No abstract

Two-stream CNN is a widely-used network for human action recognition. Two-stream CNN consists of a spatial stream and a temporal stream. The spatial stream, through which the RGB image passes, extracts the shape features of human motion. The temporal stream, through which the optical flow images pass, extracts the sequence features of the listed motions. However, because of the constraints of the optical flow, such as brightness, constancy, and piecewise smoothness, there are limitations to the performance of two-stream CNN. One of the efficient methods to solve this problem is to expand the network model to a three-stream network, fuse it with LSTM, and add a modified pooling layer. This method improves the performance of the model but it increases the computational cost. Besides, the limitations of the optical flow are still present. In this paper, without extending the network model, a binary dense SIFT flow-based two-stream CNN is used instead of the optical flow. Unlike the optical flow, binary dense SIFT flow, which is a feature-based matching flow field is robust in brightness, constancy and piecewise smoothness. To evaluate the binary dense SIFT flow-based two-stream CNN, the UCF-101 dataset was selected for human action recognition. Furthermore, to evaluate the robustness of its brightness constancy and piecewise smoothness, a custom dataset was made up of classes that were extracted from UCF-101. Finally, the proposed method was compared with the state-of-the-art, which uses an optical flow-based two-stream CNN.

This paper is aimed at the investigation of flow and oxygen-transfer characteristics in an aeration system using an annular nozzle ejector, and the experimental evaluation of the oxygen-transfer characteristics in the ejector aeration and blower aeration. The entrainment ratio decreased with the primary water flow rate of the annular nozzle ejector, with ratios ranging between 6.8 and 0.4. It was found that the turbulence level and entrainment ratio strongly affected the air bubble size and the volumetric mass-transfer coefficient. The saturation times and volumetric mass-transfer coefficients varied with the suction air flow rate in the ejector aeration, while the times decreased and the coefficients increased with the blowing air flow rate in the blower aeration. The average mass-transfer coefficient of the ejector aeration was about 3.7 times higher than that of the blower aeration. It was found that the high turbulence level and optimum entrainment ratio were needed to increase the oxygen-transfer rate.

The objective of this study is to experimentally investigate the mixed flow behaviors and oxygen transfer characteristics of a vertical orifice ejector. The experimental apparatus consisted of an electric motor-pump, an orifice ejector, a circulation water tank, an air compressor, a high speed camera unit and control or measurement accessories. The mass ratio was calculated using the measured primary flow rate and suction air flow rate with experimental parameters. The visualization images of vertically injected mixed jet issuing from the orifice ejector were qualitatively analyzed. The volumetric oxygen transfer coefficient was calculated using the measured dissolved oxygen concentration. At a constant primary flow rate, the mass ratio and oxygen transfer coefficient increase with the air pressure of compressor. At a constant air pressure of the compressor, the mass ratio decreases and the oxygen transfer coefficient increases as the primary flow rate increases. The residence time and dispersion of fine air bubbles and the penetration of mixed flow were found to be important parameters for the oxygen transfer rate owing to the contact area and time of two phases.

No abstract