Blind watermarking targets the challenging recovery of the watermark when the host is not available during the detection stage. This paper proposes Discrete Shearlet Transform (DST) as a new embedding domain for blind image watermarking. Our novel DST blind watermark detection system uses a nonadditive scheme based on the statistical decision theory. It first computes the Probability Density Function (PDF) of the DST coefficients modeled as a Laplacian distribution. The resulting likelihood ratio is compared with a decision threshold calculated using Neyman–Pearson criterion to minimize the missed detection subject to a fixed false alarm probability. Our method is evaluated in terms of imperceptibility, robustness, and payload against different attacks (Gaussian noise, blurring, cropping, compression, and rotation) using 30 standard grayscale images covering different characteristics (smooth, more complex with a lot of edges, and high detail textured regions). The proposed method shows greater windowing flexibility with more sensitive to directional and anisotropic features when compared against discrete wavelet and contourlets.

/pdf/blind-image-watermark-detection-algorithm-based-on-discrete-h0k29mwacz.pdf

Blind Image Watermark Detection Algorithm Based on Discrete Shearlet Transform Using Statistical Decision Theory

This paper proposes a blind digital audio water- marking algorithm that utilizes the quantization index modulation (QIM) and the singular value decomposition (SVD) of stereo audio signals. Conventional SVD-based blind audio watermarking algorithms lack physical interpretation since the matrix construction method for the input matrix for SVD is heuristically defined. However, in the proposed approach, because the SVD is directly applied to the stereo input signals, the resulting decomposed elements convey a conceptually meaningful inter- pretation of the original audio signal. As the proposed approach effectively utilizes the ratio of singular values, the embedded watermark is highly imperceptible and robust against volumetric scaling attacks; most QIM-based watermarking schemes are weak to these types of attacks. Experimental results under well-known practical attacks, such as compressions, resampling, and various types of signal processing, confirm that the proposed algorithm performs well compared to conventional audio watermarking algorithms.

SVD-Based Adaptive QIM Watermarking on Stereo Audio Signals

Methods, systems, devices and computer program products are provided to facilitate enhanced use and interaction with multimedia content that is based on a time-varying evaluation of the multimedia content. While a content is being presented on a first device, timing information identifying temporal locations of content segments are extracted and transmitted to a database along with a user's evaluation feedback of the multimedia content. The user then receives a response corresponding to feedback provided by a plurality of consumers of the multimedia content on a segment-by-segment basis.

Time varying evaluation of multimedia content

An optimized blind watermarking scheme based on principal component analysis in redundant discrete wavelet domain

Content screening operations are facilitated in devices that receive a content that is subject to screening obligations. When such a content is received at a device, a watermark extraction record is obtained and accessed to fulfill content screening obligations. Upon the receipt of such an extraction record, verification of the received extraction record is carried out based on a verification rate. If the verification is successful for an extraction record with permissive information, the verification rate is decreased, thereby reducing the processing load of the device. If the verification is unsuccessful, the verification rate is increased, which can adversely affect the processing load of the device.

Enhanced content management based on watermark extraction records

The present application provides a robust system and method for detecting the watermark in an electronic media, wherein the electronic media had gone through various kinds of attacks and their combinations thereof which may not be known while detecting the watermarks. The watermark detection system and method can be trained to detect or reject a particular pattern. The watermark detection system and method are based on correlation and are useful in any kind of digital watermarking applications. The watermark detection system and method perform well even when the amount of distortion is not precisely known, wherein the disclosure provides a set of templates or correlation filters, being designed for detection of watermark to cover any kind and combination of attacks. For synchronous attacks, the correlation filter designing is carried out dynamically. Particularly, the template i.e. correlation filter is an attack adaptive frequency domain pattern for watermark detection.

System and method for detecting the watermark using decision fusion

Rapidly growing use of biometrics across enterprises, consumer electronics, and applications has made biometric template protection indispensable to any practical biometric system. To provide biometric template protection, several methods have been reported but most of them have a trade-off between matching performance and the template security. To address this problem, we improvise upon existing methods to provide a method for face template protection with improved matching performance and high template security. The proposed method uses deep Convolutional Neural Network (CNN) together with random projection to maximize the inter-user variations, reduce the dimensionality (thus eliminating redundancy) of the extracted feature vector of each face image and minimize the intra-user variations. The proposed method is robust enough to perform even with one-shot enrollment of users. Three publicly available face datasets, namely, CMU-PIE, FEI and Color-FERET are used for evaluation. The proposed method achieves ~99.5% Genuine Accept Rate (GAR) at zero False Accept Rate (FAR), improving the matching performance by ~10% over the comparable state-of-the-art methods, while providing high template security.

Securing Face Templates using Deep Convolutional Neural Network and Random Projection

An optical character recognition (OCR) system recognizes either printed or handwritten text. Hence it is required to seperate machine printed text from handwritten text in scanned documents before feeding it to a OCR system. We can discriminate these two types of text word images by their visual impression and shape structures. The intensity values distribution features gives us the visual impression and the shapes can be represented by the structural features. This paper proposes an approach for machine print and handwritten text classification at word level using intensity and shape structural features of scanned text. The proposed method achieved impressive classification efficiency on IAM dataset.

A System for Handwritten and Printed Text Classification

Visible light face recognition has been well researched but still its performance is limited by varying illumination conditions. Illumination conditions are major source of the uncertainty in face recognition systems performance when it is used in outdoor setting. In order to augment the performance of visible light face recognition system's performance, infrared facial images as a new modality has been used in the literature. The other intriguing factor in using the infrared, especially thermal infrared imaging for facial recognition is the night time surveillance with little or no light to illuminate the faces. Thermal facial recognition is used in several covert military applications. The covert data acquisition has its own environmental constraints. The choice of infrared makes the system less dependent on external light sources and more robust with respect to incident angle and light variations. In this paper we focus on reviewing recent developments in thermal IR face recognition and suggest the approaches that could be useful in matching visible and thermal IR images.

Thermal Infrared Face Recognition: A Review

Digital watermarking is a process to provide authenticity by hiding a data into an image or audio or document. Hiding of data in an image can be done in frequency domain. Since frequency domain based techniques are more robust against signal processing and geometric attacks than time domain based techniques and watermark can also be extracted with(non blind) and without(blind) original cover image. In this paper we proposed a blind image watermarking technique which embeds watermark into image in frequency domain using discrete wavelet transform, singular value decomposition and torus automorphism techniques. This technique extracts watermark without cover image and also proved that this method is robust against different signal and non signal processing attacks.

Srinivasa Rao Chalamala

Papers

System and method for detecting the watermark using decision fusion

Securing Face Templates using Deep Convolutional Neural Network and Random Projection

A System for Handwritten and Printed Text Classification

Thermal Infrared Face Recognition: A Review

Block based robust blind image watermarking using discrete wavelet transform