Comparison of CO2 Reduction Performance with NH3 and H2O between Cu/TiO2 and Pd/TiO2

The aim of this study is to clarify the effect of doped metal type on CO2 reduction characteristics of TiO2 with NH3 and H2O. Cu and Pd have been selected as dopants for TiO2. In addition, the impact of molar ratio of CO2 to reductants NH3 and H2O has been investigated. A TiO2 photocatalyst was prepared by a sol-gel and dip-coating process, and then doped with Cu or Pd fine particles by using the pulse arc plasma gun method. The prepared Cu/TiO2 film and Pd/TiO2 film were characterized by SEM, EPMA, TEM, STEM, EDX, EDS and EELS. This study also has investigated the performance of CO2 reduction under the illumination condition of Xe lamp with or without ultraviolet (UV) light. As a result, it is revealed that the CO2 reduction performance with Cu/TiO2 under the illumination condition of Xe lamp with UV light is the highest when the molar ratio of CO2/NH3/H2O = 1:1:1 while that without UV light is the highest when the molar ratio of CO2/NH3/H2O = 1:0.5:0.5. It is revealed that the CO2 reduction performance of Pd/TiO2 is the highest for the molar ratio of CO2/NH3/H2O = 1:1:1 no matter the used Xe lamp was with or without UV light. The molar quantity of CO per unit weight of photocatalyst for Cu/TiO2 produced under the illumination condition of Xe lamp with UV light was 10.2 μmol/g, while that for Pd/TiO2 was 5.5 μmol/g. Meanwhile, the molar quantity of CO per unit weight of photocatalyst for Cu/TiO2 produced under the illumination condition of Xe lamp without UV light was 2.5 μmol/g, while that for Pd/TiO2 was 3.5 μmol/g. This study has concluded that Cu/TiO2 is superior to Pd/TiO2 from the viewpoint of the molar quantity of CO per unit weight of photocatalyst as well as the quantum efficiency.

ICONet: A Lightweight Network with Greater Environmental Adaptivity

With the increasing popularity of artificial intelligence, deep learning has been applied to various fields, especially in computer vision. Since artificial intelligence is migrating from cloud to edge, deep learning nowadays should be edge-oriented and adaptive to complex environments. Aiming at these goals, this paper proposes an ICONet (illumination condition optimized network). Based on OTSU segmentation algorithm and fuzzy c-means clustering algorithm, the illumination condition classification subnet increases the environmental adaptivity of our network. The reduced time complexity and optimized size of our convolutional neural network (CNN) model enables the implementation of ICONet on edge devices. In the field of fatigue driving, we test the performance of ICONet on YawDD and self-collected datasets. Our network achieves a general accuracy of 98.56% and our models are about 590 kilobytes. Compared to other proposed networks, the ICONet shows significant success and superiority. Applying ICONet to fatigue driving detection is helpful to solve the symmetry of the needs of edge-oriented detection under complex illumination condition environments and the scarcity of related approaches.

Detecting Plant Invasion in Urban Parks with Aerial Image Time Series and Residual Neural Network

Invasive plants are a major agent threatening biodiversity conservation and directly affecting our living environment. This study aims to evaluate the potential of deep learning, one of the fastest-growing trends in machine learning, to detect plant invasion in urban parks using high-resolution (0.1 m) aerial image time series. Capitalizing on a state-of-the-art, popular architecture residual neural network (ResNet), we examined key challenges applying deep learning to detect plant invasion: relatively limited training sample size (invasion often confirmed in the field) and high forest contextual variation in space (from one invaded park to another) and over time (caused by varying stages of invasion and the difference in illumination condition). To do so, our evaluations focused on a widespread exotic plant, autumn olive (Elaeagnus umbellate), that has invaded 20 urban parks across Mecklenburg County (1410 km2) in North Carolina, USA. The results demonstrate a promising spatial and temporal generalization capacity of deep learning to detect urban invasive plants. In particular, the performance of ResNet was consistently over 96.2% using training samples from 8 (out of 20) or more parks. The model trained by samples from only four parks still achieved an accuracy of 77.4%. ResNet was further found tolerant of high contextual variation caused by autumn olive’s progressive invasion and the difference in illumination condition over the years. Our findings shed light on prioritized mitigation actions for effectively managing urban invasive plants.