Fingerprinting of deformed paper images acquired by scanners

Abstract

Images texture extraction is a core step in image pattern recognition applications such as paper texture identification or fingerprinting. Different methods are applied for paper images texture extraction. Subsequently, one of the wellknown methods in images texture extraction is the Locale Binary Pattern (LBP) method. However, the LBP method show a number of drawbacks in paper images texture extraction and two of which are neglecting some texture information of the images and incompetent to some images deformation due to its local view. In this paper, combinations of Gabor filters and a LBP operator are proposed to reduce the effects of the mentioned drawbacks in papers fingerprinting domain. We use self-collected textures from 102 paper images in the test. Consequently, the testing results of the proposed combinations improve paper images identification rate by 28.45% when the Gabor filters have a scale of 9 and an orientation of π/2 degree. This paper finds that applying Gabor filters prior to LBP method improve the LBP description and the papers fingerprinting accuracy.

Comparative analysis of classification techniques for leaves and land cover texture

“…3 [26]. Based on the articles related to k-NN, the probability of error of simple classification rule is bounded with the Bayes minimum probability of error is better that make the most impact paper in pattern recognition and texture classification applications such as the document authentication texture features [27]. The k-NN algorithm is one of the supervised machine learning algorithms that can solve both classification and regression problems.…”

Review of local binary pattern operators in image feature extraction

“…Using benchmark data, the authors have demonstrated that the performance of their proposed method is superior regarding the grayscale variations and noise-resistance. Khaleefah et al [17] have utilized texture classification method by using LBP descriptors for the task of paper texture identification or fingerprinting, which are well-known technique, in the texture classification that shown superior performance in authenticating documents. Wan et al [18] have proposed an enhanced LBP, namely, average-LBP, for the process of texture analysis of human breast tissue images.…”

“…Table 1 summarizes all LBP texture analysis related work that has been illustrated in subsequent paragraphs. [16] Extended LBP Structure Image classification Wan et al [18] Average-LBP Averaging Medical image classification Kim et al [19] Adaptive LBP Structure Image classification for handwritten recognition Dey et al [21] LBP and spatial sampling Structure Image segmentation for handwritten recognition Almezoghy et al [22] PCA-LBP Averaging Image classification for palm recognition Bian et al [23] Multistructure LBP Structure High-resolution image classification Jia et al [24] LBP superpixel-level Structure Image classification for hyperspectral images Yuan et al [25] HDLBP for spatial structure Structure Image classification for material recognition Xu et al [26] PCBP Averaging Face recognition Kou et al [27] PCLBP Structure Image texture classification Khaleefah et al [17,20,28] LBP, ULBP Parameter tuning Paper fingerprinting…”

A survey on local binary pattern and gabor filter as texture descriptors of smart profiling systems

“…Experimental results showed that the Gabor filter had the ability to detect and remove the footprint which has contributed toward better identification of structural features such as the sand and channels. Khaleefah et al [24] have utilized texture classification method by using LBP descriptors for the task of paper texture identification or fingerprinting, which are well-known technique, in the texture classification that shown superior performance in authenticating documents. Tadic et al [25] have proposed a fuzzy Gabor filter for the license plates detection.…”

“…Table 2 summarizes all the Gabor filter related work that has been illustrated in the latter paragraphs. [22] Variable thresholding Gabor algorithm Parameter tuning Medical image classification Ang et al [23] Modified Gabor filter Noisy data Geographical image classification Khaleefah et al [24] Gabor filter Structure Texture document image Tadic et al [25] Fuzzy-based Gabor filter Parameter tuning Pattern recognition Low et al [26] FGFC Parameter tuning Face recognition Dora et al [28] PSO with Gabor filter Parameter tuning Face Recognition Ji et al [29] Gabor with DFT Structure Image classification Babashakoori and Ezoji [30] Gabor and Hough…”