2results about How to "Comprehensive measurement" patented technology

The invention relates to the technical field of wire coating thickness measurement by an electron microscope, in particular to a wire coating thickness measurement and uniformity evaluation method based on image processing. Comprising the following steps: vertically cutting a wire sample and preparing an observation surface; acquiring a clear image of the section of the coating by using a scanning electron microscope back scattering electron mode; carrying out binarization processing and interface repair on the image, and extracting a plating layer area; carrying out equidistant segmentation on the regional image along the extension direction of the plating layer, measuring the area of each segmented region, and calculating the local thickness; based on all local thickness data, an average thickness and a standard deviation are calculated. According to the method, image segmentation and statistics are combined, so that the average thickness can be obtained more accurately, the standard deviation of the coating uniformity can be directly output in single measurement, and comprehensive and objective data support is provided for product quality control and process optimization. The method is suitable for various materials and coating systems, and has the advantages of high scientificity, high repeatability and comprehensive information.

The application discloses a kind of multi-modal face anti-fraud methods for domain generalization, it is related to face recognition technical field, multiple source domains under multi-modal face image are collected, and data set is constructed;Face anti-fraud model is constructed, and multiple modal image coding module is configured for each source domain, and multiple modal fusion features are generated, and classification head is configured for each source domain, and the prediction result of corresponding domain is output according to multiple modal fusion features, domain relationship extraction module is configured, the inter-domain relationship weight between each source domain is extracted, and the prediction result of each source domain is fused according to inter-domain relationship weight, and the final prediction result is obtained;The model is trained, including intra-domain modal level optimization and domain level optimization.The application introduces the center difference expansion convolution module of adaptive receptive field, combines inter-domain relationship weighted multi-head fusion framework, solves the problem that multi-modal feature extraction is insufficient in the prior art, and the cross-domain generalization ability is weak, realizes complex scene, unknown domain under accurate face anti-fraud detection.