Video analytics is usually time-consuming as it not only requires video decoding as a first step but also usually applies complex computer vision and machine learning algorithms to the decoded frame. To achieve high efficiency in video analytics with ever increasing frame size, many researches have been conducted for distributed video processing using Hadoop. However, most approaches focused on processing multiple video files on multiple nodes. Such approaches require a number of video files to achieve any speedup, and could easily result in load imbalance when the size of video files is reasonably long since a video file itself is processed sequentially. In contrast, we propose a distributed video decoding method with an extended FFmpeg and VideoRecordReader, by which a single large video file can be processed in parallel across multiple nodes in Hadoop. The experimental results show that a case study of face detection and SURF system achieve 40.6 times and 29.1 times of speedups respectively on a four-node cluster with 12 mappers in each node, showing good scalability.
For developing subminiature and highly integrated multilayer inductors, soft magnetic powder was used; however, its ferrite magnetic component is characterized by high resistivity and reduced direct current saturation, leading to the deterioration of the inductor under high currents. Therefore, herein, to improve the electromagnetic properties of thin-film inductors, Fe nanopowder was used to increase the volume fraction of magnetic sheets. Surface treatment was performed by using silane coupling agents, which improved the bonding strength and dispersibility of the Fe nanopowder with a heterogeneous epoxy binder. For uniform surface treatment on the nanopowder, the silane-treated powder was aged for 24 h, at a temperature of 3 °C. The surface-treated Fe nanopowder was used with a mixing ratio of the soft magnetic powder (coarse:fine:nano) of 7:2.5:0.5 wt.%; this was successful in producing a flexible and highly densified magnetic sheet. As a result, the volume fraction of the magnetic sheet for thin-film inductors to which a low-temperature aging-treated nanopowder was applied was significantly improved.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.