6results about How to "Retain structure" patented technology

This invention discloses a virtual immunohistochemical staining method and system based on relation generative adversarial networks (GANs) and sequence slide data, comprising the following steps: Step S1, constructing a sequence slide dataset, wherein each H&E-stained slide is followed by three consecutive Ki67-stained slides; Step S2, constructing a relation generative adversarial network (GAN) model, the model comprising two ResNet generators and two relation discriminators, used to train and evaluate the credibility of the sequence relationship between image pairs; Step S3, performing two-stage training on the relation generative adversarial network model, namely, basic style pre-training and Z-axis alignment relation adaptation. This invention fundamentally solves the Z-axis alignment problem, resulting in higher image fidelity and pathological accuracy, significantly improving clinical application value, and providing new research resources and evaluation paradigms.

This utility model belongs to the field of UAV power motor technology, and discloses an external rotor structure, a brushless motor, and a UAV. The external rotor structure is composed of a shaft, a fan, a housing, a cage, and magnets. The cage is ring-shaped and fixed inside the housing, with multiple protrusions extending downwards on its lower side to hold the magnets in place. The fan has a top plate on its upper side and fan blades on its lower side. The top plate of the fan is fixed to the top surface of the housing, and the shaft is fixed inside the housing via a shaft mounting part. This external rotor structure achieves the installation of the shaft and magnets through the assembly structure of the fan, housing, and cage. Furthermore, the structure of the fan, housing, and cage is relatively simple, allowing the cage and fan to be manufactured using injection molding, and the housing to be manufactured using deep drawing, effectively reducing manufacturing costs and improving the production efficiency of the UAV. At the same time, the fan structure is retained to ensure the motor's heat dissipation performance.

This invention discloses an enhanced antistatic polyetheretherketone (PEEK) composite material and its preparation method. The composite material is prepared from the following raw materials in parts by weight: 60-85 parts of PEEK resin, 5-15 parts of a first synergistic component, 3-10 parts of a second synergistic component, 2-8 parts of a third synergistic component, and 1-5 parts of a high-temperature processing aid. The first synergistic component is a mixture of a metal-based conductive filler and a high-temperature resistant antistatic synergist; the second synergistic component is ionic liquid-functionalized modified graphene; the third synergistic component is a composite of carbon nanotubes and conductive nanofibers; and the high-temperature processing aid includes a high-temperature lubricant and a high-temperature antioxidant. The preparation method includes steps such as preparing various premixes, melt blending, extrusion granulation, and supercritical fluid in-mold foaming. This invention constructs a stable three-dimensional conductive network through the synergistic effect of multi-scale conductive fillers, giving the composite material excellent antistatic properties.

The invention discloses a digital holographic microscopic imaging coherent noise suppression network model and method based on deep learning. The network model adopts a double-branch encoder-single decoder structure, and a double-branch encoder extracts local details and global noise features of an image in parallel; the innovative double-branch intensive attention fusion module fuses multi-scale features through an adaptive weighting and enhancement mechanism; and the decoder integrates information through jump connection and reconstructs a clear image. The training method generates a high-fidelity data set based on a physical simulation system. According to the method, speckle noise and parasitic fringes can be efficiently suppressed, the phase details of the object are effectively reserved while the peak signal-to-noise ratio and the structural similarity index are remarkably improved, and the method has excellent generalization ability and calculation efficiency and is suitable for real-time high-quality imaging of a digital holographic microscopy system.

This application discloses a composite high-flowability, high-conductivity conductive masterbatch and its preparation method, relating to the technical field of conductive materials. The masterbatch uses a fiber-forming polymer as a matrix, compounded with conductive carbon black, graphene, silver nanosheets, carbon nanotubes, and modified MXene nanosheets, along with coupling agents, dispersants, antioxidants, and lubricants. The modified MXene nanosheets are modified with an amino-functionalized modifier and a polysaccharide mixture of sodium alginate and agar. Amino grafting improves interfacial compatibility, while polysaccharide coating inhibits filler agglomeration and oxidation, optimizing the dispersion of the conductive filler in the matrix. This masterbatch constructs a multi-level continuous conductive network, significantly improving melt flowability, conductive stability, and mechanical properties, effectively solving the problems of easy agglomeration, poor flowability, and insufficient spinnability of traditional conductive masterbatches.

This invention proposes a fuzzing method and system for WebAssembly engines, relating to the field of software testing. It involves acquiring a large number of WebAssembly binary samples as initial seed samples; parsing the initial seed samples into an attribute abstract syntax tree (API) with node position and type information; applying a mutation method to the API using a type-aware mutation strategy and a structure-aware instruction mutation method to generate a new API with valid structure and type; pre-validating the mutated API and quickly correcting invalid node positions; and generating WebAssembly binary samples from the pre-validated API, which are then input into the target engine for fuzzing. This invention, by using an API based on attribute abstract syntax trees and a WebAssembly mutation method with pre-validation and rapid correction, more effectively generates WebAssembly module mutations with correct binary format, legality, and compliance with security constraints, enabling efficient fuzzing of the WebAssembly engine.