Search

Abstract Lightweight deep learning models for face recognition are becoming increasingly crucial for deployment on resource‐constrained devices such as embedded systems or mobile devices. This paper presents a highly efficient and compact deep learning (DL) model that achieves state‐of‐the‐art performance on various face recognition benchmarks. The developed DL model employs one‐ or few‐shot learning to obtain effective feature embeddings and draws inspiration from FaceNet with significant refinements to achieve a memory size of only 3.5 MB—about 30 times smaller than FaceNet—while maintaining high accuracy and real‐time performance. The study demonstrates the model's effectiveness through extensive experiments, which include testing on public datasets and the model's ability to recognize occluded faces in uncontrolled environments using grayscale input images. Compared to the state‐of‐the‐art lightweight models, the proposed model requires fewer FLOPs (0.06G), has a smaller number of parameters (1.2 M), and occupies a smaller model size (3.5 MB) while achieving a competitive level of recognition accuracy and real‐time performance. The results show that the model is well‐suited for deployment in embedded domains, including live entrance security checks, driver authorization, and in‐class attendance systems. The entire code of FN8 is available on GitHub.

A fully connected layer is essential for a CNN, i.e., convolutional neural network, which has been shown to be successful in classifying images in several related applications. A CNN begins with convolution and pooling operations for decomposing an input image into features. The result of this process is then fed into a fully connected neural network, driving the final classification decision for the input image. However, it has been found that the learned feature maps in a CNN are sometimes not good enough for being fed into the fully connected layers to get good classification results. In this article, a visual attention learning module is proposed to enhance the classification capability of the fully connected layers in a CNN. By learning better feature maps to emphasize salient regions and weaken meaningless regions, better classification performance can be obtained with integrating the proposed module into the fully connected layers. The proposed visual attention learning module can be imposed on any existed CNN-based image classification models to achieve incremental improvements with negligible overhead. Based on our experiments, the proposed method achieves the top-1 accuracies of 95.32%, 92.73%, and 66.50% on average, respectively, obtained on our collected Underwater Fish dataset, the public Animals-10 dataset, and the public Stanford Cars dataset.

Abstract The decomposition of Chinese characters is difficult and has been rarely investigated in the literature. In this paper, we propose a novel non-negative matrix factorization (NMF) technique to decompose a Chinese character into several graphical components without considering the strokes of the character or any semantic or phonetic properties of the components. Chinese characters can usually be represented as binary images. However, traditional NMF is only suitable for representing general gray-level or color images. To decompose a binary image using NMF, we force all of the elements of the two matrices (obtained by factorizing the binary image/matrix to be decomposed) as close to 0 or 1 as possible. As a result, a Chinese character can be efficiently decomposed into several components, where each component is semantically unreadable. Moreover, our NMF-based Chinese character decomposition method is suitable for applications in visual secret sharing by distributing the shares (different character components) among multiple parties, so that only when the parties are taken together with their respective shares can the secret (the original Chinese character(s)) be reconstructed. Experimental results have verified the decomposition performance and the feasibility of the proposed method.

Marsdeniae tenacissimae Caulis is a traditional Chinese medicine, named Tongguanteng (TGT), that is often used for the adjuvant treatment of cancer. In our previous study, we reported that an ethyl acetate extract of TGT had inhibitory effects against adenocarcinoma A549 cells growth. To identify the components of TGT with anti-tumor activity and to elucidate their underlying mechanisms of action, we developed a technique for isolating compounds, which was then followed by cytotoxicity screening, network pharmacology analysis, and cellular and molecular experiments. We isolated a total of 19 compounds from a TGT ethyl acetate extract. Two novel steroidal saponins were assessed using an ultra-performance liquid chromatography-photodiode array coupled with quadrupole time-of-flight mass (UPLC-ESI-Q/TOF-MS). Then, we screened these constituents for anti-cancer activity against non-small cell lung cancer (NSCLC) in vitro and obtained six target compounds. Furthermore, a compound-target-pathway network of these six bioactive ingredients was constructed to elucidate the potential pathways that controlled anticancer effects. Approximately 205 putative targets that were associated with TGT, as well as 270 putative targets that were related to NSCLC, were obtained from online databases and target prediction software. Protein–protein interaction networks for drugs as well as disease putative targets were generated, and 18 candidate targets were detected based on topological features. In addition, pathway enrichment analysis was performed to identify related pathways, including PI3K/AKT, VEGF, and EGFR tyrosine kinase inhibitor resistance, which are all related to metabolic processes and intrinsic apoptotic pathways involving reactive oxygen species (ROS). Then, various cellular experiments were conducted to validate drug-target mechanisms that had been predicted using network pharmacology analysis. The experimental results showed the four C21 steroidal saponins could upregulate Bax and downregulate Bcl-2 expression, thereby changing the mitochondrial membrane potential, producing ROS, and releasing cytochrome C, which finally activated caspase-3, caspase-9, and caspase-8, all of which induced apoptosis in A549 cells. In addition, these components also downregulated the expression of MMP-2 and MMP-9 proteins, further weakening their degradation of extracellular matrix components and type IV collagen, and inhibiting the migration and invasion of A549 cells. Our study elucidated the chemical composition and underlying anti-tumor mechanism of TGT, which may be utilized in the treatment of lung cancer.