Free hand interaction with devices is a promising trend with the advent of Internet of Things (IoT). The unmodulated ambient light, which can be an exciting modality for interaction, is still deficient in research and practice when most of the efforts in the field of visible light sensing are put into solutions based on modulated light. In this paper, we propose a low-cost ambient light-based system which performs finger gesture recognition in real-time. The system relies on a recurrent neural network (RNN) architecture without complicated pre-processing algorithms for the gesture classification task. The results of experimental evaluation proves that the solution that we put forward achieves a rather high recognition accuracy with our proposed sensor layout across a certain group of users.

As an essential requirement of pervasive smart devices, free hand gestural input considered as necessary for user interactions has attracted lots of research attention for nearly decades. Nevertheless, existing proposals heavily rely on either expensive pre-deployed equipment or user on-body sensors, thus confine their application scenarios. In this paper, we propose a novel hand gesture recognition system which purely relies on ubiquitous ambient light and low-cost photodiodes. The proposed system does not need any modification to existing lighting infrastructure. While without complex signal pre-processing for modulated light, very low-cost photodiodes and processors can capture and process the light variations caused by hand gesture. To produce accurate hand gesture recognition, we design efficient algorithms based on recurrent neural network to process sensing data collected by a photodiode array. We implement a prototype consisting of an array of 8 photodiodes and extensive experiments demonstrate that the proposed solution can achieve a very high overall recognition accuracy of 99.31%.

A component is the basic unit of Tibetan characters, and component decomposition of Tibetan characters is a fundamental step for Tibetan informatization. According to the character formation rules and writing features in the vertical and horizontal dimensions of Tibetan characters, a matrix-based algorithm for the component decomposition of Tibetan characters is proposed. Firstly, the algorithm decomposes Tibetan characters in vertical and horizontal dimensions, based on their structural features and writing order. Secondly, it goes to decomposition each dimension respectively. Finally, 48 different structures of modern Tibetan characters were tested. The test results show that the accuracy of the algorithm reaches 100%.