Spectral domain-based data-embedding mechanisms for display-to-camera communication

Abstract

Recently, digital displays and cameras have been extensively used as new data transmission and reception devices in conjunction with optical camera communication (OCC) technology. This paper presents three types of frequency-based data-embedding mechanisms for a display-to-camera (D2C) communication system, in which a commercial digital display transmits information and an off-the-shelf smartphone camera receives it. For the spectral embedding, subband coefficients obtained from a discrete cosine transform (DCT) image and predetermined embedding factors of three embedding mechanisms are used. This allows the data to be recovered from several types of noises induced in wireless optical channels, such as analog-to-digital (A/D) and digital-to-analog (D/A) conversion, rotation, scaling, and translation (RST) effects, while also maintaining the image quality to normal human eyes. We performed extensive simulations and real-world D2C experiments using several performance metrics. Through the analysis of the experimental results, it was shown that the proposed method can be considered as a suitable candidate for the D2C system in terms of the achievable data rate (ADR), peak signal-to-noise ratio (PSNR), and the bit error rate (BER).