Explore the vast range of titles available.

The results of collapsed building detection act as an important reference for damage assessment after an earthquake, which is crucial for governments in order to efficiently determine the affected area and execute emergency rescue. For this task, unmanned aerial vehicle (UAV) images are often used as the data sources due to the advantages of high flexibility regarding data acquisition time and flying requirements and high resolution. However, collapsed buildings are typically distributed in both connected and independent pieces and with arbitrary shapes, and these are generally more obvious in the UAV images with high resolution; therefore, the corresponding detection is restricted by using conventional convolutional neural networks (CNN) and the detection results are difficult to evaluate. In this work, based on faster region-based convolutional neural network (Faster R-CNN), deformable convolution was used to improve the adaptability to the arbitrarily shaped collapsed buildings. In addition, inspired by the idea of pixelwise semantic segmentation, in contrast to the intersection over union (IoU), a new method which estimates the intersected proportion of objects (IPO) is proposed to describe the degree of the intersection of bounding boxes, leading to two improvements: first, the traditional non-maximum suppression (NMS) algorithm is improved by integration with the IPO to effectively suppress the redundant bounding boxes; second, the IPO is utilized as a new indicator to determine positive and negative bounding boxes, and is introduced as a new strategy for precision and recall estimation, which can be considered a more reasonable measurement of the degree of similarity between the detected bounding boxes and ground truth bounding boxes. Experiments show that compared with other models, our work can obtain better precision and recall for detecting collapsed buildings for which an F1 score of 0.787 was achieved, and the evaluation results from the suggested IPO are qualitatively closer to the ground truth. In conclusion, the improved NMS with the IPO and Faster R-CNN in this paper is feasible and efficient for the detection of collapsed buildings in UAV images, and the suggested IPO strategy is more suitable for the corresponding detection result’s evaluation.

Background Dysbiosis of the gut microbiome is involved in the pathogenesis of various diseases, but the contribution of gut microbes to the progression of chronic obstructive pulmonary disease (COPD) is still poorly understood. Methods We carried out 16S rRNA gene sequencing and short-chain fatty acid analyses in stool samples from a cohort of 73 healthy controls, 67 patients with COPD of GOLD stages I and II severity, and 32 patients with COPD of GOLD stages III and IV severity. Fecal microbiota from the three groups were then inoculated into recipient mice for a total of 14 times in 28 days to induce pulmonary changes. Furthermore, fecal microbiota from the three groups were inoculated into mice exposed to smoke from biomass fuel to induce COPD-like changes. Results We observed that the gut microbiome of COPD patients varied from that of healthy controls and was characterized by a distinct overall microbial diversity and composition, a Prevotella -dominated gut enterotype and lower levels of short-chain fatty acids. After 28 days of fecal transplantation from COPD patients, recipient mice exhibited elevated lung inflammation. Moreover, when mice were under both fecal transplantation and biomass fuel smoke exposure for a total of 20 weeks, accelerated declines in lung function, severe emphysematous changes, airway remodeling and mucus hypersecretion were observed. Conclusion These data demonstrate that altered gut microbiota in COPD patients is associated with disease progression in mice model.

All-inorganic metal halide perovskites are showing promising development towards efficient long-term stable materials and solar cells. Element doping, especially on the lead site, has been proved to be a useful strategy to obtain the desired film quality and material phase for high efficient and stable inorganic perovskite solar cells. Here we demonstrate a function by adding barium in CsPbI 2 Br. We find that barium is not incorporated into the perovskite lattice but induces phase segregation, resulting in a change in the iodide/bromide ratio compared with the precursor stoichiometry and consequently a reduction in the band gap energy of the perovskite phase. The device with 20 mol% barium shows a high power conversion efficiency of 14.0% and a great suppression of non-radiative recombination within the inorganic perovskite, yielding a high open-circuit voltage of 1.33 V and an external quantum efficiency of electroluminescence of 10 −4 . Element doping has been proven a useful strategy to tune the properties of halide perovskites. Here Xiang et al. show that barium unexpectedly does not incorporate in perovskite lattice but induces phase segregation and bandgap reduction and inhibits non-radiative recombination.

Ship targets are high-value military and civilian targets with broad application prospects. However, the precise focusing of ships is still a difficult issue because of their complicated six-degree-of-freedom motions on the sea surface. This paper focused on investigating the effect of ship six-degree-of-freedom oscillation on Synthetic Aperture Radar imaging. Firstly, based on the six-degree-of-freedom motions, the accurate range models for ship linear oscillation and angular oscillation were built, and the superiority was verified by comparing them with the models described in published literature. Secondly, we used the Taylor formula and Bessel function to expand the phase error introduced by ship oscillation, then their effects on imaging were further analyzed. Finally, based on the measured ship attitude data, we generated the semi-physical echoes of the oscillatory ship to validate the analysis throughout this article. Based on the proposed range model, we also made some tentative on the phase compensation method by fitting ship attitude angles with multiple sinusoidal functions.

Target detection on the sea-surface has always been a high-profile problem, and the detection of weak targets is one of the most difficult problems and the key issue under this problem. Traditional techniques, such as imaging, cannot effectively detect these types of targets, so researchers choose to start by mining the characteristics of the received echoes and other aspects for target detection. This paper proposes a false alarm rate (FAR) controllable deep forest model based on six-dimensional feature space for efficient and accurate detection of weak targets on the sea-surface. This is the first attempt at the deep forest model in this field. The validity of the model was verified on IPIX data, and the detection probability was compared with other proposed methods. Under the same FAR condition, the average detection accuracy rate of the proposed method could reach over 99.19%, which is 9.96% better than the results of the current most advanced method (K-NN FAR-controlled Detector). Experimental results show that multi-feature fusion and the use of a suitable detection framework have a positive effect on the detection of weak targets on the sea-surface.

Synthetic aperture radar (SAR) has become one of the most important means of information acquisition in today’s society and shows great potential in many fields. Target identification and classification of SAR images are also the focus of research. With the vigorous development of deep learning, many researchers apply this method to SAR target classification to obtain a more automatic process and more accurate results. In this paper, a novel deep forest model constructed by multi-grained cascade forest (gcForest), which is different from the traditional neural network (NN) model, is employed to classify ten types of SAR targets in the moving and stationary target acquisition and recognition (MSTAR) dataset. Considering that the targets of input images may be off-center and of different sizes in practical applications, two improved models based on varying weights by image features have been put forward, and both obtain better results. A series of experiments have been conducted to optimize model parameters, and final results with the MSTAR dataset illustrate that the two improved models are both superior to the original gcForest model. This is the first attempt to classify SAR targets using the non-NN model.

Regularization has become an important method in adversarial defense. However, the existing regularization-based defense methods do not discuss which features in convolutional neural networks (CNN) are more suitable for regularization. Thus, in this paper, we propose a multi-stage feature fusion network with a feature regularization operation, which is called Enhanced Multi-Stage Feature Fusion Network (EMSF2Net). EMSF2Net mainly combines three parts: multi-stage feature enhancement (MSFE), multi-stage feature fusion (MSF2), and regularization. Specifically, MSFE aims to obtain enhanced and expressive features in each stage by multiplying the features of each channel; MSF2 aims to fuse the enhanced features of different stages to further enrich the information of the feature, and the regularization part can regularize the fused and original features during the training process. EMSF2Net has proved that if the regularization term of the enhanced multi-stage feature is added, the adversarial robustness of CNN will be significantly improved. The experimental results on extensive white-box attacks on the CIFAR-10 dataset illustrate the robustness and effectiveness of the proposed method.

Soybean bacterial blight disease is a threat to soybean production. Multispectral technology has shown good potential in detecting this disease and can overcome the limitations of traditional methods. The aim of this study was to perform field monitoring of the dynamics of this disease in Northeast China in 2022. The correlation between the soybean chlorophyll content index (CCI) and disease grade was obtained using artificial inoculation of the pathogen. The correlation between the soybean CCI, disease grade, green normalized difference vegetation index (GNDVI), and soybean yield was analyzed using a drone-mounted spectrometer platform for image acquisition and preprocessing. The soybean CCI was negatively correlated with the disease grade. The GNDVI declined with disease progression, which allowed for an indirect determination of the disease grade. The soybean yield loss was significant at disease grade 4 for soybean bacterial blight disease. The random forest regression model was more accurate than the regression model in estimating the yield based on the GNDVI. Therefore, the GNDVI could be used to survey the disease class and estimate the yield using the random forest model. This study provides support for field trials of drone-mounted multispectral equipment. This surveillance approach holds the potential to bring about precision plant protection in the future.

Diatoms often outnumber other eukaryotic algae in the oceans, especially in coastal environments characterized by frequent fluctuations in light intensity. The identities and operational mechanisms of regulatory factors governing diatom acclimation to high light stress remain largely elusive. Here, we identified the AUREO1c protein from the coastal diatom Phaeodactylum tricornutum as a crucial regulator of non-photochemical quenching (NPQ), a photoprotective mechanism that dissipates excess energy as heat. AUREO1c detects light stress using a light-oxygen-voltage (LOV) domain and directly activates the expression of target genes, including LI818 genes that encode NPQ effector proteins, via its bZIP DNA-binding domain. In comparison to a kinase-mediated pathway reported in the freshwater green alga Chlamydomonas reinhardtii , the AUREO1c pathway exhibits a faster response and enables accumulation of LI818 transcript and protein levels to comparable degrees between continuous high-light and fluctuating-light treatments. We propose that the AUREO1c-LI818 pathway contributes to the resilience of diatoms under dynamic light conditions. Diatoms thrive in dynamic environments that frequently confer high light stress. In this study, the authors report a diatom photoreceptor that triggers immediate transcription of photoprotective genes upon high light signals. The fast response kinetics of this pathway confer advantages in environments with rapid light fluctuations.

In our digitally driven society, advances in software and hardware to capture video data allow extensive gathering and analysis of large datasets. This has stimulated interest in extracting information from video data, such as buildings and urban streets, to enhance understanding of the environment. Urban buildings and streets, as essential parts of cities, carry valuable information relevant to daily life. Extracting features from these elements and integrating them with technologies such as VR and AR can contribute to more intelligent and personalized urban public services. Despite its potential benefits, collecting videos of urban environments introduces challenges because of the presence of dynamic objects. The varying shape of the target building in each frame necessitates careful selection to ensure the extraction of quality features. To address this problem, we propose a novel evaluation metric that considers the video-inpainting-restoration quality and the relevance of the target object, considering minimizing areas with cars, maximizing areas with the target building, and minimizing overlapping areas. This metric extends existing video-inpainting-evaluation metrics by considering the relevance of the target object and interconnectivity between objects. We conducted experiment to validate the proposed metrics using real-world datasets from Japanese cities Sapporo and Yokohama. The experiment results demonstrate feasibility of selecting video frames conducive to building feature extraction.