YOLSO: You Only Look Small Object

Small object detection is challenging and far from satisfactory. Most general object detectors suffer from two critical issues with small objects: (1) Feature extractor based on classification network cannot express the characteristics of small objects reasonably due to insufficient appearance information of targets and a large amount of background interference around them. (2) The detector requires a much higher location accuracy for small objects than for general objects. This paper proposes an effective and efficient small object detector YOLSO to address the above problems. For feature representation, we analyze the drawbacks in previous backbones and present a Half-Space Shortcut(HSSC) module to build a background-aware backbone. Furthermore, a coarse-to-fine Feature Pyramid Enhancement(FPE) module is introduced for layer-wise aggregation at a granular level to enhance the semantic discriminability. For loss function, we propose an exponential L1 loss to promote the convergence of regression, and a focal IOU loss to focus on prime samples with high classification confidence and high IOU. Both of them significantly improves the location accuracy of small objects. The proposed YOLSO sets state-of-the-art results on two typical small object datasets, MOCOD and VeDAI, at a speed of over 200 FPS. In the meantime, it also outperforms the baseline YOLOv3 by a wide margin on the common COCO dataset.     

The effects of washing a collagen sample prior to TEM examination

Transmission electron microscopy (TEM) is an important analysis technique to visualize (bio)macromolecules and their assemblies, including collagen fibers. Many protocols for TEM sample preparation of collagen involve one or more washing steps to remove excess salts from the dispersion that could hamper analysis when dried on a TEM grid. Such protocols are not standardized and washing times as well as washing solvents vary from procedure to procedure, with each research group typically having their own protocol. Here, we investigate the influence of washing with water, ethanol, but also methanol and 2-propanol, for both mineralized and unmineralized collagen samples via a protocol based on centrifugation. Washing with water maintains the hydrated collagen structure and the characteristic banding pattern can be clearly observed. Conversely, washing with ethanol results in dehydration of the fibrils, often leading to aggregation of the fibers and a less obvious banding pattern, already within 1 min of ethanol exposure. As we show, this process is fully reversible. Similar observations were made for methanol and propanol. Based on these results, a standardized washing protocol for collagenous samples is proposed.     

Effect of high hydrostatic pressure pretreatment on flavour and physicochemical properties of freeze-dried carambola slices

Ripe carambolas are hard to store and transport, while freeze-dried ones are easy to store. However, its long production time leads to higher costs. This study shows that high hydrostatic pressure (HHP) treatment could shorten the freeze-drying time of carambola slices. After HHP treatment (25–250 MPa), the drying time of the fresh sample can be shortened by 33.3–44.4% and the distribution of water and pigment in tissues is much uniform. With the increment of the pressure, 2,2-diphenyl-1-picrylhydrazyl (DPPH) and hydroxyl radical scavenging rate are increased. At 250 MPa, the total phenolic content (TPC) increased from 11.34 to 13.36 mg GAE g⁻¹, and the total flavonoid content (TFC) of the control sample was increased from 10.77 to 12.73 mg RE g⁻¹. Compared with the untreated sample, HHP treatment can enhance the flavour and shorten the freeze-drying time. This work guides the application of HHP technology for drying food processing.     

Tabular data: Deep learning is not all you need

A key element in solving real-life data science problems is selecting the types of models to use. Tree ensemble models (such as XGBoost) are usually recommended for classification and regression problems with tabular data. However, several deep learning models for tabular data have recently been proposed, claiming to outperform XGBoost for some use cases. This paper explores whether these deep models should be a recommended option for tabular data by rigorously comparing the new deep models to XGBoost on various datasets. In addition to systematically comparing their performance, we consider the tuning and computation they require. Our study shows that XGBoost outperforms these deep models across the datasets, including the datasets used in the papers that proposed the deep models. We also demonstrate that XGBoost requires much less tuning. On the positive side, we show that an ensemble of deep models and XGBoost performs better on these datasets than XGBoost alone.     

LinkingPark: An automatic semantic table interpretation system

In this paper, we present LinkingPark, an automatic semantic annotation system for tabular data to knowledge graph matching. LinkingPark is designed as a modular framework which can handle Cell-Entity Annotation (CEA), Column-Type Annotation (CTA), and Columns-Property Annotation (CPA) altogether. It is built upon our previous SemTab 2020 system, which won the 2nd prize among 28 different teams after four rounds of evaluations. Moreover, the system is unsupervised, stand-alone, and flexible for multilingual support. Its backend offers an efficient RESTful API for programmatic access, as well as an Excel Add-in for ease of use. Users can interact with LinkingPark in near real-time, further demonstrating its efficiency.     

广义正交码索引调制系统的性能分析

摘 要：为了提高码索引调制（code index modulation，CIM）系统的传输效率，提出了一种具有更低复杂度的单输入单输出（single input single output，SISO）的广义正交码索引调制（generalized orthogonal code index modulation，GQCIM）系统。CIM 系统使用扩频码和星座符号传输信息，但只能激活两个扩频码索引和一个调制符号。而 GQCIM 系统以一种新颖的方式克服了只激活一个调制符号的限制，同时充分利用了调制符号的正交性，增加扩频码索引以传输更多的额外信息位，提高了系统的传输效率。此外，分析了GQCIM系统的理论性能，推导了误码率性能的上界。通过蒙特卡罗仿真验证了GQCIM系统的性能，对比发现GQCIM系统的理论和仿真性能一致。而且在相同的传输效率下，结果显示GQCIM系统的性能优于同样具有正交性的调制系统，如广义码索引调制（generalized code index modulation，GCIM）系统、CIM系统、码索引调制-正交空间调制（code index modulation aided quadrature spatial modulation，CIM-QSM）系统、码索引调制-正交空间调制（code index modulation aided spatial modulation，CIM-SM）系统、脉冲索引调制（pulse index modulation，PIM）系统。     

磨料有序排布电镀砂轮的制备及其把持力研究

为解决电镀砂轮磨削加工中容屑空间不足的问题，采用点胶微粘接的方法制备了磨料有序排布的电镀砂轮，分析了磨料粘接效果和镀层力学性能。通过SEM分析了磨料/镀层/导电胶的结合界面，并进行了干磨削试验。研究结果表明，直径约为磨料粒径40%的胶点可粘接住磨料，单个胶点上粘接多颗磨料的占比小于6%；双脉冲电镀工艺制备的镀层显微硬度大于500HV，表层残余应力小于100MPa，磨料/镀层/导电胶之间的界面贴合紧密，无明显缺陷；砂轮在磨削时没有出现磨料脱落现象。     

Effects of the Anti-Tumorigenic Agent AT101 on Human Glioblastoma Cells in the Microenvironmental Glioma Stem Cell Niche

Glioblastoma (GBM) is a barely treatable disease due to its profound chemoresistance. A distinct inter- and intratumoral heterogeneity reflected by specialized microenvironmental niches and different tumor cell subpopulations allows GBMs to evade therapy regimens. Thus, there is an urgent need to develop alternative treatment strategies. A promising candidate for the treatment of GBMs is AT101, the R(-) enantiomer of gossypol. The present study evaluates the effects of AT101, alone or in combination with temozolomide (TMZ), in a microenvironmental glioma stem cell niche model of two GBM cell lines (U251MG and U87MG). AT101 was found to induce strong cytotoxic effects on U251MG and U87MG stem-like cells in comparison to the respective native cells. Moreover, a higher sensitivity against treatment with AT101 was observed upon incubation of native cells with a stem-like cell-conditioned medium. This higher sensitivity was reflected by a specific inhibitory influence on the p-p42/44 signaling pathway. Further, the expression of CXCR7 and the interleukin-6 receptor was significantly regulated upon these stimulatory conditions. Since tumor stem-like cells are known to mediate the development of tumor recurrences and were observed to strongly respond to the AT101 treatment, this might represent a promising approach to prevent the development of GBM recurrences.     

Benchmark experiment of large-angle scattering reaction cross section of iron at 14 MeV using two shadow bars – Comparison of experimental results with ENDF/B-VIII –

ABSTRACT

It is important to perform neutron transport simulations with accurate nuclear data in the neutronics design of a fusion reactor. However, absolute values of large-angle scattering cross sections vary among nuclear data libraries even for well-examined nuclide of iron. Benchmark experiments focusing on large-angle scattering cross sections were thus performed to confirm the correctness of nuclear data libraries. The series benchmark experiments were performed at a DT neutron source facility, OKTAVIAN of Osaka University, Japan, by the unique experimental system established by the authors’ group, which can extract only the contribution of large-angle scattering reactions. This system consists of two shadow bars, target plate (iron), and neutron detector (niobium). Two types of shadow bars were used and four irradiations were conducted for one experiment, so that contribution of room-return neutrons was effectively removed and only large-angle scattering neutrons were extracted from the measured four Nb reaction rates. The obtained experimental results were compared with calculations for five nuclear data libraries including JENDL-4.0, JEFF.-3.3, FENDL-3.1, ENDF/B- VII, and recently released ENDF/B-VIII. It was found from the comparison that ENDF/B-VIII showed the best result, though ENDF/B-VII showed overestimation and others are in large underestimation at 14 MeV.     

基于K-means和SOM的水下传感器数据采集方法

随着海洋资源勘探和海洋污染物监控工作的开展，水文数据的监测和采集等已经成为重要的研究方向。其中，水下无线传感器网络在水文数据采集过程中起着举足轻重的作用。本文研究的是水下无线传感器二维监测网络模型中，传感器节点数据采集的问题，其设计方法是通过自组织映射（Self-organizing mapping，SOM）对传感器节点进行路径最优化处理，结合优化的路径图形和K-means算法找到路径内部聚合点，利用聚合点和传感器的节点得到传感器通信半径内的数据采集点，最后通过SOM得到水下机器人（Autonomous underwater vehicle，AUV）到各个数据采集点采集数据的最优路径。经过实验验证，在水下1 200 m×1 750 m范围内布置52个传感器节点的情景下，数据采集点相比于传感器节点路径规划采用相同的采集顺序得到的路径优化了6.7%；对数据采集点重新进行自组织路径规划得到的路径比传感器结点路径的最优解提高了12.2%。增加传感器节点的数量，其结果也大致相同，因此采用该方法可以提高水下机器人采集数据的效率。