Recent Advances in Open Set Recognition: A Survey Geng, Chuanxing; Huang, Sheng-Jun; Chen, Songcan
IEEE transactions on pattern analysis and machine intelligence,
10/2021, Letnik:
43, Številka:
10
Journal Article
Recenzirano
In real-world recognition/classification tasks, limited by various objective factors, it is usually difficult to collect training samples to exhaust all classes when training a recognizer or ...classifier. A more realistic scenario is open set recognition (OSR), where incomplete knowledge of the world exists at training time, and unknown classes can be submitted to an algorithm during testing, requiring the classifiers to not only accurately classify the seen classes, but also effectively deal with unseen ones. This paper provides a comprehensive survey of existing open set recognition techniques covering various aspects ranging from related definitions, representations of models, datasets, evaluation criteria, and algorithm comparisons. Furthermore, we briefly analyze the relationships between OSR and its related tasks including zero-shot, one-shot (few-shot) recognition/learning techniques, classification with reject option, and so forth. Additionally, we also review the open world recognition which can be seen as a natural extension of OSR. Importantly, we highlight the limitations of existing approaches and point out some promising subsequent research directions in this field.
During their lifetime, plants encounter numerous biotic and abiotic stresses with diverse modes of attack. Phytohormones, including salicylic acid (SA), ethylene (ET), jasmonate (JA), abscisic acid ...(ABA), auxin (AUX), brassinosteroid (BR), gibberellic acid (GA), cytokinin (CK) and the recently identified strigolactones (SLs), orchestrate effective defense responses by activating defense gene expression. Genetic analysis of the model plant
has advanced our understanding of the function of these hormones. The SA- and ET/JA-mediated signaling pathways were thought to be the backbone of plant immune responses against biotic invaders, whereas ABA, auxin, BR, GA, CK and SL were considered to be involved in the plant immune response through modulating the SA-ET/JA signaling pathways. In general, the SA-mediated defense response plays a central role in local and systemic-acquired resistance (SAR) against biotrophic pathogens, such as Pseudomonas syringae, which colonize between the host cells by producing nutrient-absorbing structures while keeping the host alive. The ET/JA-mediated response contributes to the defense against necrotrophic pathogens, such as
, which invade and kill hosts to extract their nutrients. Increasing evidence indicates that the SA- and ET/JA-mediated defense response pathways are mutually antagonistic.
Binary Relevance is a well-known framework for multi-label classification, which considers each class label as a binary classification problem. Many existing multi-label algorithms are constructed ...within this framework, and utilize identical data representation in the discrimination of all the class labels. In multi-label classification, however, each class label might be determined by some specific characteristics of its own. In this paper, we seek to learn label-specific data representation for each class label, which is composed of label-specific features. Our proposed method LLSF can not only be utilized for multi-label classification directly, but also be applied as a feature selection method for multi-label learning and a general strategy to improve multi-label classification algorithms comprising a number of binary classifiers. Inspired by the research works on modeling high-order label correlations, we further extend LLSF to learn class-Dependent Labels in a sparse stackingway, denoted as LLSF-DL. It incorporates both second-order- and high-order label correlations. A comparative study with the state-of-the-art approaches manifests the effectiveness and efficiency of our proposed methods.
Image denoising aims to restore a clean image from an observed noisy one. Model-based image denoising approaches can achieve good generalization ability over different noise levels and are with high ...interpretability. Learning-based approaches are able to achieve better results, but usually with weaker generalization ability and interpretability. In this paper, we propose a wavelet-inspired invertible network (WINNet) to combine the merits of the wavelet-based approaches and learning-based approaches. The proposed WINNet consists of K -scale of lifting inspired invertible neural networks (LINNs) and sparsity-driven denoising networks together with a noise estimation network. The network architecture of LINNs is inspired by the lifting scheme in wavelets. LINNs are used to learn a non-linear redundant transform with perfect reconstruction property to facilitate noise removal. The denoising network implements a sparse coding process for denoising. The noise estimation network estimates the noise level from the input image which will be used to adaptively adjust the soft-thresholds in LINNs. The forward transform of LINNs produces a redundant multi-scale representation for denoising. The denoised image is reconstructed using the inverse transform of LINNs with the denoised detail channels and the original coarse channel. The simulation results show that the proposed WINNet method is highly interpretable and has strong generalization ability to unseen noise levels. It also achieves competitive results in the non-blind/blind image denoising and in image deblurring.
•Fate of heavy metals during the hydrothermal treatment of sewage sludge.•Assessment indices of the contamination level and ecological risk of heavy metals.•The effect of reaction temperature and ...reaction time.•The effect of the addition of catalyst/other biomass.•The effect of the type of liquefaction solvent.
Various hydrothermal treatment methods, including hydrothermal carbonization, liquefaction and sub/super-critical water gasification, have been applied to the disposal of sewage sludge for producing bio-materials or bio-fuels. It has become a research hotspot whether the heavy metals contained in sewage sludge can be well treated/stabilized after the hydrothermal treatments. This review firstly summarized the methods of assessing heavy metals’ contamination level/risk and then discussed the migration and transformation behaviors of heavy metals from the following aspects: the effect of reaction temperature, the effect of additives (catalysts and other biomass), the effect of the type of solvent and the effect of reaction time. This review can provide an important reference for the further study of the migration and transformation behaviors of heavy metals during the hydrothermal treatment of sewage sludge.
Fog computing extends the facility of cloud computing from the center to edge networks. Although fog computing has the advantages of location awareness and low latency, the rising requirements of ...ubiquitous connectivity and ultra-low latency challenge real-time traffic management for smart cities. As an integration of fog computing and vehicular networks, vehicular fog computing (VFC) is promising to achieve real-time and location-aware network responses. Since the concept and use case of VFC are in the initial phase, this article first constructs a three-layer VFC model to enable distributed traffic management in order to minimize the response time of citywide events collected and reported by vehicles. Furthermore, the VFC-enabled offloading scheme is formulated as an optimization problem by leveraging moving and parked vehicles as fog nodes. A real-world taxi-trajectory-based performance analysis validates our model. Finally, some research challenges and open issues toward VFC-enabled traffic management are summarized and highlighted.
Acoustofluidics, the integration of acoustics and microfluidics, is a rapidly growing research field that is addressing challenges in biology, medicine, chemistry, engineering, and physics. In ...particular, acoustofluidic separation of biological targets from complex fluids has proven to be a powerful tool due to the label-free, biocompatible, and contact-free nature of the technology. By carefully designing and tuning the applied acoustic field, cells and other bioparticles can be isolated with high yield, purity, and biocompatibility. Recent advances in acoustofluidics, such as the development of automated, point-of-care devices for isolating sub-micron bioparticles, address many of the limitations of conventional separation tools. More importantly, advances in the research lab are quickly being adopted to solve clinical problems. In this review article, we discuss working principles of acoustofluidic separation, compare different approaches of acoustofluidic separation, and provide a synopsis of how it is being applied in both traditional applications, such as blood component separation, cell washing, and fluorescence activated cell sorting, as well as emerging applications, including circulating tumor cell and exosome isolation.
Acute graft-versus-host disease (aGVHD) is a common complication of allogeneic hematopoietic stem cell transplantation (alloHCT) and is a major cause of morbidity and mortality. Systemic steroid ...therapy is the first-line treatment for aGVHD, although about half of patients will become refractory to treatment. As the number of patients undergoing alloHCT increases, developing safe and effective treatments for aGVHD will become increasingly important, especially for those whose disease becomes refractory to systemic steroid therapy. This paper reviews current treatment options for patients with steroid-refractory aGVHD and discusses data from recently published clinical studies to outline emerging therapeutic strategies.
The lithium–air battery (LAB) is envisaged as an ultimate energy storage device because of its highest theoretical specific energy among all known batteries. However, parasitic reactions bring about ...vexing issues on the efficiency and longevity of the LAB, among which the formation and decomposition of lithium carbonate Li2CO3 is of paramount importance. The discovery of Li2CO3 as the main discharge product in carbonate‐based electrolytes once brought researchers to “the end of the idyll“ in the early 2010s. In the past few years, tremendous efforts have been made to understand the formation and decomposition mechanisms of Li2CO3, as well as to conceive novel chemical/material strategies to suppress the Li2CO3 formation and to facilitate the Li2CO3 decomposition. Moreover, the study on Li2CO3 in LABs is opening up a new research field in energy technology. Considering the rapid development and innumerous emerging issues, it is timely to recapitulate the current understandings, define the ambiguities and the scientific gaps, and discuss topics of high priority for future research, which is the aim of this Minireview.
Formation and decomposition of Li2CO3: In lithium–air batteries, Li2CO3 is a major by‐product that can lead to cell dry‐out and early failure. Therefore, understanding the formation and decomposition mechanisms of Li2CO3 lays the basis for a better design of lithium–air batteries.
This paper presents a comprehensive analysis of the current-stress optimization and soft-switching operation of the isolated bidirectional dual active bridge (DAB) dc-dc converter with the unified ...triple-phase-shift (UTPS) control. On this basis, the current-stress-optimized modulation scheme is proposed for DAB, which leads to the minimum current stress with the required transmission power and voltage conversion ratio in the whole load range. Moreover, the full soft-switching operation is achieved for the converter simultaneously. Distinct from the previous modulation schemes, the proposed optimized modulation scheme is deduced from a unified analysis of TPS where all effective switching modes are investigated. A novel algorithm based on the Karush-Kuhn-Tucker (KKT) conditions is originally proposed to derive closed-form solutions for the global optimal control parameters. This paper also presents a typical closed-loop control strategy for DAB with TPS control and detailed descriptions about the closed-loop operation. A laboratory prototype is applied, and the experimental results validate that the current-stress optimization and efficiency improvement are realized by applying the optimized modulation scheme. The experimental results also verify the effectiveness of the closed-loop control strategy for DAB with TPS control.