We synthesized g-C3N4/nano-InVO4 heterojunction-type photocatalyts by in situ growth of InVO4 nanoparticles onto the surface of g-C3N4 sheets via a hydrothermal process. The results of SEM and TEM showed that the obtained InVO4 nanoparticles 20 nm in size dispersed uniformly on the surface of g-C3N4 sheets, which revealed that g-C3N4 sheets was probably a promising support for in situ growth of nanosize materials. The achieved intimate interface promoted the charge transfer and inhibited the recombination rate of photogenerated electron-hole pairs, which significantly improved the photocatalytic activity. A possible growth process of g-C3N4/nano-InVO4 nanocomposites was proposed based on different mass fraction of g-C3N4 content. The obtained g-C3N4/nano-InVO4 nanocomposites could achieve effective separation of charge-hole pairs and stronger reducing power, which caused enhanced H2 evolution from water-splitting compared with bare g-C3N4 sheets and g-C3N4/micro-InVO4 composites, respectively. As a result, the g-C3N4/nano-InVO4 nanocomposite with a mass ratio of 80:20 possessed the maximum photocatalytic activity for hydrogen production under visible-light irradiation.

Compared with single image based crowd counting, video provides the spatial-temporal information of the crowd that would help improve the robustness of crowd counting. But translation, rotation and scaling of people lead to the change of density map of heads between neighbouring frames. Meanwhile, people walking in/out or being occluded in dynamic scenes leads to the change of head counts. To alleviate these issues in video crowd counting, a Locality-constrained Spatial Transformer Network (LSTN) is proposed. Specifically, we first leverage a Convolutional Neural Networks to estimate the density map for each frame. Then to relate the density maps between neighbouring frames, a Locality-constrained Spatial Transformer (LST) module is introduced to estimate the density map of next frame with that of current frame. To facilitate the performance evaluation, a large-scale video crowd counting dataset is collected, which contains 15K frames with about 394K annotated heads captured from 13 different scenes. As far as we know, it is the largest video crowd counting dataset. Extensive experiments on our dataset and other crowd counting datasets validate the effectiveness of our LSTN for crowd counting. All our dataset are released in https://github.com/sweetyy83/Lstn_fdst_ dataset.

No abstract

No abstract

An accurate measurement method to extract the common mode (CM) and the differential mode (DM) noise source impedances of a switched-mode power supply (SMPS) under its operating condition is developed and validated. With a proper premeasurement calibration process, the proposed method allows extraction of both the CM and the DM noise source impedances with very good accuracy. These noise source impedances come in handy to design an electromagnetic interference filter for an SMPS systematically with minimum hassle.