To better solve scale variance problem, deep multi-scale methods usually detect objects of different scales by different in-network layers. However, the semantic levels of features from different ...layers are usually inconsistent. In this paper, we propose a multi-branch and high-level semantic network by gradually splitting a base network into multiple different branches. As a result, the different branches have same depth and the output features of different branches have similarly high-level semantics. Due to the difference of receptive fields, the different branches are suitable to detect objects of different scales. Meanwhile, the multi-branch network does not introduce additional parameters by sharing the convolutional weights of different branches. To further improve detection performance, skip-layer connections are used to add context to the branch of relatively small receptive field, and dilated convolution is incorporated to enlarge the resolutions of output feature maps. When they are embedded into Faster RCNN architecture, the weighted scores of proposal generation network and proposal classification network are further proposed. Experiments on three pedestrian datasets (i.e., the KITTI dataset, the Caltech dataset, and the Citypersons dataset), one face dataset (i.e., the WIDER FACE dataset), and two general object datasets (i.e., the COCO benchmark and the PASCAL VOC dataset) demonstrate the effectiveness and generality of proposed method. On these datasets, our method achieves state-of-the-art performance.
On benchmark images, modern dehazing methods are able to achieve very comparable results whose differences are too subtle for people to qualitatively judge. Thus, it is imperative to adopt ...quantitative evaluation on a vast number of hazy images. However, existing quantitative evaluation schemes are not convincing due to a lack of appropriate datasets and poor correlations between metrics and human perceptions. In this work, we attempt to address these issues, and we make two contributions. First, we establish two benchmark datasets, i.e., the BEnchmark Dataset for Dehazing Evaluation (BeDDE) and the EXtension of the BeDDE (exBeDDE), which had been lacking for a long period of time. The BeDDE is used to evaluate dehazing methods via full reference image quality assessment (FR-IQA) metrics. It provides hazy images, clear references, haze level labels, and manually labeled masks that indicate the regions of interest (ROIs) in image pairs. The exBeDDE is used to assess the performance of dehazing evaluation metrics. It provides extra dehazed images and subjective scores from people. To the best of our knowledge, the BeDDE is the first dehazing dataset whose image pairs were collected in natural outdoor scenes without any simulation. Second, we provide a new insight that dehazing involves two separate aspects, i.e., visibility restoration and realness restoration, which should be evaluated independently; thus, to characterize them, we establish two criteria, i.e., the visibility index (VI) and the realness index (RI), respectively. The effectiveness of the criteria is verified through extensive experiments. Furthermore, 14 representative dehazing methods are evaluated as baselines using our criteria on BeDDE. Our datasets and relevant code are available at https://github.com/xiaofeng94/BeDDE-for-defogging .
Super-resolution semantic segmentation (SRSS) is a technique that aims to obtain high-resolution semantic segmentation results based on resolution-reduced input images. SRSS can significantly reduce ...computational cost and enable efficient, high-resolution semantic segmentation on mobile devices with limited resources. Some of the existing methods require modifications of the original semantic segmentation network structure or add additional and complicated processing modules, which limits the flexibility of actual deployment. Furthermore, the lack of detailed information in the low-resolution input image renders existing methods susceptible to misdetection at the semantic edges. To address the above problems, we propose a simple but effective framework called multi-resolution learning and semantic edge enhancement-based super-resolution semantic segmentation (MS-SRSS) which can be applied to any existing encoder-decoder based semantic segmentation network. Specifically, a multi-resolution learning mechanism (MRL) is proposed that enables the feature encoder of the semantic segmentation network to improve its feature extraction ability. Furthermore, we introduce a semantic edge enhancement loss (SEE) to alleviate the false detection at the semantic edges. We conduct extensive experiments on the three challenging benchmarks, Cityscapes, Pascal Context, and Pascal VOC 2012, to verify the effectiveness of our proposed MS-SRSS method. The experimental results show that, compared with the existing methods, our method can obtain the new state-of-the-art semantic segmentation performance.
In this letter, effects of top electrodes (TEs) on ferroelectric properties of Hf 0.5 Zr 0.5 O 2 (HZO) thin films are examined systematically. The remnant polarization (P r ) of HZO thin films ...increases by altering TEs with lower thermal expansions coefficient (<inline-formula> <tex-math notation="LaTeX">\alpha </tex-math></inline-formula>). The largest 2P r value of 38.72 <inline-formula> <tex-math notation="LaTeX">\mu \text{C} </tex-math></inline-formula>/cm 2 is observed for W TE with <inline-formula> <tex-math notation="LaTeX">\alpha = 4.5\times 10^{\mathsf {-6}} </tex-math></inline-formula>/K, while the 2P r value is only <inline-formula> <tex-math notation="LaTeX">22.83~\mu \text{C} </tex-math></inline-formula>/cm 2 for Au TE with <inline-formula> <tex-math notation="LaTeX">\alpha = 14.2\times 10^{\mathsf {-6}} </tex-math></inline-formula>/K. Meanwhile, coercive field (E c ) shifts along the electric field axis and the offset is found to be dependent on the difference of workfunctions (WFs) between TE and TiN bottom electrode (BE). E c shifts toward negative/positive direction, when the WF of TE is larger/smaller (Pt, Pd, Au/W, Al, Ta) than TiN BE. This letter provides an effective way to modulate HfO 2 -based device performance for different requirements in actual application.
For heterogeneous network, which has been viewed as one pioneering technology for making cellular networks be evolved into 5G systems, reducing energy consumption by dynamically switching off base ...stations (BSs) has attracted increasing attention recently. With aiming at optimization on energy saving only or another energy-related performance tradeoffs, several BS switch-off strategies have been proposed from different design perspectives, such as random, distance-aware, load-aware, and auction-based strategies. Furthermore, work has been done to consider joint design for BS switch-off strategy and another strategies, such as user association, resource allocation, and physical-layer interference cancellation strategies. Finally, there have been research results about this topic in emerging cloud radio access networks. In this paper, we take an overview on these technologies and present the state of the art on each aspect. Some challenges that need to be solved in this research filed for future work are also described.
With the explosive growth of data communications, existing infrastructure networks are under ever‐increasing pressure. Due to the advantages of fully controllable mobility, rapid deployment, and low ...cost, the unmanned aerial vehicles (UAVs) have attracted much attentions from both industry and academia in recent years, and it has become an inevitable trend to employ UAVs to enhance the network performance in different environments. As an important paradigm of UAV‐assisted communications, UAV relaying communications has been regarded as a promising solution in enhancing connectivity and improving transmission rate. This paper for the first time comprehensively summarizes UAV relaying communications and its application scenarios, including single UAV relaying networks, multi‐user UAV relaying networks, multi‐hop UAV relaying networks, as well as Internet of UAVs, and deeply analyzes the key technologies and challenges to be solved under this topic. Furthermore, the state‐of‐the‐art researches and opportunities of UAV relaying communications are discussed in detail.
A seafloor observation network (SON) consists of a large number of heterogeneous devices that monitor the deep sea and communicate with onshore data centers. Due to the long-distance information ...transmission and the risk of malicious attacks, ensuring the integrity of data in transit is essential. A cryptographically secure frame check sequence (FCS) has shown great advantages in protecting data integrity. However, the commonly used FCS has a collision possibility, which poses a security risk; furthermore, reducing the encryption calculation cost is a challenge. In this paper, we propose a secure, lightweight encryption scheme for transmitted data inspired by mimic defense from dynamic heterogeneous redundancy theory. Specifically, we use dynamic keys to encrypt a data block and generate multiple encrypted heterogeneous blocks for transmission. These continuously changing encrypted data blocks increase the confusion regarding the original encoded data, making it challenging for attackers to interpret and modify the data blocks. Additionally, the redundant information from the multiple blocks can identify and recover tampered data. Our proposed scheme is suitable for resource-constrained environments where lightweight encryption is crucial. Through experimental demonstrations and analysis methods, we determine the effectiveness of our encryption scheme in reducing computational costs and improving security performance to protect data integrity.
Fruit characteristics of sweet watermelon are largely the result of human selection. Here we report an improved watermelon reference genome and whole-genome resequencing of 414 accessions ...representing all extant species in the Citrullus genus. Population genomic analyses reveal the evolutionary history of Citrullus, suggesting independent evolutions in Citrullus amarus and the lineage containing Citrullus lanatus and Citrullus mucosospermus. Our findings indicate that different loci affecting watermelon fruit size have been under selection during speciation, domestication and improvement. A non-bitter allele, arising in the progenitor of sweet watermelon, is largely fixed in C. lanatus. Selection for flesh sweetness started in the progenitor of C. lanatus and continues through modern breeding on loci controlling raffinose catabolism and sugar transport. Fruit flesh coloration and sugar accumulation might have co-evolved through shared genetic components including a sugar transporter gene. This study provides valuable genomic resources and sheds light on watermelon speciation and breeding history.
Containers and microservices have become the most popular method for hosting IoT applications in cloud servers. However, one major security issue of this method is that if a container image contains ...software with security vulnerabilities, the associated microservices also become vulnerable at run-time. Existing works attempted to reduce this risk with vulnerability-scanning tools. They, however, demand an up-to-date database and may not work with unpublished vulnerabilities. In this paper, we propose a novel system to strengthen container security from unknown attack using the mimic defense framework. Specifically, we constructed a resource pool with variant images and observe the inconsistency in execution results, from which we can identify potential vulnerabilities. To avoid continuous attack, we created a graph-based scheduling strategy to maximize the randomness and heterogeneity of the images used to replace the current images. We implemented a prototype using Kubernetes. Experimental results show that our framework makes hackers have to send 54.9% more random requests to complete the attack and increases the defence success rate by around 8.16% over the baseline framework to avoid the continuous unknown attacks.
Display omitted
•A novel load path-based lattice design and optimization methodology is proposed.•Reinforced strut orientations of anisotropic unit cells are tailored by load paths.•Variable radii of ...the lattice struts are optimized by topology optimization.•Proposed lattice structures show significant improvements in stiffness and strength.
Lattice structures have been increasingly used in load-carrying applications due to their exceptional mechanical performance. This study presents a novel load path methodology for designing and optimizing functionally graded lattice structures composed of anisotropic unit cells with directional reinforcement struts. Firstly, the optimal density distribution of the lattice structure is obtained by solid isotropic material with penalization (SIMP) topology optimization. Secondly, pointing stress vectors of the structure are calculated to determine the orientations of the unit cells. Lastly, the lattice model is constructed using tapered beams for a smooth transition between struts with different radii. Two examples of a simply supported beam and a 3-dimensional base support structure are provided. The experimental validation showcases that the proposed design improves the specific stiffness by 75 % compared to the uniform body-centered cubic design. Furthermore, the strength-to-weight ratio is increased by 232 % due to a more desirable stress distribution.