面向分拣机器人的珍珠形状视觉检测方法

针对人工进行珍珠形状分拣效率低、精度不稳定等问题,提出基于机器视觉的珍珠形状检测方法;采用背光成像方式消除珍珠表面纹理和光泽的影响,对获取的珍珠图像进行同态滤波等预处理算法,提高图像对比度;为了解决相互接触珍珠影响珍珠轮廓提取的问题,采用分水岭算法对珍珠图像进行分割,得到了独立存在的珍珠个体,再通过连通域标记、质心算法对珍珠进行定位;根据国家标准对珍珠形状的规定,基于珍珠图像信息建立珍珠形状参数模型,对珍珠形状进行量化;实验结果表明,不同形状的珍珠样本的检测误差为0.63%,形状统计精度为100%,算法耗时24 ms;该方法可准确高效地对珍珠进行分拣分级,具有一定的实用价值。     

Extraction of protein from food waste: An overview of current status and opportunities

The chief intent of this review is to explain the different extraction techniques and efficiencies for the recovery of protein from food waste (FW) sources. Although FW is not a new concept, increasing concerns about chronic hunger, nutritional deficiency, food security, and sustainability have intensified attention on alternative and sustainable sources of protein for food and feed. Initiatives to extract and utilize protein from FW on a commercial scale have been undertaken, mainly in the developed countries, but they remain largely underutilized and generally suited for low-quality products. The current analysis reveals the extraction of protein from FW is a many-sided (complex) issue, and that identifies for a stronger and extensive integration of diverse extraction perspectives, focusing on nutritional quality, yield, and functionality of the isolated protein as a valued recycled ingredient.     

Image-Based Lifelogging: User Emotion Perspective

Lifelog is a digital record of an individual’s daily life. It collects, records, and archives a large amount of unstructured data; therefore, techniques are required to organize and summarize those data for easy retrieval. Lifelogging has been utilized for diverse applications including healthcare, self-tracking, and entertainment, among others. With regard to the image-based lifelogging, even though most users prefer to present photos with facial expressions that allow us to infer their emotions, there have been few studies on lifelogging techniques that focus upon users’ emotions. In this paper, we develop a system that extracts users’ own photos from their smartphones and configures their lifelogs with a focus on their emotions. We design an emotion classifier based on convolutional neural networks (CNN) to predict the users’ emotions. To train the model, we create a new dataset by collecting facial images from the CelebFaces Attributes (CelebA) dataset and labeling their facial emotion expressions, and by integrating parts of the Radboud Faces Database (RaFD). Our dataset consists of 4,715 high-resolution images. We propose Representative Emotional Data Extraction Scheme (REDES) to select representative photos based on inferring users’ emotions from their facial expressions. In addition, we develop a system that allows users to easily configure diaries for a special day and summaize their lifelogs. Our experimental results show that our method is able to effectively incorporate emotions into lifelog, allowing an enriched experience.     

长江经济带工业集聚对水环境污染的影响——基于空间计量与门槛效应的检验

经济发展与生态环境动态良性循环已经成为社会经济发展的重要议题。为实现长江经济带高质量发展，以2007—2018年长江经济带11个省市为样本，采用空间计量和面板门槛模型考察工业集聚与水环境污染之间的关系。研究表明：长江经济带各省市水环境污染具有显著的空间自相关性；长江经济带工业集聚水平和水环境污染表现出先降低后增强的“U”形变化特征，说明工业集聚水平对水环境污染的影响存在一个阈值；长江经济带工业集聚产生的规模效应和结构效应对水环境污染存在显著的双门槛效应，在不同的工业集聚水平下，规模效应和结构效应表现出不同的外部性特征；工业集聚产生的技术溢出效应对水环境污染的影响同样具有显著的双门槛作用；随着各地区工业集聚水平的提高，技术溢出效应对水环境污染的减排作用表现出先强后弱的趋势。     

Heat transfer trait simulation of H finned tube in ventilation methane oxidation steam generator for hydrogen production

With the steam obtained by the energy released from ventilation air methane oxidation, the hydrogen production through gasification method is considered more commercial. In order to constrain the steam parameter fluctuation, the shunt honeycomb ceramics are adopted to fill between the heat exchange tubes. However, transient heat transfer characteristics of this kind of heat exchanger have not been fully studied. This paper carried out a numerical simulation study on the transient heat transfer characteristics of H finned tube under periodic reverse-flow conditions. Results show that the existence of shunt honeycomb ceramics enhances the effect of radiation. Gas flow direction reversing destructs the original boundary layer, achieving a sudden rise of the convective heat flux in the new upstream. It takes about 12s to form a new relative stable boundary layer. The apparent heat transfer coefficient achieves a maximum of 108.77 W m⁻² K⁻¹ and an average of 99.1 W m⁻² K⁻¹.     

Effect of element Ru on microstructure and creep behaviour of single crystal nickel-based superalloy

By means of creep property measurement and microstructure observation, the influence of element Ru on the microstructure and creep behaviour of single crystal nickel-based superalloy is investigated. The results show that the plate-like μ phase is precipitated along {111} plane of the Ru-free alloy during thermal exposure at 1080°C and creep at 980°C. And the precipitation of μ phase in the alloy with 6(wt)%Mo and 6(wt)%W may be restrained by adding 2% Ru element. Compared to Ru-free alloy, the creep life of the 2% Ru single crystal alloy at 980°C200 MPa increases form 123 h to 333h. Compared to the 2% Ru alloy, the precipitated plate-like μ phase in the Ru-free alloy may promote the initiation and propagation of cracks along the γ matrix up to fracture, which is thought to be the main reason of the alloy having a lower creep resistances and shorter lifetime.     

基于改进网格运动统计特征的图像匹配算法

为了解决尺度不变特征变换（SIFT）算法在图像匹配中匹配正确率低、耗时长等问题，提出一种基于改进网格运动统计特征RANSAC-GMS的图像匹配算法。首先，利用快速旋转不变性特征（ORB）算法对图像进行预匹配，对预匹配的特征点采用网格运动统计（GMS）来支持估计量以实现正确匹配点与错误匹配点的区分；然后，采用改进的随机抽样一致性（RANSAC）算法通过匹配点间的距离相似性对特征点进行筛选，并采用评价函数对筛选后的新数据集进行重新整理，进而实现对误匹配点的剔除。采用Oxford标准图库和现实中拍摄的图像对图像匹配算法进行测试对比，实验结果表明，所提算法在图像匹配中的平均匹配正确率达到91%以上；与GMS、SIFT、ORB等算法相比，该改进算法的近景匹配正确率和远景匹配正确率分别最少提高了16.15个百分点和3.56个百分点，说明它能有效剔除误匹配点，进一步提高图像匹配精度。     

Distant Supervised Relation Extraction with Cost-Sensitive Loss

Recently, many researchers have concentrated on distant supervision relation extraction (DSRE). DSRE has solved the problem of the lack of data for supervised learning, however, the data automatically labeled by DSRE has a serious problem, which is class imbalance. The data from the majority class obviously dominates the dataset, in this case, most neural network classifiers will have a strong bias towards the majority class, so they cannot correctly classify the minority class. Studies have shown that the degree of separability between classes greatly determines the performance of imbalanced data. Therefore, in this paper we propose a novel model, which combines class-to-class separability and cost-sensitive learning to adjust the maximum reachable cost of misclassification, thus improving the performance of imbalanced data sets under distant supervision. Experiments have shown that our method is more effective for DSRE than baseline methods.