A Blind Restoration Approach for Defocused Barcode Images

A Blind Restoration Approach for Defocused Barcode Images

Shamik Tiwari
DOI: 10.4018/IJSITA.2017070103
(Individual Articles)
No Current Special Offers


Use of a mobile camera for barcode decoding provides high portability and availability but it requires that the recorded barcode image must be accurate representation of the barcode that is available on the product. Barcode scanning is challenging because images may be degraded due to out-of-focus blur at the time of image acquisition. Therefore, image restoration is essential in making image sharp and useful. In case of blind restoration of such barcode images accurate estimation of out-of-focus blur parameter is highly desirable. In this article, a robust method has been proposed for estimating the radius of out-of-focus blur. Finite discrete ridgelet transform has been used to find the features of the blurred image and a radial basis function neural network is utilized to estimate the radius of out-of-focus blur. The experimental results reveal that proposed method more robust than the existing methods.
Article Preview

1. 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.

Complete Article List

Search this Journal:
Open Access Articles: Forthcoming
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing