Currently, how to deal with the massive garbage produced by various human activities is a hot topic all around the world. In this paper, a preliminary and essential step is to classify the garbage into different categories. However, the mainstream waste classification mode relies heavily on manual work, which consumes a lot of labor and is very inefficient. With the rapid development of deep learning, convolutional neural networks (CNN) have been successfully applied to various application fields. Therefore, some researchers have directly adopted CNNs to classify garbage through their images. However, compared with other images, the garbage images have their own characteristics (such as inter-class similarity, intra-class variance and complex background). Thus, neglecting these characteristics would impair the classification accuracy of CNN. To overcome the limitations of existing garbage image classification methods, a Depth-wise Separable Convolution Attention Module (DSCAM) is proposed in this paper. In DSCAM, the inherent relationships of channels and spatial positions in garbage image features are captured by two attention modules with depth-wise separable convolutions, so that our method could only focus on important information and ignore the interference. Moreover, we also adopt a residual network as the backbone of DSCAM to enhance its discriminative ability. We conduct the experiments on five garbage datasets. The experimental results demonstrate that the proposed method could effectively classify the garbage images and that it outperforms some classical methods.
Background: Immune-related long non-coding RNAs (irlncRNAs) might remodel the tumor immune microenvironment by changing the inherent properties of tumor cells and the expression of immune genes, which have been used to predict the efficacy of immunotherapy and the prognosis of various tumors. However, the value of irlncRNAs in breast cancer (BRCA) remains unclear.Materials and Methods: Initially, transcriptome data and immune-related gene sets were downloaded from The Cancer Genome Atlas (TCGA) database. The irlncRNAs were extracted from the Immunology Database and Analysis Portal (ImmPort) database. Differently expressed irlncRNAs (DEirlncRNAs) were further identified by utilizing the limma R package. Then, univariate and multivariate Cox regression analyses were conducted to select the DEirlncRNAs associated with the prognosis of BRCA patients. In addition, the univariate and least absolute shrinkage and selection operator (LASSO) Cox regression analyses were performed to determine the DEirlncRNA pairs with the independent prediction capability of prognosis in BRCA patients. Finally, the chosen DEirlncRNA pair would be evaluated in terms of survival time, clinicopathological characteristics, tumor-infiltrating immune cells, immune checkpoints (ICs), signaling pathways, and potential small-molecule drugs.Results: A total of 21 DEirlncRNA pairs were extracted, and among them, lncRNA MIR4435-2HG and lncRNA U62317.1 were chosen to establish a risk signature that served as an independent prognostic biomarker in BRCA patients. Patients in the high-risk group had a worse prognosis than those in the low-risk group, and they also had an abundance of infiltration of CD4+ T and CD8+ T cells to enhance the immune response to tumor cells. Furthermore, the risk signature showed a strong correlation with ICs, signaling pathways, and potential small-molecule drugs.Conclusion: Our research revealed that the risk signature independent of specific DEirlncRNA pair expression was closely associated with the prognosis and tumor immune microenvironment in BRCA patients and had the potential to function as an independent prognostic biomarker and a predictor of immunotherapy for BRCA patients, which would provide new insights for BRCA accurate treatment.
Gear shaping is a widely applied technology to produce spur gears. Generally, the pinion cutter and the gear workpiece rotate uniformly with a given gear ratio during the conventional gear shaping process, which can cause a large variation of the cutting area per stroke in cutting tooth spaces. It makes the cutting force less than the rated capacity of the gear shaper in most cutting strokes and thus reduces the process efficiency. To overcome such a shortage, a new spur gear shaping method is proposed in this article, in which the cutting area per stroke is homogenized to a target value through optimizing the circular feed rate. The new method can enhance process efficiency by keeping the cutting force equivalent to the rated capacity of the gear shaper. The specific algorithm includes a number of aspects: cutting area calculation, gear profile generation, cutting area analysis of conventional gear shaping, and cutting area homogenization. Additionally, the new spur gear shaping method is demonstrated and validated using a VERICUT simulation. From the simulation results, it is found that the process efficiency is improved up to 40% via the efficient gear shaping because of the reduced number of shaping strokes. Hence, the new spur gear shaping method is applicable for computer numerical control gear shapers to improve the process efficiency significantly without any additional hardware changes.
Artificial intelligence is one of the most popular topics in computer science. Convolutional neural network (CNN), which is an important artificial intelligence deep learning model, has been widely used in many fields. However, training a CNN requires a large amount of labeled data to achieve a good performance but labeling data is a time-consuming and laborious work. Since active learning can effectively reduce the labeling effort, we propose a new intelligent active learning method for deep learning, which is called multi-view active learning based on double-branch network (MALDB). Different from most existing active learning methods, our proposed MALDB first integrates two Bayesian convolutional neural networks (BCNNs) with different structures as two branches of a classifier to learn the effective features for each sample. Then, MALDB performs data analysis on unlabeled dataset and queries the useful unlabeled samples based on different characteristics of two branches to iteratively expand the training dataset and improve the performance of classifier. Finally, MALDB combines multiple level information from multiple hidden layers of BCNNs to further improve the stability of sample selection. The experiments are conducted on five extensively used datasets, Fashion-MNIST, Cifar-10, SVHN, Scene-15 and UIUC-Sports, the experimental results demonstrate the validity of our proposed MALDB.
In this paper, under the condition of the existing processing and detecting technology, we used coordinate measuring machine (CMM) to measure the contour of cycloid gear. In order to improve the accuracy of measured data, we used Euclidean distance and Laida criterion to preprocess measured data. After preprocessing the measured data of the cycloid gear, we used ellipse fitting based on least square approach to fit the contour data points of the cycloid gear, and used the determination coefficient method to evaluate the goodness of fitting. According to the result of the curve fitting of cycloid gear, we used Matlab to analyse and calculate the pitch errors of cycloid gear, which provides data for the subsequent matching of parts combination of the best RV reducer with genetic algorithm.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.