Object detection has been a building block in computer vision. Though considerable progress has been made, there still exist challenges for objects with small size, arbitrary direction, and dense distribution. Apart from natural images, such issues are especially pronounced for aerial images of great importance. This paper presents a novel multicategory rotation detector for small, cluttered and rotated objects, namely SCRDet. Specifically, a sampling fusion network is devised which fuses multi-layer feature with effective anchor sampling, to improve the sensitivity to small objects. Meanwhile, the supervised pixel attention network and the channel attention network are jointly explored for small and cluttered object detection by suppressing the noise and highlighting the objects feature. For more accurate rotation estimation, the IoU constant factor is added to the smooth L1 loss to address the boundary problem for the rotating bounding box. Extensive experiments on two remote sensing public datasets DOTA, NWPU VHR-10 as well as natural image datasets COCO, VOC2007 and scene text data ICDAR2015 show the state-of-the-art performance of our detector. The code and models will be available at https://github.com/DetectionTeamUCAS.
Ship detection has been playing a significant role in the field of remote sensing for a long time, but it is still full of challenges. The main limitations of traditional ship detection methods usually lie in the complexity of application scenarios, the difficulty of intensive object detection, and the redundancy of the detection region. In order to solve these problems above, we propose a framework called Rotation Dense Feature Pyramid Networks (R-DFPN) which can effectively detect ships in different scenes including ocean and port. Specifically, we put forward the Dense Feature Pyramid Network (DFPN), which is aimed at solving problems resulting from the narrow width of the ship. Compared with previous multiscale detectors such as Feature Pyramid Network (FPN), DFPN builds high-level semantic feature-maps for all scales by means of dense connections, through which feature propagation is enhanced and feature reuse is encouraged. Additionally, in the case of ship rotation and dense arrangement, we design a rotation anchor strategy to predict the minimum circumscribed rectangle of the object so as to reduce the redundant detection region and improve the recall. Furthermore, we also propose multiscale region of interest (ROI) Align for the purpose of maintaining the completeness of the semantic and spatial information. Experiments based on remote sensing images from Google Earth for ship detection show that our detection method based on R-DFPN representation has state-of-the-art performance.
Biotransformation is a critical factor that may modify the toxicity, behavior, and fate of engineered nanoparticles in the environment. CeO2 nanoparticles (NPs) are generally recognized as stable under environmental and biological conditions. The present study aims to investigate the biotransformation of CeO2 NPs in plant systems. Transmission electron microscopy (TEM) images show needlelike clusters on the epidermis and in the intercellular spaces of cucumber roots after a treatment with 2000 mg/L CeO2 NPs for 21 days. By using a soft X-ray scanning transmission microscopy (STXM) technique, the needlelike clusters were verified to be CePO4. Near edge X-ray absorption fine structure (XANES) spectra show that Ce presented in the roots as CeO2 and CePO4 while in the shoots as CeO2 and cerium carboxylates. Simulated studies indicate that reducing substances (e.g., ascorbic acids) played a key role in the transformation process and organic acids (e.g., citric acids) can promote particle dissolution. We speculate that CeO2 NPs were first absorbed on the root surfaces and partially dissolved with the assistance of the organic acids and reducing substances excreted by the roots. The released Ce(III) ions were precipitated on the root surfaces and in intercellular spaces with phosphate, or form complexes with carboxyl compounds during translocation to the shoots. To the best of our knowledge, this is the first report confirming the biotransformation and in-depth exploring the translocation process of CeO2 NPs in plants.
With the increasing applications of metal-based nanoparticles in various commercial products, it is necessary to address their environmental fate and potential toxicity. In this work, we assessed the phytotoxicity of lanthanum oxide (La₂O₃) NPs to cucumber plants and determined its distribution and biotransformation in roots by TEM and EDS, as well as STXM and NEXAFS. LaCl₃ was also studied as a reference toxicant. La₂O₃ NPs and LaCl₃ were both transformed to needle-like LaPO₄ nanoclusters in the intercellular regions of the cucumber roots. In vitro experiments demonstrated that the dissolution of La₂O₃ NPs was significantly enhanced by acetic acid. Accordingly, we proposed that the dissolution of NPs at the root surface induced by the organic acids extruded from root cells played an important role in the phytotoxicity of La₂O₃ NPs. The reactions of active NPs at the nano-bio interface should be taken into account when studying the toxicity of dissolvable metal-based nanoparticles.
The pollution arising from oil spills is a matter of great concern due to its damaging impacts on the ecological environment, which has created a tremendous need to find more efficient materials for oil spill cleanup. In this work, we reported a sorbent for oil soak-up from a water surface with a high sorption capacity, good selectivity, and excellent reusability based on the hydrophobic-oleophilic fibrous mats that were fabricated via co-axial electrospinning polystyrene (PS) solution as the shell solution and polyurethane (PU) solution as the core solution. The fine structures of as-prepared fibers were regulated by manipulating the spinning voltages, core solution concentrations, and solvent compositions in shell solutions, which were also characterized by field emission scanning electron microscopy, transmission electron microscopy, nitrogen adsorption method, and synchrotron radiation small-angle X-ray scattering. The effects of inter-fiber voids and intra-fiber porosity on oil sorption capacities were well studied. A comparison of oil sorption capacity for the single fiber with different porous structures was also investigated with the help of scanning transmission X-ray microscopy. The results showed that the sorption capacities of the as-prepared sorbent with regards to motor oil and sunflower seed oil can be 64.40 and 47.48 g g(-1), respectively, approximately 2-3 times that of conventional polypropylene (PP) fibers for these two same oils. Even after five sorption cycles, a comparable oil sorption capacity with PP fibers was still maintained, exhibiting an excellent reusability. We believe that the composite PS-PU fibrous mats have a great potential application in wastewater treatment, oil accident remediation and environmental protection.
With the increasing utilization of nanomaterials, there is a growing concern for the potential environmental and health effects of them. To assess the environmental risks of nanomaterials, better knowledge about their fate and toxicity in plants are required. In this work, we compared the phytotoxicity of nanoparticulate Yb(2)O(3), bulk Yb(2)O(3), and YbCl(3)·6H(2)O to cucumber plants. The distribution and biotransformation of the three materials in plant roots were investigated in situ by TEM, EDS, as well as synchrotron radiation based methods: STXM and NEXAFS. The decrease of biomass was evident at the lowest concentration (0.32 mg/L) when exposed to nano-Yb(2)O(3), while at the highest concentration, the most severe inhibition was from YbCl(3). The inhibition was dependent on the actual amount of toxic Yb uptake by the cucumber plants. In the intercellular regions of the roots, Yb(2)O(3) particles and YbCl(3) were all transformed to YbPO(4). We speculate that the dissolution of Yb(2)O(3) particles induced by the organic acids exuded from roots played an important role in the phytotoxicity. Only under the nano-Yb(2)O(3) treatment, YbPO(4) deposits were found in the cytoplasm of root cells, so the phytotoxicity might also be attributed to the Yb internalized into the cells.
Background Ocular changes are traditionally associated with only a few hepatobiliary diseases. These changes are non-specific and have a low detection rate, limiting their potential use as clinically independent diagnostic features. Therefore, we aimed to engineer deep learning models to establish associations between ocular features and major hepatobiliary diseases and to advance automated screening and identification of hepatobiliary diseases from ocular images.Methods We did a multicentre, prospective study to develop models using slit-lamp or retinal fundus images from participants in three hepatobiliary departments and two medical examination centres. Included participants were older than 18 years and had complete clinical information; participants diagnosed with acute hepatobiliary diseases were excluded. We trained seven slit-lamp models and seven fundus models (with or without hepatobiliary disease [screening model] or one specific disease type within six categories [identifying model]) using a development dataset, and we tested the models with an external test dataset. Additionally, we did a visual explanation and occlusion test. Model performances were evaluated using the area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, and F1* score.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.