In hyperspectral image (HSI) classification, each pixel sample is assigned to a land-cover category. In the recent past, convolutional neural network (CNN)-based HSI classification methods have ...greatly improved performance due to their superior ability to represent features. However, these methods have limited ability to obtain deep semantic features, and as the layer's number increases, computational costs rise significantly. The transformer framework can represent high-level semantic features well. In this article, a spectral-spatial feature tokenization transformer (SSFTT) method is proposed to capture spectral-spatial features and high-level semantic features. First, a spectral-spatial feature extraction module is built to extract low-level features. This module is composed of a 3-D convolution layer and a 2-D convolution layer, which are used to extract the shallow spectral and spatial features. Second, a Gaussian weighted feature tokenizer is introduced for features transformation. Third, the transformed features are input into the transformer encoder module for feature representation and learning. Finally, a linear layer is used to identify the first learnable token to obtain the sample label. Using three standard datasets, experimental analysis confirms that the computation time is less than other deep learning methods and the performance of the classification outperforms several current state-of-the-art methods. The code of this work is available at https://github.com/zgr6010/HSI_SSFTT for the sake of reproducibility.
This paper presents a new approach for hyperspectral image classification exploiting spectral-spatial information. Under the maximum a posteriori framework, we propose a supervised classification ...model which includes a spectral data fidelity term and a spatially adaptive Markov random field (MRF) prior in the hidden field. The data fidelity term adopted in this paper is learned from the sparse multinomial logistic regression (SMLR) classifier, while the spatially adaptive MRF prior is modeled by a spatially adaptive total variation (SpATV) regularization to enforce a spatially smooth classifier. To further improve the classification accuracy, the true labels of training samples are fixed as an additional constraint in the proposed model. Thus, our model takes full advantage of exploiting the spatial and contextual information present in the hyperspectral image. An efficient hyperspectral image classification algorithm, named SMLR-SpATV, is then developed to solve the final proposed model using the alternating direction method of multipliers. Experimental results on real hyperspectral data sets demonstrate that the proposed approach outperforms many state-of-the-art methods in terms of the overall accuracy, average accuracy, and kappa (k) statistic.
A permanent-magnet dual-rotor motor is particularly suitable for constructing the power split device in hybrid electric vehicle application. This paper proposes a new magnetic-geared dual-rotor motor ...(MGDRM) design with complementary structure, in which both the inner and outer rotors are divided into three modules with a proper angular displacement for each other along the axis direction. This complementary design makes the flux linkage symmetrical and total cogging torque significantly reduced, without impairing the torque production. A simplified magnetic circuit model is developed to illustrate the complementary principle. By finite-element analysis (FEA), the effectiveness of such complementary structure is verified through the comparison with the conventional design. A prototype motor has been manufactured, and experiments have been carried out. Both FEA and experiments show that this new MGDRM offers symmetrical back-EMF waveforms, smaller cogging torque, and lower torque ripple.
An investment factor, long in low-investment stocks and short in high-investment stocks, helps explain the new issues puzzle. Adding the investment factor into standard factor regressions reduces the ...SEO underperformance by about 75%, the IPO underperformance by 80%, the underperformance following convertible debt offerings by 50%, and Daniel and Titman's (2006) composite issuance effect by 40%. The reason is that issuers invest more than nonissuers, and the investment factor earns a significantly positive average return of 0.57% per month.
Deep learning has been extensively applied to segmentation in medical imaging. U-Net proposed in 2015 shows the advantages of accurate segmentation of small targets and its scalable network ...architecture. With the increasing requirements for the performance of segmentation in medical imaging in recent years, U-Net has been cited academically more than 2500 times. Many scholars have been constantly developing the U-Net architecture. This paper summarizes the medical image segmentation technologies based on the U-Net structure variants concerning their structure, innovation, efficiency, etc.; reviews and categorizes the related methodology; and introduces the loss functions, evaluation parameters, and modules commonly applied to segmentation in medical imaging, which will provide a good reference for the future research.
Modeling the processes of neuronal progenitor proliferation and differentiation to produce mature cortical neuron subtypes is essential for the study of human brain development and the search for ...potential cell therapies. We demonstrated a novel paradigm for the generation of vascularized organoids (vOrganoids) consisting of typical human cortical cell types and a vascular structure for over 200 days as a vascularized and functional brain organoid model. The observation of spontaneous excitatory postsynaptic currents (sEPSCs), spontaneous inhibitory postsynaptic currents (sIPSCs), and bidirectional electrical transmission indicated the presence of chemical and electrical synapses in vOrganoids. More importantly, single-cell RNA-sequencing analysis illustrated that vOrganoids exhibited robust neurogenesis and that cells of vOrganoids differentially expressed genes (DEGs) related to blood vessel morphogenesis. The transplantation of vOrganoids into the mouse S1 cortex resulted in the construction of functional human-mouse blood vessels in the grafts that promoted cell survival in the grafts. This vOrganoid culture method could not only serve as a model to study human cortical development and explore brain disease pathology but also provide potential prospects for new cell therapies for nervous system disorders and injury.
The ON-OFF direction selective ganglion cells (DSGCs) in the mammalian retina code image motion by responding much more strongly to movement in one direction. They do so by receiving inhibitory ...inputs selectively from a particular sector of processes of the overlapping starburst amacrine cells, a type of retinal interneuron. The mechanisms of establishment and regulation of this selective connection are unknown. Here, we report that in the rat retina, the morphology, physiology of the ON-OFF DSGCs and the circuitry for coding motion directions develop normally with pharmacological blockade of GABAergic, cholinergic activity and/or action potentials for over two weeks from birth. With recent results demonstrating light independent formation of the retinal DS circuitry, our results strongly suggest the formation of the circuitry, i.e., the connections between the second and third order neurons in the visual system, can be genetically programmed, although emergence of direction selectivity in the visual cortex appears to require visual experience.
In the convolutional neural network, the precise segmentation of small-scale objects and object boundaries in remote sensing images is a great challenge. As the model gets deeper, low-level features ...with geometric information and high-level features with semantic information cannot be obtained simultaneously. To alleviate this problem, a successive pooling attention network (SPANet) was proposed. The SPANet mainly consists of ResNet50 as the backbone, successive pooling attention module (SPAM), and feature fusion module (FFM). Specifically, the SPANet uses two parallel branches to extract high-level features by ResNet50 and low-level features by the first 11 layers of ResNet50. Then, both the high- and low-level features are fed to the SPAM, which is mainly composed of a successive pooling operator and a self-attention submodule, for further extracting deeper multiscale and salient features. In addition, the low- and high-level features after the SPAM are fused by the FFM to achieve the complementarity of spatial and geometric information. This fusion module alleviates the problem of the accurate segmentation of object edges. Finally, the high-level features and enhanced low-level features of the two branches are fused to obtain the final prediction results. Experiments show that the proposed SPANet achieves a good segmentation effect compared with other models on two remotely sensed datasets.
Digital inclusive finance not only reduces the threshold of rural financial services but also plays an important role in promoting the rural economy. Under the background of the digital economy in ...China, we studied how digital inclusive finance affects the high-quality development of the rural economy. For this purpose, we assembled the provincial panel data from 2009 to 2018 and constructed an index system of rural high-quality development and then calculated its index using the entropy weight method. FE regression model and RE regression model were employed for empirical analysis, and we constructed a nonlinear regression model to investigate the nonlinear relationship as well. Furthermore, systematic GMM and iterative GMM were used to solve the endogenous problem. We found that digital inclusive finance plays a significant and positive role in promoting rural high-quality development, mainly through the channel of economic efficiency, urban and rural structure, green ecological development, harmony of people’s livelihood, and innovative development potential. Each subdimension of digital inclusive finance has a positive impact on rural high-quality development to different degrees. There is also a direct nonlinear relationship between the core variables, which act as a disincentive in the initial stage and boost after reaching a certain inflection point. Therefore, it is recommended to increase the construction of rural digital infrastructure, improve the digitalization of inclusive finance with the help of digital technologies such as cloud computing, and expand the breadth and depth of digital inclusive finance services in rural areas, especially in economically underdeveloped provinces and cities, to effectively promote the comprehensive and high-quality development of rural areas.
PDF
