Inverse of Low Resolution Line Halftone Images for Document Inspection

Abstract

In this paper, a new inverse half toning method has been proposed for reconstructing low resolution line halftone images. This reconstruction is done in order to authenticate an image in question. The reconstructed image is compared with its original image in terms of standard image quality metrics such as peak signal to noise ratio (PSNR) and structural similarity index measure (SSIM). The existing inverse halftone methods have rarely considered line halftone images which are normally of low resolution and the quality of the inverse halftone largely depends on the characteristics like frequency or shape of halftone dots. Our proposed inverse halftone technique consists of two parts: at first, the resolution (in lines per inch, lpi) of an input image is estimated and a low level image from the binary line halftone image is constructed. In the second phase, gray level continuous image is generated from the low level description and the lpi information. The method is based on learning based pattern classification techniques namely, neural nets. A comparative study shows that the proposed method outperforms many existing inverse halftone techniques while dealing with line halftone images.