a variational inference-based Bayesian method to improve a SSD backbone network. The present application can improve the accuracy of defect detection without losing the processing speed, not only saving a large amount of manpower and material resources, but also effectively providing reliable technical support for finding the root causes of defect types. Thus, it achieves intelligent quality risk management and control in a cigarette production workshop.

Abstract: An example includes: generating component carbon footprint data of a component, wherein the data comprises data of a material for manufacturing the component and addon component carbon footprint data generated in the production and processing of the component; storing an encryption result of the addon data in a first blockchain to trigger generation and storage of encrypted component carbon footprint data based on the encryption result of the addon component carbon footprint data and pre-stored first encrypted material carbon footprint data; acquiring a first verifiable credential for average footprint data of a batch of components containing the component, wherein a public key for verifying the credential is stored in a second blockchain; and sending the component data and the presentation generated to the manufacturer, so the manufacturer verifies the presentation and the footprint data based on the first encrypted component data and the public key.

Abstract: Flexible optical fiber ribbon includes multiple optical fibers. The multiple optical fibers are optical fibers arranged side by side, and multiple bonded portions are arranged between adjacent optical fibers and distributed intermittently along the longitudinal direction, the multiple bonded portions being made of the same material; each optical fiber has a ink layer at the underlying outermost layer and a transparent coating layer at the outermost layer; the transparent coating layers of the multiple optical fibers are made of the same material, and the transparent coating layers are continuous in the axial direction and continuous in the circumferential direction; the colors and/or shapes of the ink layers of optical fibers are observed through the transparent coating layers, and there is at least one optical fiber, among the multiple optical fibers, of which the color and/or shape of the ink layer thereof is different from that of another optical fiber among the multiple optical fibers.

Abstract: An integrated plugging and perforating operation system includes an operation truck, a wellhead equipment transport frame a centralized control room and wellhead blowout preventers. A mechanical mechanism and a power mechanism are arranged on a chassis of the operation truck. The mechanical mechanism includes a turntable assembly, where a telescopic arm assembly is installed on the turntable assembly, a mechanical arm assembly for logging operation is installed at an operating end of the telescopic arm assembly, and a logging winch assembly and a follow-up winch assembly are arranged at a counterweight end of the telescopic arm assembly to form a counter weight of the telescopic arm assembly. The power mechanism is configured for providing power output to the mechanical mechanism.

Abstract: A flexible optical fiber ribbon includes multiple optical fibers arranged side by side, with intermittent bonding parts between adjacent optical fibers. The transmittance of the intermittent bonding parts for a wavelength band of a detection light is 50% or lower. The optical fiber coating is covered with a color layer, and the color layer is covered with a light-transmitting layer. The transmittance of the color layer for the wavelength band of the detection light is 30% or lower, and the transmittance of the light-transmitting layer for the wavelength band of the detection light is 85% or higher. The bonding parts with qualified bonding performance cover the light-transmitting layer in the detection light direction.

Abstract: An olefin polymerization catalyst component, a catalyst system, a prepolymerization catalyst composition and an olefin polymerization method are provided. The catalyst component contains magnesium, titanium, halogen, and an internal electron donor. The internal electron donor includes a compound A, a compound B and a compound C. The compound A is selected from a hydroxybenzoyl compound represented by formula (I), the compound B is selected from one or more of an ester compound other than the compound A and an ether compound, and the compound C is selected from an alkoxysilane represented by formula RnSi(OR6)4-n. A content of the compound A may be zero. The catalyst system contains the catalyst component and a co-catalyst component or a reaction product of the two, or a prepolymerization catalyst composition prepared by further prepolymerization is used for olefin copolymerization.

Abstract: An example includes: generating component carbon footprint data of a component, wherein the data comprises data of a material for manufacturing the component and addon component carbon footprint data generated in the production and processing of the component; storing an encryption result of the addon data in a first blockchain to trigger generation and storage of encrypted component carbon footprint data based on the encryption result of the addon component carbon footprint data and pre-stored first encrypted material carbon footprint data; acquiring a first verifiable credential for average footprint data of a batch of components containing the component, wherein a public key for verifying the credential is stored in a second blockchain; and sending the component data and the presentation generated to the manufacturer, so the manufacturer verifies the presentation and the footprint data based on the first encrypted component data and the public key.

Abstract: The present application discloses a method for detecting cigarette appearance defects based on variational Bayesian inference. In a cigarette appearance defect detection scenario, in order to solve the problem that a current point estimation-based machine learning algorithm leads to an overconfidence decision in a data scarcity area, the present application proposes a variational inference-based Bayesian method to improve a SSD backbone network, preferably including calculating a real posterior probability by the variational inference method, replacing a point estimation mechanism of a convolution layer weight in the original backbone network by using a form of probability distribution, then detecting the improved extracted features by an SSD target detection algorithm, and furthermore, accurately determining whether the appearance of the cigarette is defective and the type and location of the defect.

Abstract: An integrated plugging and perforating operation system is provided, including an operation truck, a wellhead equipment pry, and a centralized control room. A mechanical mechanism and a power mechanism are arranged on a chassis of the operation truck. The mechanical mechanism includes a turntable assembly, where a telescopic arm assembly is installed on the turntable assembly, a mechanical arm assembly for logging operation is installed at an operating end of the telescopic arm assembly, and a logging winch assembly and a follow-up winch assembly are arranged at a counterweight end of the telescopic arm assembly to form a counter weight of the telescopic arm assembly. The power mechanism is configured for providing power output to the mechanical mechanism.

Abstract: Methods and systems for treatment prioritization using a neural network model are provided. Training the model includes generating an enriched genetic knowledge graph with edges characterizing relationships between genetic variants, genes, diseases, treatments and symptoms, embedding the graph vertices in genetic knowledge vectors, receiving training patient feature value sets, generating a training dataset of training patient vectors labelled with target values comprising treatment efficacity by scaling the genetic knowledge vectors by the patient feature value sets, and training the model to process an input patient vector and generate a predicted efficacity of the at least one treatment option.

Abstract: Methods and systems for phenotyping based on a neural network model are provided. Training the model includes generating a dataset of individual-specific variant graphs associated representing dependencies among variants in individuals' genomes based on biomedical domain knowledge labelled according to target phenotype values to be predicted, and training the model using the training dataset to generate a phenotype prediction from an input individual-specific variant graph. Measure the contribution of a genetic variant to a phenotype of a target individual includes generating an individual-specific variant graph for the individual, modifying the individual-specific variant graph to remove a genetic variant, providing the initial and the modified individual-specific variant graphs as input to the phenotyping neural network model to generate two phenotype predictions, and calculating a difference between the two predictions, providing a measure of the contribution of the genetic variant to the phenotype.

Abstract: An authentication method and apparatus are provided. In an embodiment, the authentication method includes: receiving, by a first network node, an enrollment request from a second network node; obtaining, by the first network node, position information of the second network node; and authenticating, by the first network node, the second network node according to the obtained position information of the second network node. A flexible simple authentication solution is provided, having low deployment costs.

Abstract: An autonomous driving control method for a vehicle may include: obtaining a road surface image from a sensor of the vehicle; detecting road surface damage from the road surface image using an image detection algorithm; determining a damage degree of the detected road surface damage and a damage distance of the road surface damage from the vehicle; and adjusting an autonomous driving strategy for the vehicle based on the damage degree and the damage distance. The autonomous driving control method for the vehicle can detect potholes or cracks in front of road accurately in real time and adjust the autonomous driving strategy based on a detection result, ensuring safety of passenger in the vehicle and improving passenger experience during the autonomous driving process.

Abstract: A structure for performing analysis includes a first opening formed on a back side of a substrate and passing through the substrate, a second opening connected with a bottom of the first opening and penetrating into a first dielectric layer formed on a front side of the substrate, a first conductive layer formed on a sidewall of the second opening and a contact element in the first dielectric layer, and a second conductive layer formed on a second dielectric layer. The first conductive layer contacts the second conductive layer electrically.

Abstract: A catalyst system for olefin polymerization contains a main catalyst and a cocatalyst. The cocatalyst contains a twelve-membered ring compound represented by formula (M). The catalyst system is suitable for preparing polypropylene products having high stereoregularity and low ash, and can regulate the melt index of the products within a wide range by adjusting the amount of hydrogenation. It is also suitable for copolymerization systems to improve the copolymerization yield.

Abstract: A magnesium-based solid, by means of determination based on a nitrogen adsorption method, has a multimodal pore distribution and a specific surface area of not less than 50 m2/g, and the pore size distribution of the solid is in a range of 1 nm to 300 nm. There is at least one peak within a pore size range of less than 10 nm, and there is at least another peak within a pore size range of not less than 10 nm. A catalyst is formed using the solid catalyst component is used for propylene polymerization.

Abstract: A structure for performing analysis includes a first opening formed on a back side of a substrate and passing through the substrate, a second opening connected with a bottom of the first opening and penetrating into a first dielectric layer formed on a front side of the substrate, a first conductive layer formed on a sidewall of the second opening and a contact element in the first dielectric layer, and a second conductive layer formed on a second dielectric layer. The first conductive layer contacts the second conductive layer electrically.

Abstract: A network traffic sending method and apparatus, and a hybrid honeypot system are provided. The method includes receiving a first attack traffic flow; determining that a request type of the first attack traffic flow is a first request type and determining maturity of a virtual honeypot model for the first request type. Upon the maturity of the virtual honeypot model for the first request type being higher than a threshold set for the first request type, the method includes forwarding the first attack traffic flow to a virtual honeypot using the model, or forwarding the first attack traffic flow to a virtual honeypot using the model and a physical honeypot. Otherwise, it includes forwarding the first attack traffic flow to a physical honeypot. Therefore, a virtual honeypot using a virtual honeypot model of relatively high maturity is used to respond to an attack traffic flow.

Abstract: A device for separating light flaky cut rolled stems (CRSs) by a throwing roller is provided. The device includes a vertically provided separation bin, and a CRS guide device and a CRS distribution plate which are arranged in the separation bin. The present disclosure further provides a method for separating light flaky CRSs by using the device. The present disclosure achieves full separation of finished CRSs from light flaky CRSs. The device of the present disclosure is designed for the first time, and can be directly connected to a pneumatic CRS feeding and discharging section of a production line. The device of the present disclosure is compact and suitable for industrial production.

Abstract: A device for separating light flaky cut rolled stems (CRSs) by a throwing roller is provided. The device includes a vertically provided separation bin, and a CRS guide device and a CRS distribution plate which are arranged in the separation bin. The present disclosure further provides a method for separating light flaky CRSs by using the device. The present disclosure achieves full separation of finished CRSs from light flaky CRSs. The device of the present disclosure is designed for the first time, and can be directly connected to a pneumatic CRS feeding and discharging section of a production line. The device of the present disclosure is compact and suitable for industrial production.