To tackle the negative effect of the arbitrary crowd distribution on the counting task, in this article, we propose a novel RGB-D crowd counting approach, including a cross-modal cycle-attention fusion (CmCaF) model and a novel fine-coarse (FC) supervision. In the feature level, the CmCaF model combines the RGB feature and depth feature in a cycle-attention way so as to model the crowd distribution effectively. In the supervision level, the novel design of FC supervision could optimize the counting model from both the fine pixel-aware level and coarse region-aware level to enhance its sensitivity to the whole crowd distribution and the instance location. Extensive evaluations on benchmarks well illustrate the feasibility of the proposed approach for the RGB-D crowd counting, as well as RGB and RGB-T counting. And the ablation study demonstrates the effectiveness of its main components on both the feature representation of cross-modal data and the accurate estimation of the crowd distribution.

From the actual development and testing requirements of horizontal wells in domestic oilfields, according to the problems that the common sensors have relatively low resolution during the water cut measurement for the oil–water flow in the horizontal wells with medium and low production, a trielectrode capacitance sensor (TECS) is designed and a water cut measurement system based on TECS is developed for different conditions of horizontal oil–water flow. Specifically, the electrostatic field distribution of the annular measurement region is investigated under different conditions of horizontal stratified flow. Then, its equivalent capacitance on the electrodes is analyzed. Based on the simulated results, the effects of structural parameters on sensitivity distribution are analyzed, including the electrode length, the radius of the flow channel, and the thickness of the insulation, determining its optimal structure. Moreover, the static and dynamic experiments, as well as the field testing, validate the designed TECS, indicating that the TECS enables measuring the water cut of horizontal oil–water flow, especially under low flow rate and oil–water stratified conditions. The developed TECS could realize the water cut measurement with a wide flow rate range (3–60 m $^3$ /d) and the full water cut range (0%–100%) for the horizontal oil–water flow.

This paper proposes an automatic scale-adaptive approach with attention mechanism-based crowd spatial information addressing the crowd counting task, i.e. a novel cascaded crowd counting network. The proposed network is composed of a classification sub-network to estimate crowd scales and a main network to predict the corresponding density maps. Firstly, the image serves as the input of the classification network and the main network. Secondly, according to the estimated crowd scale results, the main network structure is adjusted; simultaneously, the feature in intermediate layer of the classification network is added stepwise into the main crowd counting network. Then, the semantic feature of the classification network is converted into the crowd spatial information mask via the proposed Spatial Attention Conversion module, and the crowd spatial information mask is weighted into the specific feature of the main network. Lastly, the crowd density map and the crowd counting result are obtained. Experiments on challenging Mall, ShanghaitechA and ShanghaitechB datasets prove the effectiveness, feasibility and robustness of the proposed method, and the ablation study demonstrates the structure rationality of the proposed network.