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.     

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.     

Feature selection methods for root-cause analysis among top-level product attributes

Manufacturing companies not only strive to deliver flawless products but also monitor product failures in the field to identify potential quality issues. When product failures occur, quality engineers must identify the root cause to improve any affected product and process. This root-cause analysis can be supported by feature selection methods that identify relevant product attributes, such as manufacturing dates with an increased number of product failures. In this paper, we present different methods for feature selection and evaluate their ability to identify relevant product attributes in a root-cause analysis. First, we compile a list of feature selection methods. Then, we summarize the properties of product attributes in warranty case data and discuss these properties regarding the challenges they pose for machine learning algorithms. Next, we simulate datasets of warranty cases, which emulate these product properties. Finally, we compare the feature selection methods based on these simulated datasets. In the end, the univariate filter information gain is determined to be a suitable method for a wide range of applications. The comparison based on simulated data provides a more general result than other publications, which only focus on a single use case. Due to the generic nature of the simulated datasets, the results can be applied to various root-cause analysis processes in different quality management applications and provide a guideline for readers who wish to explore machine learning methods for their analysis of quality data.     

Marine Phospholipids from Fishery By-Products Attenuate Atherosclerosis

In this study, the anti-atherosclerotic properties of three marine phospholipids (MPLs) extracts from fishery by-products including codfish roe, squid gonad, and shrimp head are verified. Their effects on key factors involved in atherosclerosis are examined and compared to explore whether the differences in their constitutions lead to the differences in the function. All three MPLs dampen oxidation of low- density lipoproteins (LDL) in vitro. Treating RAW264.7 macrophages and HUVECs endothelial cells with each MPLs ranging 10–100 µg mL⁻¹ does not decrease cell viability, yet ox-LDL caused cytotoxicity of both cells are alleviated by 50 or 100 µg mL⁻¹ MPLs treatment. In addition, the three MPLs reduce ox-LDL induced macrophage foam-like transition, mainly through inhibition of lipid uptake. Of the three MPLs, the one from squid gonad exhibits the best effect. On the other hand, all three MPLs modulate inflammatory responses, equally, by inhibiting the adhesion of monocytes to endothelial cells, and decreasing secretion of pro-inflammatory cytokines IL-6 and MCP-1. Using a high-cholesterol diet induced zebrafish model, it is found that all three MPLs, especially the one from squid gonad, alleviates cholesterol accumulation in early plaques, and decreases total cholesterol as well as lipid peroxide in vivo. Practical Applications: As a way of making the best of the increasingly scarce marine resources, valuable lipid components can be recovered from by-products and wastes from the fishery industry. Here, we tested the anti-atherosclerotic effects and the mechanisms of three MPLs extracted from codfish roe, squid gonad, and shrimp head. Our study provides further evidence that marine phospholipids extracted from fishery by-products could protect against atherosclerosis, and helps to elucidate the structure-function relationship of MPLs.     

Understanding the asymmetric perceptions of smartphone security from security feature perspective: A comparative study

Smartphones are being used and relied on by people more than ever before. The open connectivity brings with it great convenience and leads to a variety of risks that cannot be overlooked. Smartphone vendors, security policy designers, and security application providers have put a variety of practical efforts to secure smartphones, and researchers have conducted extensive research on threat sources, security techniques, and user security behaviors. Regrettably, smartphone users do not pay enough attention to mobile security, making many efforts futile. This study identifies this gap between technology affordance and user requirements, and attempts to investigate the asymmetric perceptions toward security features between developers and users, between users and users, as well as between different security features. These asymmetric perceptions include perceptions of quality, perceptions of importance, and perceptions of satisfaction. After scoping the range of smartphone security features, this study conducts an improved Kano-based method and exhaustively analyzes the 245 collected samples using correspondence analysis and importance satisfaction analysis. The 14 security features of the smartphone are divided into four Kano quality types and the perceived quality differences between developers and users are compared. Correspondence analysis is utilized to capture the relationship between the perceived importance of security features across different groups of respondents, and results of importance-satisfaction analysis provide the basis for the developmental path and resource reallocation strategy of security features. This article offers new insights for researchers as well as practitioners of smartphone security.     

Soft sensor for real-time cement fineness estimation

This paper describes the design and implementation of soft sensors to estimate cement fineness. Soft sensors are mathematical models that use available data to provide real-time information on process variables when the information, for whatever reason, is not available by direct measurement. In this application, soft sensors are used to provide information on process variable normally provided by off-line laboratory tests performed at large time intervals. Cement fineness is one of the crucial parameters that define the quality of produced cement. Providing real-time information on cement fineness using soft sensors can overcome limitations and problems that originate from a lack of information between two laboratory tests. The model inputs were selected from candidate process variables using an information theoretic approach. Models based on multi-layer perceptrons were developed, and their ability to estimate cement fineness of laboratory samples was analyzed. Models that had the best performance, and capacity to adopt changes in the cement grinding circuit were selected to implement soft sensors. Soft sensors were tested using data from a continuous cement production to demonstrate their use in real-time fineness estimation. Their performance was highly satisfactory, and the sensors proved to be capable of providing valuable information on cement grinding circuit performance. After successful off-line tests, soft sensors were implemented and installed in the control room of a cement factory. Results on the site confirm results obtained by tests conducted during soft sensor development.     

Questioning numerical integration methods for microsphere (and microplane) constitutive equations

In the last few years, more and more complex microsphere models have been proposed to predict the mechanical response of various polymers. Similarly than for microplane models, they consist in deriving a one-dimensional force vs. stretch equation and to integrate it over the unit sphere to obtain a three-dimensional constitutive equation. In this context, the focus of authors is laid on the physics of the one-dimensional relationship, but in most of the case the influence of the integration method on the prediction is not investigated.Here we compare three numerical integration schemes: a classical Gaussian scheme, a method based on a regular geometric meshing of the sphere, and an approach based on spherical harmonics. Depending on the method, the number of integration points may vary from 4 to 983,040! Considering simple quantities, i.e. principal (large) strain invariants, it is shown that the integration method must be carefully chosen. Depending on the quantities retained to described the one-dimensional equation and the required error, the performances of the three methods are discussed. Consequences on stress–strain prediction are illustrated with a directional version of the classical Mooney–Rivlin hyperelastic model. Finally, the paper closes with some advices for the development of new microsphere constitutive equations.     

基于深度学习的秀丽隐杆线虫显微图像分割方法

利用计算机实现自动、准确的秀丽隐杆线虫(C.elegans)的各项形态学参数分析,至关重要的是从显微图像上分割出线虫体态,但由于显微镜下的图像噪声较多,线虫边缘像素与周围环境相似,而且线虫的体态具有鞭毛和其他附着物需要分离,多方面因素导致设计一个鲁棒性的C.elegans分割算法仍然面临着挑战。针对这些问题,提出了一种基于深度学习的线虫分割方法,通过训练掩模区域卷积神经网络(Mask R-CNN)学习线虫形态特征实现自动分割。首先,通过改进多级特征池化将高级语义特征与低级边缘特征融合,结合大幅度软最大损失(LMSL)损失算法改进损失计算;然后,改进非极大值抑制;最后,引入全连接融合分支等方法对分割结果进行进一步优化。实验结果表明,相比原始的Mask R-CNN,该方法平均精确率(AP)提升了4.3个百分点,平均交并比(mIOU)提升了4个百分点。表明所提出的深度学习分割方法能够有效提高分割准确率,在显微图像中更加精确地分割出线虫体。     

Towards a Better Prediction of Cell Settling on Nanostructure Arrays—Simple Means to Complicated Ends

Vertical arrays of nanostructures (NSs) are emerging as promising platforms for probing and manipulating live mammalian cells. The broad range of applications requires different types of interfaces, but cell settling on NS arrays is not yet fully controlled and understood. Cells are both seen to deform completely into NS arrays and to stay suspended like tiny fakirs, which have hitherto been explained with differences in NS spacing or density. Here, a better understanding of this phenomenon is provided by using a model that takes into account the extreme membrane deformation needed for a cell to settle into a NS array. It is shown that, in addition to the NS density, cell settling depends strongly on the dimensions of the single NS, and that the settling can be predicted for a given NS array geometry. The predictive power of the model is confirmed by experiments and good agreement with cases from the literature. Furthermore, the influence of cell‐related parameters is evaluated theoretically and a generic method of tuning cell settling through surface coating is demonstrated experimentally. These findings allow a more rational design of NS arrays for the numerous exciting biological applications where the mode of cell settling is crucial.     

State-space ARIMA for supply-chain forecasting

ARIMA is seldom used in supply chains in practice. There are several reasons, not the least of which is the small sample size of available data, which restricts the usage of the model. Keeping in mind this restriction, we discuss in this paper a state-space ARIMA model with a single source of error and show how it can be efficiently used in the supply-chain context, especially in cases when only two seasonal cycles of data are available. We propose a new order selection algorithm for the model and compare its performance with the conventional ARIMA on real data. We show that the proposed model performs well in terms of both accuracy and computational time in comparison with other ARIMA implementations, which makes it efficient in the supply-chain context.