Cross-scene hyperspectral image classification (HSIC) is a challenging topic in remote sensing, especially when there are no labels target domain. Domain adaptation (DA) techniques for cross-scene HSIC aim to label domain by associating it with labeled source Most existing DA methods learn domain-invariant features reducing feature distance across domains. Recently, contrastive learning has shown excellent performance computer vision tasks, but little or research on the of HSIC. Considering...

Although deep reinforcement learning (DRL) methods are promising for making behavioral decisions in autonomous vehicles (AVs), their low training efficiency and difficulty to adapt untrained cases hinder applications. Introducing a human role the DRL paradigm could improve by using prior knowledge overcome deployment online takeover. In this study, novel value-based algorithm that leverages guidance its performance is proposed addressing high-level decision-making problems driving. We...

Recent studies in cross-domain classification have shown that discriminant information of both source and target domains is very important. In this article, we propose a new domain adaptation (DA) method for hyperspectral image (HSI) classification, called graph embedding distribution alignment (GEDA). GEDA uses the pseudo-label learning to learn interclass intraclass divergence matrices domains, which preserves local domains. Meanwhile, spatial spectral features HSI are used, subspace...

While most recent autonomous driving system focuses on developing perception methods ego-vehicle sensors, people tend to overlook an alternative approach leverage intelligent roadside cameras extend the ability beyond visual range. We discover that state-of-the-art vision-centric detection perform poorly cameras. This is because these mainly focus recovering depth regarding camera center, where difference between car and ground quickly shrinks while distance increases. In this paper, we...

Multimodally distributed data is very common in remote sensing images, such as hyperspectral images (HSIs). It important to capture the local manifold structure while preserving global discriminant information multimodal data. In this paper, an adaptive analysis and distribution matching (ALDADM) method designed for domain adaptation (DA) HSI classification. ALDADM uses extract features that are discriminative robust multimodally Meanwhile, considering spectral properties of HSI, shift...

The purpose of this study was to assess the suitability lactic acid bacteria (LABs) isolated from Muscovy duck as a potential probiotic.Isolates were identified by targeted polymerase chain reaction and assessed in vitro for probiotic characteristics such autoaggregation; surface-charge; hydrophobicity; tolerance acidic pH, bile salts protease; expression genes involved Caco-2 cell adhesion.The LAB isolates exhibited strong resistance high concentration produced acid, bacteriostatic (P <...

In smart classroom environments, accurately recognizing students' facial expressions is crucial for teachers to efficiently assess learning states, timely adjust teaching strategies, and enhance quality effectiveness. this paper, we propose a student expression recognition model based on multi-scale deep fine-grained feature attention enhancement (SFER-MDFAE) address the issues of inaccurate extraction poor robustness in scenarios. Firstly, construct novel dual-pooling aggregation module...

A novel soft-switching buck boost converter using a pulse current regenerative snubber is presented. The proposed circuit consists of passive components that configure and resonant circuit. It shown the effective operation region its efficiency improvement result on basis experimental evaluations.

This paper introduces the theoretical analysis and derivation of localizability estimation for mobile robots based on point cloud observation. Combined with observation model, pre-established map is firstly clustered segmented a newly proposed method considering multiple constraints depth map. Then set to be equal strength associated 3D features. In addition, using information matrix theory, this integrates Fisher's features estimate localizability.

&lt;p class="Abstract"&gt;A 4&lt;sup&gt;th&lt;/sup&gt;-order Kalman filter (KF) algorithm is developed based on the thermal model of an asynchronous machine. The parameters are identified and KF implemented in a wireless sensor network (WSN) to estimate temperatures stator windings, rotor cage, core speed, coolant air temperature, effective current voltage acquired by WTIM (wireless transducer interface module) separately transmitted NCAP (network capable application processor) where...

Real-time pedestrian crossing prediction is an essential prerequisite for automated driving systems, especially urban driving. Vision-based deep neural network models have been investigated to solve the problem. However, certain challenges still need be addressed, such as optimal strategy fusing different features from heterogeneous sensors. In this letter, a novel architecture proposed fuse spatial and temporal prediction. The divided into vision nonvision paths corresponding information...

Abstract. In this paper, a fourth-order Kalman filter (KF) algorithm is implemented in the wireless sensor node to estimate temperatures of stator winding, rotor cage and core induction machine. Three separate nodes are used as data acquisition systems for different input signals. Six Hall sensors acquire three-phase currents voltages All them processed root mean square (rms) ampere volt. A rotary encoder mounted speed Pt-1000 temperature coolant air. The signals physical unit transmitted...

Heart related health conditions, such as heart attacks, are the primary reasons for millions of deaths worldwide. Most existing heart-disease diagnosis techniques invasive and need trained medical professionals. The current paper reports development a rapid, non-invasive sound method describes its functionality in Health Social Networks (HSNs). HSNs platforms professionals, organizations patients to use IT applications share information, including e.g., allow users conveniently inexpensively...

The paper quantizes the joint effects of finite rate and delay feedback channel on performance downlink transmissions coverage in co-channel multi-antenna tiered macro/femto networks. per-tier probability successful reception that signal to interference ratio (SIR) is larger than minimum SIR target derived. Based quality service (QoS) constraints, radius macrocell maximum no-coverage femtocell are obtained respectively. Simulation results indicate both reduced by imperfect feedback. While if...

Recently, great interest has been generated in monitoring and increasing the level of antioxidants human body, as higher antioxidant levels are linked to better overall health lower risk various ailments. Not only have researchers gained a understanding roles development antioxidants, they also explored methods detecting humans. While research suggests that greater consumption plant based food positively correlates with levels, general public, well professionals lack readily available, low...

Abstract Background Recent studies indicated that exosomal programmed death-ligand 1 (PD-L1) derived from cancers could induce immunosuppression and tumor pathogenesis. However, it is unclear how the exosome influences osteosarcoma (OS) progression whether PD-L1 also exists in serum exosomes (Sr-exosomes) of patients with osteosarcoma. Methods We examined 70 osteosarcoma, 9 benign tumors 22 healthy donors. Osteosarcoma-derived exosomal-PD-L1 were functionally evaluated using vivo vitro...