RESUMO
Image clustering is a research hotspot in machine learning and computer vision. Existing graph-based semi-supervised deep clustering methods suffer from three problems: 1) because clustering uses only high-level features, the detailed information contained in shallow-level features is ignored; 2) most feature extraction networks employ the step odd convolutional kernel, which results in an uneven distribution of receptive field intensity; and 3) because the adjacency matrix is precomputed and fixed, it cannot adapt to changes in the relationship between samples. To solve the above problems, we propose a novel graph-based semi-supervised deep clustering method for image clustering. First, the parity cross-convolutional feature extraction and fusion module is used to extract high-quality image features. Then, the clustering constraint layer is designed to improve the clustering efficiency. And, the output layer is customized to achieve unsupervised regularization training. Finally, the adjacency matrix is inferred by actual network prediction. A graph-based regularization method is adopted for unsupervised training networks. Experimental results show that our method significantly outperforms state-of-the-art methods on USPS, MNIST, street view house numbers (SVHN), and fashion MNIST (FMNIST) datasets in terms of ACC, normalized mutual information (NMI), and ARI.
RESUMO
Peroxidase (POD) Nanozyme-based hydrogen peroxide (H2 O2 ) detection is popular, but hardly adapt to high concentration of H2 O2 owing to narrow linear range (LR) and low LR maximum. Here, a solution of combining POD and catalase (CAT) is raised to expand the LR of H2 O2 assay via decomposing part of H2 O2 . As a proof of concept, a cascade enzyme system (rGRC) is constructed by integrating ruthenium nanoparticles (RuNPs), CAT and graphene together. The rGRC-based sensor does perform an expanded LR and higher LR maximum for H2 O2 detection. Meanwhile, it is confirmed that LR expansion is closely associated with apparent Km of rGRC, which is determined by the relative enzyme activity between CAT and POD both in theory and in experiment. At last, rGRC is successfully used to detect high concentration of H2 O2 (up to 10 mm) in contact lens care solution, which performs higher assay accuracy (close to 100% recovery at 10 mm of H2 O2 ) than traditional POD nanozymes. This study brings up a kind of POD/CAT cascade enzyme system and provides a new concept for accurate and facile H2 O2 detection. Additionally, it replenishes a new enzyme-substrate model of achieving the same pattern with competitive inhibition in enzyme reactions.
