Search Results (2)

China’s grain production has been on a pathway parallel to urbanization in the last 30 years. When the balance between the grain supply and demand is considered, contradiction between farming suitability and actual deviation provides warning of a crisis regarding China’s food security. In this study, we constructed a set of topologic maps to summarize the basic distribution of the farming conditions in China, and Kernel density and Local Moran’s I analyses were con-ducted to investigate the spatial-temporal pattern of China’s regional grain production based on the grain output at the county level from 1985 to 2019. The results show that the high-output zones were concentrated in the regions with superior physical conditions in 1985, and by 2019, the high-output zones had increased in the northern regions (i.e., Northeast China Plain) and decreased in the southern regions (i.e., Southern China). The surplus zones of per capita grain output were concentrated in the regions with high total outputs during 1985–2019. The shortage zones were distributed in the regions with low total outputs and low total outputs or large populations. Based on the above three results, several typical commodity grain bases have lost their dominant role (i.e., the Pearl River Delta); furthermore, the qualified commodity grain bases were compiled at both the national and regional level based on overlay analysis (i.e., the Tarbagatay Prefecture as well as eastern and central Inner Mongolia). Full article

Gesture recognition acts as a key enabler for user-friendly human-computer interfaces (HCI). To bridge the human-computer barrier, numerous efforts have been devoted to designing accurate fine-grained gesture recognition systems. Recent advances in wireless sensing hold promise for a ubiquitous, non-invasive and low-cost system with existing Wi-Fi infrastructures. In this paper, we propose DeepNum, which enables fine-grained finger gesture recognition with only a pair of commercial Wi-Fi devices. The key insight of DeepNum is to incorporate the quintessence of deep learning-based image processing so as to better depict the influence induced by subtle finger movements. In particular, we make multiple efforts to transfer sensitive Channel State Information (CSI) into depth radio images, including antenna selection, gesture segmentation and image construction, followed by noisy image purification using high-dimensional relations. To fulfill the restrictive size requirements of deep learning model, we propose a novel region-selection method to constrain the image size and select qualified regions with dominant color and texture features. Finally, a 7-layer Convolutional Neural Network (CNN) and SoftMax function are adopted to achieve automatic feature extraction and accurate gesture classification. Experimental results demonstrate the excellent performance of DeepNum, which recognizes 10 finger gestures with overall accuracy of 98% in three typical indoor scenarios. Full article