A Semi-Supervised Reduced-Space Method for Hyperspectral Imaging Segmentation

Improving the Performance Properties of Eutectoid Steel Products by a Complex Effect

This study focuses on the assessment of possible hypereutectoid steel carbide mesh crushing. It is used for tools production, including forming rolls of various diameters, with modification and cyclic heat treatment methods. For steel containing 1.79-1.83% C, we studied the effect of 0.35-1.15% Si on the possible crushing of the cementite mesh within crystallization by introducing modifiers Ti, V, N, as well as simultaneously modifying V with N and Ti with N. The obtained castings of Ø200 mm, 400 mm high were cut into discs, from which we made samples for tests on wear, determining mechanical properties, thermal resistance, and susceptibility to brittle fracture. The assessment was performed in the as-cast and after double and triple normalizing and annealing with drawback. With additional fans blowing, we changed the cooling rate from 25 °C/h to 100-150 °C/h. We performed the microstructure analyses using traditional metallographic, micro-X-ray spectral analyses, and also used the segmentation process based on 2D image markers. It was found that the as-cast modifying additives infusion is insufficient for carbide mesh crushing. It can be made by multi-stage normalizing with accelerated cool-down for products up to 600 mm in diameter to cycle temperatures above the steel transfer from a plastic to elastic state (above 450 °C).

Hyperspectral Image Classification Based on Spectral Multiscale Convolutional Neural Network

In recent years, convolutional neural networks (CNNs) have been widely used for hyperspectral image classification, which show good performance. Compared with using sufficient training samples for classification, the classification accuracy of hyperspectral images is easily affected by a small number of samples. Moreover, although CNNs can effectively classify hyperspectral images, due to the rich spatial and spectral information of hyperspectral images, the efficiency of feature extraction still needs to be further improved. In order to solve these problems, a spatial–spectral attention fusion network using four branch multiscale block (FBMB) to extract spectral features and 3D-Softpool to extract spatial features is proposed. The network consists of three main parts. These three parts are connected in turn to fully extract the features of hyperspectral images. In the first part, four different branches are used to fully extract spectral features. The convolution kernel size of each branch is different. Spectral attention block is adopted behind each branch. In the second part, the spectral features are reused through dense connection blocks, and then the spectral attention module is utilized to refine the extracted spectral features. In the third part, it mainly extracts spatial features. The DenseNet module and spatial attention block jointly extract spatial features. The spatial features are fused with the previously extracted spectral features. Experiments are carried out on four commonly used hyperspectral data sets. The experimental results show that the proposed method has better classification performance than some existing classification methods when using a small number of training samples.