[Establishment and Application of Foshan Ozone Concentration Forecast Equation].

The formation and changes of ozone (O3), a secondary pollutant in the atmosphere, are complex, and ozone forecasting has become one of the current problems in air pollution prevention and control. In this study, the relationships between the near-surface O3 concentration and meteorological elements (high- and low-level) in Foshan from 2014 to 2017 were analyzed, and the concentration forecasting equation was established, tested, and applied. The results showed that the near-surface O3 changed closely related to high- and low-level meteorological elements. Meteorological elements such as temperature and sunshine hours were significantly positively correlated with O3 concentration, whereas relative humidity, total (low) cloud cover, and wind speed were negatively correlated with O3. Heavy O3 pollution often occurred with meteorological conditions of low wind speed, sunny days and few clouds, low relative humidity, longer sunshine time, and higher temperature. The definitions of high-concentration O3 potential index (HOPI) and wind direction index (WDI) in the Foshan area could better characterize the meteorological conditions of O3 pollution. Considering 13 meteorological elements, such as HOPI and WDI at different heights, the O3 concentration forecasting equation in the Foshan area was established using multi-indicator stacking and multiple stepwise regression methods. Using the 2018 data, it was found that the correlation coefficient R between the simulated values and the measured values reached 0.82, and the forecast equation had a good fitting effect and predictability.

Retinal blood vessel segmentation based on Densely Connected U-Net.

The segmentation of blood vessels from retinal images is an important and challenging task in medical analysis and diagnosis. This paper proposes a new architecture of the U-Net network for retinal blood vessel segmentation. Adding dense block to U-Net network makes each layer's input come from the all previous layer's output which improves the segmentation accuracy of small blood vessels. The effectiveness of the proposed method has been evaluated on two public datasets (DRIVE and CHASE_DB1). The obtained results (DRIVE： Acc = 0.9559， AUC = 0.9793， CHASE_DB1： Acc = 0.9488， AUC = 0.9785) demonstrate the better performance of the proposed method compared to the state-of-the-art methods. Also, the results show that our method achieves better results for the segmentation of small blood vessels and can be helpful to evaluate related ophthalmic diseases.