Search

In this paper, novel humidity sensors based on montmorillonite, kaolinite, and composite films coated on micro-cantilevers were prepared to measure the relative humidity (RH) values by the deflection of a micro-cantilever (MC) at room temperature. The humidity-sensing properties, such as response and recovery, sensitivity, repeatability, humidity hysteresis, and long-term stability, were investigated in the range of working humidity (10–80% RH). The humidity response in the close humidity range of 10% RH to 80% RH revealed a linear increase in water absorption of montmorillonite, kaolinite, and montmorillonite/kaolinite mixed dispersant (1:1) as a function of RH with linear correlation factors between the humidity change and deflection estimated to be 0.994, 0.991, and 0.946, respectively. Montmorillonite's sensitivity was better than kaolinite's, with the mixed-clay mineral film's response falling somewhere in between. This research provides a feasible and effective approach to constructing high-performance MC humidity sensors that can be operated at room temperature based on clay minerals. [ABSTRACT FROM AUTHOR]

For a long time, object detection has been a popular but difficult research problem in the field of pattern recognition. In recent years, object detection algorithms based on convolutional neural networks have achieved excellent results. However, neural networks are computationally intensive and parameter redundant, so they are difficult to deploy on resource-limited embedded devices. Especially for two-stage detectors, operations and parameters are mainly clustered on feature fusion of proposals after the region of interest (ROI) pooling layer, and they are enormous. In order to deal with these problems, we propose a subnetwork—efficient feature fusion module (EFFM) to reduce the number of operations and parameters for a two-stage detector. In addition, we propose a multi-scale dilation region proposal network (RPN) to further improve detection accuracy. Finally, our accuracy is higher than Faster RCNN based on VGG16, the number of operations is only half of the latter, and the number of parameters is only one third. [ABSTRACT FROM AUTHOR]