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 decreasing effect of ammonia enrichment on the combustion emission of hydrogen,methane, and propane fuels

In the present study, non-premixed combustion and NOx emission of H₂, NH₃, C₃H₈, and CH₄ fuels have been studied in a combustion test unit under lean mixture conditions (λ = 4) at 8.6 kW thermal capacity. Furthermore, the combustion and NOx emission of the H₂, C₃H₈, and CH₄ fuels have been investigated for various NH₃ enrichment ratios (5, 10, 20, and 50%) and excess air coefficients (λ = 1.1, 2, 3, and 4) at the same thermal capacity. The obtained results have been compared for each fuel. Numerical simulation results show that H₂ emits intense energy through the reaction zone despite the lowest fuel consumption in mass, among others, due to its high calorific value. Therefore, it has a higher flame temperature than others. At the same time, C₃H₈ has the lowest flame temperature. Besides, NH₃ has the shortest flame length among others, while C₃H₈ has the most extended flame form. The highest level of NOx is released from the NH₃ flame in the combustion chamber, while the lowest NOx is released from the CH₄. However, the lowest NOx emission at the combustion chamber exit is obtained in NH₃ combustion, while the highest NOx emission is obtained with H₂ combustion. It results from the shortest flame length of NH₃, short residence time, and backward NOx reduction to N₂ for NH₃. As for H₂, high flame temperature and relatively long flame, and high residence time of the products trigger NOx formation and keep the NOx level high. On the other hand, excess air coefficient from 1.1 to 2 increases NOx for H₂, CH₄, and NH₃ due to their large flame diameters, unlike propane. Then, NOx emission levels decrease sharply as the excess air coefficient increases to 4 for each fuel. NH₃ fuel also emits minimum NOx in other excess air coefficients at the exit, while H₂ emits too much emission. With NH₃ enrichment, the NOx emissions of H₂, CH₄, and C₃H₈ fuels at the combustion chamber exit decrease gradually almost every excess air coefficient apart from λ = 1.1. As a general conclusion, like renewable fuels, H₂ appears to be a source of pollution in terms of NOx emissions in combustion applications. In contrast, NH₃ appears to be a relatively modest fuel with a low NOx level. In addition, the high amount of NOx emission released from H₂ and other fuels during the combustion can be remarkably reduced by NH₃ enrichment with an excess air combustion.     

Investigation of La0.6Sr0.4Co1-xNixO3-δ (x=0, 0.2, 0.4, 0.6, 0.8) catalysts on solid oxide fuel cells anode for biogas dry reforming

The introduction of catalyst on anode of solid oxide fuel cell (SOFC) has been an effective way to alleviate the carbon deposition when utilizing biogas as the fuel. A series of La_0.6Sr_0.4Co_1-xNi_xO_3-δ (x = 0, 0.2, 0.4, 0.6, 0.8) oxides are synthesized by sol-gel method and used as catalysts precursors for biogas dry reforming. The phase structure of La_0.6Sr_0.4Co_1-xNi_xO_3-δ oxides before and after reduction are characterized by X-ray diffraction (XRD). The texture properties, carbon deposition, CH₄ and CO₂ conversion rate of La_0.6Sr_0.4Co_1-xNi_xO_3-δ catalysts are evaluated and compared. The peak power density of 739 mW cm^?2 is obtained by a commercial SOFC with La_0.6Sr_0.4Co_0.4Ni_0.6O_3-δ catalyst at 850 °C when using a mixture of CH₄: CO₂ = 2:1 as fuel. This shows a great improvement from the cell without catalyst for internal dry reforming, which is attributed to the formation of NiCo alloy active species after reduction in H₂ atmosphere. The results indicate the benefits of inhibiting the carbon deposition on Ni-based anode through introducing the La_0.6Sr_0.4Co_0.4Ni_0.6O_3-δ catalyst precursor. Additionally, the dry reforming technology will also help to convert part of the exhaust heat into chemical energy and improve the efficiency of SOFC system with biogas fuel.     

用于电弧检测的电力专用SoC设计

针对传统的电弧电路故障检测结果不准确的问题,设计用于电弧检测的SoC系统,并且在55nm工艺下进行流片验证。采用包含两种结构的模数转换器的片上电压源,设计了锁相环以及复位电路,精度最高可达8.67 bit。验证结果表明,本设计可提高电弧检测的准确性。     

基于误差估值累加开环校正的诱导式欺骗检测方法

诱导式卫星欺骗干扰可诱导航空器逐渐偏离预定航迹，难以被发现，因此及时有效地检测干扰是飞行安全的保障。在现有紧组合导航体制基础上，设计了一种基于误差估值累加开环校正的紧组合导航结构，并证明了其性能与传统闭环校正紧组合导航性能等效。在此结构中，将紧组合导航系统与自适应序贯概率比检测方法结合，提出了一种基于误差估值累加开环校正的诱导式欺骗检测方法，融合紧组合导航信息与其他不受欺骗影响的导航信息，构建欺骗检测统计量进行诱导式欺骗检测。仿真结果表明，开环校正结构可避免随时间累加的惯性导航系统误差所导致的组合导航滤波器发散问题，同时欺骗检测方法可进一步提高算法对“最坏”情形下微小诱导式欺骗的检测效果。     

Catalytic behavior in CH4 decomposition of catalysts derived from red mud: Impact of residual Na2O

Catalyst samples for CH₄ decomposition were prepared from red mud (RM) by an acid-leaching neutralization precipitation approach. Water-washing the resultant precipitates multiple times, followed by drying at 105 °C and calcination at 500 °C, resulted in a threshold of residual Na₂O, equivalent to 96% Na₂O removal. Drying the precipitate at a higher temperature of 200 °C, followed by repeated water washing, provided a deeper Na₂O removal of 99% and made the resultant samples more active for the targeted reaction. Subsequently, four catalyst samples with a simulated red mud composition and NaOH contents from 0 to 0.3 wt% were prepared and the catalytic test results revealed that the Na₂O remaining in the RM-derived catalysts did not only inhibit their activation in CH₄ but also lower their maximal activities for CH₄ decomposition. Finally, two catalysts with the same simulated red mud composition and their Na impregnated respectively on Fe₂O₃ and a mixture support of Al₂O₃-SiO₂-TiO₂ were prepared and tested to explore the effect of Na distribution on the activation behavior of RM-derived catalysts for CH₄ decomposition. The activity testing results showed that it was the Na residual dispersed on iron oxides in the RM-derived samples to significantly inhibit the activation of CH₄ decomposition.     

Applications of sketches in network traffic measurement: A survey

Accurate and timely network traffic measurement is essential for network status monitoring, network fault analysis, network intrusion detection, and network security management. With the rapid development of the network, massive network traffic brings severe challenges to network traffic measurement. However, existing measurement methods suffer from many limitations for effectively recording and accurately analyzing big-volume traffic. Recently, sketches, a family of probabilistic data structures that employ hashing technology for summarizing traffic data, have been widely used to solve these problems. However, current literature still lacks a thorough review on sketch-based traffic measurement methods to offer a comprehensive insight on how to apply sketches for fulfilling various traffic measurement tasks. In this paper, we provide a detailed and comprehensive review on the applications of sketches in network traffic measurement. To this end, we classify the network traffic measurement tasks into four categories based on the target of traffic measurement, namely cardinality estimation, flow size estimation, change anomaly detection, and persistent spreader identification. First, we briefly introduce these four types of traffic measurement tasks and discuss the advantages of applying sketches. Then, we propose a series of requirements with regard to the applications of sketches in network traffic measurement. After that, we perform a fine-grained classification for each sketch-based measurement category according to the technologies applied on sketches. During the review, we evaluate the performance, advantages and disadvantages of current sketch-based traffic measurement methods based on the proposed requirements. Through the thorough review, we gain a number of valuable implications that can guide us to choose and design proper traffic measurement methods based on sketches. We also review a number of general sketches that are highly expected in modern network systems to simultaneously perform multiple traffic measurement tasks and discuss their performance based on the proposed requirements. Finally, through our serious review, we summarize a number of open issues and identify several promising research directions.     

基于全U网络的混凝土路面裂缝检测算法

针对现有的混凝土裂缝检测算法在各种复杂环境中检测精度不够、鲁棒性不强的问题,根据深度学习理论和U-net模型,提出一种全U型网络的裂缝检测算法。首先,依照原U-net模型路线构建网络;然后,在每个池化层后都进行一次上采样,恢复其在池化层之前的特征图规格,并将其与池化之前的卷积层进行融合,将融合之后的特征图作为新的融合层与原U-net网络上采样之后的网络层进行融合;最后,为了验证算法的有效性,在测试集中进行实验。结果表明,所提算法的平均精确率可达到83.48%,召回率为85.08%,F1为84.11%,相较于原U-net分别提升了1.48%,4.68%和3.29%,在复杂环境中也能提取完整裂缝,保证了裂缝检测的鲁棒性。     

A rapid detection method for the surface defects of mosaic ceramic tiles

Due to its unique artistic value, mosaic ceramics are widely used in construction-related fields. To meet the artist's demand for high-quality mosaic ceramic to create artistic works, it is necessary to meet the needs for efficient screening of mosaic ceramic tiles. Different from the ordinary large-target ceramics, mosaic ceramics exhibit characteristics of small tile sizes, a variety of colors, large demand for quantities, and easy reflection on the surface. Common manual detection methods show problems of low efficiency or accuracy, easy to fatigue, and many others. To solve these problems, this paper proposes a new detection method to identify surface defects of mosaic ceramic tiles and designs a detection system platform to achieve rapid detection. The experiment proves that the detection system has a detection rate of 93.99% for small defects on the surface of mosaic ceramic tiles, and the detection time of a single mosaic ceramic tile is less than 0.06 s. The detection method can quickly and accurately screen out high-quality, defect-free mosaic ceramic tiles, which can effectively improve the quality and artistic value of mosaic ceramic art creation.