Xiamen University

About: Xiamen University is a education organization based out in Amoy, Fujian, China. It is known for research contribution in the topics: Catalysis & Population. The organization has 50472 authors who have published 54480 publications receiving 1058239 citations. The organization is also known as: Amoy University & Xiàmén Dàxué.

Random Erasing Data Augmentation

Abstract: In this paper, we introduce Random Erasing, a new data augmentation method for training the convolutional neural network (CNN). In training, Random Erasing randomly selects a rectangle region in an image and erases its pixels with random values. In this process, training images with various levels of occlusion are generated, which reduces the risk of over-fitting and makes the model robust to occlusion. Random Erasing is parameter learning free, easy to implement, and can be integrated with most of the CNN-based recognition models. Albeit simple, Random Erasing is complementary to commonly used data augmentation techniques such as random cropping and flipping, and yields consistent improvement over strong baselines in image classification, object detection and person re-identification. Code is available at: this https URL.

Reactive oxygen species generating systems meeting challenges of photodynamic cancer therapy

Abstract: The reactive oxygen species (ROS)-mediated mechanism is the major cause underlying the efficacy of photodynamic therapy (PDT). The PDT procedure is based on the cascade of synergistic effects between light, a photosensitizer (PS) and oxygen, which greatly favors the spatiotemporal control of the treatment. This procedure has also evoked several unresolved challenges at different levels including (i) the limited penetration depth of light, which restricts traditional PDT to superficial tumours; (ii) oxygen reliance does not allow PDT treatment of hypoxic tumours; (iii) light can complicate the phototherapeutic outcomes because of the concurrent heat generation; (iv) specific delivery of PSs to sub-cellular organelles for exerting effective toxicity remains an issue; and (v) side effects from undesirable white-light activation and self-catalysation of traditional PSs. Recent advances in nanotechnology and nanomedicine have provided new opportunities to develop ROS-generating systems through photodynamic or non-photodynamic procedures while tackling the challenges of the current PDT approaches. In this review, we summarize the current status and discuss the possible opportunities for ROS generation for cancer therapy. We hope this review will spur pre-clinical research and clinical practice for ROS-mediated tumour treatments.

Surface-Enhanced Raman Scattering: From Noble to Transition Metals and from Rough Surfaces to Ordered Nanostructures

Abstract: In the mid-1970s, surface-enhanced Raman scattering (SERS) was discovered which impacted on surface science and spectroscopy because of its extremely high surface sensitivity. However, SERS had not developed as many people had hoped to be a powerful surface diagnostic technique that can be widely used because of some obstacles. For example, only three noble metals Au, Ag, and Cu could provide large enhancement, severely limiting the widespread applications involving other metallic materials of both fundamental and practical importance. In this article, emphasis is put on the recent work of our group to directly generate SERS on net transition metals (e.g., Pt, Ru, Rh, Pd, Fe, Co, Ni, and their alloys) by developing various roughening procedures and optimizing the performance of the confocal Raman microscope. An approach of replacing the randomly roughened surface with ordered nanorod arrays of transition metals is introduced as a promising class of highly SERS-active substrates. The surface enhancement fa...

NTIRE 2017 Challenge on Single Image Super-Resolution: Methods and Results

Abstract: This paper reviews the first challenge on single image super-resolution (restoration of rich details in an low resolution image) with focus on proposed solutions and results. A new DIVerse 2K resolution image dataset (DIV2K) was employed. The challenge had 6 competitions divided into 2 tracks with 3 magnification factors each. Track 1 employed the standard bicubic downscaling setup, while Track 2 had unknown downscaling operators (blur kernel and decimation) but learnable through low and high res train images. Each competition had ∽100 registered participants and 20 teams competed in the final testing phase. They gauge the state-of-the-art in single image super-resolution.

Microbial production of recalcitrant dissolved organic matter: long-term carbon storage in the global ocean.

Abstract: The biological pump is a process whereby CO(2) in the upper ocean is fixed by primary producers and transported to the deep ocean as sinking biogenic particles or as dissolved organic matter. The fate of most of this exported material is remineralization to CO(2), which accumulates in deep waters until it is eventually ventilated again at the sea surface. However, a proportion of the fixed carbon is not mineralized but is instead stored for millennia as recalcitrant dissolved organic matter. The processes and mechanisms involved in the generation of this large carbon reservoir are poorly understood. Here, we propose the microbial carbon pump as a conceptual framework to address this important, multifaceted biogeochemical problem.