Article Preview
Top1. Introduction
These days, almost all products in the market are utilizing a barcode technology. Barcode scanning with dedicated scanners is an established technology. Recently, the ease of use of cell phones with digital camera facility delivers a handy way for decoding barcode without make use of the traditional laser scanner which has poor portability. Camera phones can get an image of the barcode and later, it can transfer decoded information to a consumer product server to get product details (Gallo & Manduchi, 2011). The application of a camera phone in this area is thrilling if the captured image suffers with some type of degradation. Image blurring is often an issue that affects the performance of a bar code identification system. The out-of-focus (defocus) blur is appeared because of the inaccurate focal length adjustment. Image restoration methods (Tiwari, 2017; Tiwari, Shukla, Biradar, & Singh, 2014) available in the literature can be classified as blind deconvolution, where the blur kernel is not known and non-blind deconvolution, where the blur kernel is known. The first and foremost step in any blind image restoration technique is blur estimation. Numerous methods have been presented during the last decades, which attempt to estimate Point Spread Function (PSF) of blur concurrently with the image deconvolution (Kundur & Hatzinakos, 1996; Cannon, 1976). However, a number of efficient methods have suggested that blind deconvolution can be handled better with separate PSF estimation and after that non-blind deconvolution can be used as the consequent step (Gennery, 1973; Hummel, K. Zucker, and S. Zucker, 1987; Lane and R. Bates, 1987; Tekalp, Kaufman, & Wood, 1986). The work presented in this paper falls in the former category where PSF parameters are estimated before image deconvolution.