Low intensity visual data improvement using DWT and LWT based regularizer

Image Completion plays a vital role in compressed sensing, machine learning, and computer vision applications. The Rank Minimization algorithms are used to perform the image completion. The major problem with rank minimization algorithms is the loss of information in the recovered image at high corruption ratios. To overcome this problem Lifting wavelet transform based Rank Minimization (LwRM), and Discrete wavelet transform based Rank Minimization (DwRM) methods are proposed, which can recover the image, if the corrupted observations are more than 80%. The evaluation of the proposed methods are accomplished by Full Reference Image Quality Assessment (FRIQA) and No Reference Image Quality Assessment (NR-IQA) metrics. The simulation results of proposed methods are superior to state-of-the-art methods.

     

Target based image fusion using multi-scale feature selection in three regions

Image captured by low dynamic range (LDR) camera fails to capture entire exposure level of scene, and instead only covers certain range of exposures. In order to cover entire exposure level in single image, bracketed exposure LDR images are combined. The range of exposures in different images results in information loss in certain regions. These regions need to be addressed and based on this motive a novel methodology of layer based fusion is proposed to generate high dynamic range image. High and low-frequency layers are formed by dividing each image based on pixel intensity variations. The regions are identified based on information loss section created in differently exposed images. High-frequency layers are combined using region based fusion with Dense SIFT which is used as activity level testing measure. Low-frequency layers are combined using weighted sum. Finally combined high and low-frequency layers are merged together on pixel to pixel basis to synthesize fused image. Objective analysis is performed to compare the quality of proposed method with state-of-the-art. The measures indicate superiority of the proposed method.

     

Compression based clustering technique for enhancing accuracy in web scale videos

Multimedia Tools and Applications - Detection and clustering of commercial advertisements plays an important role in multimedia indexing also in the creation of personalized user content. In...     

Entropy based spatial domain image watermarking and its performance analysis

Digital image watermarking technique based on LSB Substitution and Hill Cipher is presented and examined in this paper. For better imperceptibility watermark is inserted in the spatial domain. Further the watermark is implanted in the Cover Image block having the highest entropy value. To improve the security of the watermark hill cipher encryption is used. Both subjective and objective image quality assessment technique has been used to evaluate the imperceptibility of the proposed scheme.Further, the perceptual perfection of the watermarked pictures accomplished in the proposed framework has been contrasted and some state-of-art watermarking strategies. Test results demonstrates that the displayed method is robust against different image processing attacks like Salt and Peppers, Gaussian filter attack, Median filter attacks, etc.

     

A novel approach to design optimal 2-D digital differentiator using vortex search optimization algorithm

Multimedia Tools and Applications - The designing of 2-D digital differentiator is multimodal and high dimensional problem which requires large number of differentiator coefficients to be...     

Quote examiner: verifying quoted images using web-based text similarity

Over the last few years, there has been a rapid growth in digital data. Images with quotes are spreading virally through online social media platforms. Misquotes found online often spread like a forest fire through social media, which highlights the lack of responsibility of the web users when circulating poorly cited quotes. Thus, it is important to authenticate the content contained in the images being circulated online. So, there is a need to retrieve the information within such textual images to verify quotes before its usage in order to differentiate a fake or misquote from an authentic one. Optical Character Recognition (OCR) is used in this paper, for converting textual images into readable text format, but none of the OCR tools are perfect in extracting information from the images accurately. In this paper, a method of post-processing on the retrieved text to improve the accuracy of the detected text from images has been proposed. Google Cloud Vision has been used for recognizing text from images. It has also been observed that using post-processing on the extracted text improved the accuracy of text recognition by 3.5% approximately. A web-based text similarity approach (URLs and domain name) has been used to examine the authenticity of the content of the quoted images. Approximately, 96.26% accuracy has been achieved in classifying quoted images as verified or misquoted. Also, a ground truth dataset of authentic site names has been created. In this research, images with quotes by famous celebrities and global leaders have been used. A comparative analysis has been performed to show the effectiveness of our proposed algorithm.

     

Forensic image analysis using inconsistent noise pattern

With the advancement of image acquisition devices and social networking services, a huge volume of image data is generated. Using different image and video processing applications, these image data are manipulated, and thus, original images get tampered. These tampered images are the prime source of spreading fake news, defaming the personalities and in some cases (when used as evidence) misleading the law bodies. Hence before relying totally on the image data, the authenticity of the image must be verified. Works of the literature are reported for the verification of the authenticity of an image based on noise inconsistency. However, these works suffer from limitations of confusion between edges and noise, post-processing operation for localization and need of prior knowledge about an image. To handle these limitations, a noise inconsistency-based technique has been presented here to detect and localize a false region in an image. This work consists of three major steps of pre-processing, noise estimation and post-processing. For the experimental purpose two, publicly available datasets are used. The result is discussed in terms of precision, recall, accuracy and f1-score on the pixel level. The result of the presented work is also compared with the recent state-of-the-art techniques. The average accuracy of the proposed work on datasets is 91.70%, which is highest among state-of-the-art techniques.

     

HSIC-based affinity measure for learning on graphs

Pattern Analysis and Applications - The performance of graph-based learning techniques largely relies on the edges defined between the vertices of the graph. These edges that represent the affinity...     

Model reduction based on limited‐time interval impulse response Gramians

In this paper, three new Gramians are introduced namely ‐ limited‐time interval impulse response Gramians (LTIRG), generalized limited‐time Gramians (GLTG) and generalized limited‐time impulse response Gramians (GLTIRG). GLTG and GLTIRG are applicable to both unstable systems and also to systems which have eigenvalues of opposite polarities and equal magnitude. The concept of these Gramians is utilized to develop model reduction algorithms for linear time‐invariant continuous‐time single‐input single‐output (SISO) systems. In the cases of GLTIRG and GLTG based model reduction, the standard time‐limited Gramians are generalized to be applied to unstable systems by transforming the original system into a new system which requires the solution of two Riccati equations. Two numerical examples are included to illustrate the proposed methods. The results are also compared with standard techniques.     

A smart application to detect pupil for small dataset with low illumination

Biometric applications are very sensitive to the process because of its complexity in presenting unstructured input to the processing. The existing applications of image processing are based on the implementation of different programing segments such as image acquisition, segmentation, extraction, and final output. The proposed model is designed with 2 convolution layers and 3 dense layers. We examined the module with 5 datasets including 3 benchmark datasets, namely CASIA, UBIRIS, MMU, random dataset, and the live video. We calculated the FPR, FNR, Precision, Recall, and accuracy of each dataset. The calculated accuracy of CASIA using the proposed system is 82.8%, for UBIRIS is 86%, MMU is 84%, and the random dataset is 84%. On live video with low resolution, calculated accuracy is 72.4%. The proposed system achieved better accuracy compared to existing state-of-the-art systems.