Banks have been already offering a wide variety of products and services, integrated with technology automation, the most familiar being ATM machines all around us.Now moving to next level in present Industry 4.0 era, Banking sector is set amplify its strategy implementation by leveraging latest digital technologies so that customers may experience swift secure processing transactions.This study examined evolution, adoption, future opportunities Artificial Intelligence (AI) for successful...

A semi-blind watermarking scheme based on singular-value-decomposition (SVD) and homomorphic transform is pro-posed. This ensures the digital security of an eight bit gray scale image by inserting invisible wa-termark into it. The key approach to apply host obtain its reflectance component. watermark embedded singular values that are obtained applying value decomposition Peak-signal-to-noise-ratio (PSNR), normalized-correlation-coefficient (NCC) mean-structural-similarity-index-measure...

This paper attempts to solve the integrity issues of a compromised face biometric system using two watermarking schemes. Two new blind schemes, namely S <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> and xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> , are proposed ensure training database test images, respectively. Scheme is fragile spatial-domain based scheme works in discrete cosine transformation (DCT) domain robust channel...

This paper presents the Roman Characters Recognition in clean and noisy environment using geometrical regional features. The features are extracted from standard dataset combined to achieve better recognition. combination of these is classified neural network (NN) random forest (RF) classifiers. In our experiments, we have achieved recognition accuracy 100% for some characters. However, average 85.7% has been recorded by NN 88% RF classifier, respectively.

In this paper, we present an efficient and robust technique for the recognition of offline roman characters.The main strategy is to extract statistical similarity features using a combination grey level co-occurrence matrix (GLCM) complementary measure (CSM) method.In work, CSM method used from binary images combined with GLCM boost accuracy character recognition.The has been done four different classifiers i.e. artificial neural network (ANN), Naive Bayes classifier, random forest (RF)...

Images are used to analyse the solution of two dimensional heat equation. A remarkable observation is that image detected an ambiguity in existing solution. Therefore better proposed. The images formed at three different sampling rates 1000 × 1000, 100 and 10 pixels per unit for space variables. Peak Signal Noise Ratio (PSNR) as quantitative measure relative analysis shown users It has been observed both by using rate rational with mathematical Average computation time (in seconds) form a...

The solution of the two-dimensional heat equation is represented as images. considered 2 . U t x y           , 0  1, 1 Boundary conditions are U(0, y,t) 0, U(1, U(x,0,t) and U(x,1,t) initial condition U(x, y,0) sin4 xcos4 y images provide quick but approximate insights to solution. Approximation by image visualization depends on many factors such display device, human eye factors, resolution (sampling rate bit depth image), resize methods. We have provided for different...

This paper presents the 3D motion trajectories (lower case alphabetic characters) recognition using optimal set of geometric primitives, angular and statistical features. It has been observed that different combinations these features have not used in literature for characters. The standard dataset named “CHAR3D” analysis purpose. consists 2858 character samples each sample is 3 dimensional pen tip velocity trajectory. In this only single down segmented characters considered. performed...

This present paper represents a simple framework for the novel study of effect finite word–length (FWL) on performance different watermarking methods in computer environment. We have evaluated discrete wavelet transform (DWT) and singular value decomposition (SVD)–based at word–lengths found reliable (the up to which change is significant beyond this, unnoticeable) each method. The can be used as new benchmark comparing their real–time implementation, where optimisation with respect memory,...

Abstract Air‐writing recognition is relevant in areas such as natural human–computer interaction, augmented reality, and virtual reality. A trajectory the most way to represent air writing. We analyze accuracy of words written considering five features, namely, writing direction, curvature, trajectory, orthocenter, ellipsoid, well different parameters a hidden Markov model classifier. Experiments were performed on two representative datasets, whose sample trajectories collected using Leap...

Quantization parameter based blind watermarking schemes are very popular. These work in spatial domain as well various transform domains such wavelet, discrete cosine, and real oriented wavelet. utilize one bit (binary) image a watermark. However, general, watermarks any n image, where ≥ 1. Therefore, these not suitable for the with > In this paper, we extend existing on bits watermark images. The 'division algorithm' is used key component to develop extended schemes. proposed scheme...

The performance of a watermarking system depends on all its factors, such as watermark embedding strength, size and repetition bits. In this paper, mathematical formula is designed, that governs the relation with amount Performance measured by using extracted bit accuracy rate. designed verified experimentally. A spatial domain scheme used for experimental verification. Further, verification done different kind attacks cropping, Gaussian filter, noise salt &amp; pepper noise. Accuracy rate...