International Journal of Image and Graphics

Identification of motion in videos is a fundamental task for several computer vision problems. One of the main tools for motion identification is optical flow, which estimates the projection of the 3D velocity of the objects onto the plane of the camera. In this work, we propose a differential optical flow method based on the wave equation. The optical flow is computed by minimizing a functional energy composed by two terms: a data term based on brightness constancy and a regularization term based on energy of the wave. Flow is determined by solving a system of linear equations. The decoupling of the pixels in the solution allows solving the system by a direct or iterative approach and makes the method suitable for parallelization. We present the convergence conditions for our method since it does not converge for all the image points. For comparison purposes, we create a global video descriptor based on histograms of optical flow for the problem of action recognition. Despite its sparsity, results show that our method improves the average motion estimation, compared with classical methods. We also evaluate optical flow error measures in image sequences of a classical dataset for method comparison.

Moving object detection and tracking have various applications, including surveillance, anomaly detection, vehicle navigation, etc. The literature on object detection and tracking is rich enough, and there exist several essential survey papers. However, the research on camouflage object detection and tracking is limited due to the complexity of the problem. Existing work on this problem has been done based on either biological characteristics of the camouflaged objects or computer vision techniques. In this paper, we review the existing camouflaged object detection and tracking techniques using computer vision algorithms from the theoretical point of view. This paper also addresses several issues of interest as well as future research direction in this area. We hope this paper will help the reader to learn the recent advances in camouflaged object detection and tracking.

In this paper, we describe our work in Spoken language Identification using Visual Speech Recognition (VSR) and analyze the effect of various visual speech units used to transcribe the visual speech on language recognition. We have proposed a new approach of word recognition followed by the word N-gram language model (WRWLM), which uses high-level syntactic features and the word bigram language model for language discrimination. Also, as opposed to the traditional visemic approach, we propose a holistic approach of using the signature of a whole word, referred to as a “Visual Word” as visual speech unit for transcribing visual speech. The result shows Word Recognition Rate (WRR) of 88% and Language Recognition Rate (LRR) of 94% in speaker dependent cases and 58% WRR and 77% LRR in speaker independent cases for English and Marathi digit classification task. The proposed approach is also evaluated for continuous speech input. The result shows that the Spoken Language Identification rate of 50% is possible even though the WRR using Visual Speech Recognition is below 10%, using only 1s of speech. Also, there is an improvement of about 5% in language discrimination as compared to traditional visemic approaches.

The medical field has been revolutionized by the medical imaging system, which plays a key role in providing information on the early life-saving detection of dreadful diseases. Diabetic retinopathy is a chronic visual disease that is the primary reason for the vision loss in most of the patients, who left undiagnosed at the initial stage. As the count of the diabetic retinopathy affected people kept on increasing, there is a necessity to have an automated detection method. The accuracy of the diagnosis of the automatic detection model is related to image acquisition as well as image interpretation. In contrast to this, the analysis of medical images by using computerized models is still a limited task. Thus, different kinds of detection methods are being developed for early detection of diabetic retinopathy. Accordingly, this paper focuses on the various literature analyses on different detection algorithms and techniques for diagnosing diabetic retinopathy. Here, it reviews several research papers and exhibits the significance of each detection method. This review deals with the analysis on the segmentation as well as classification algorithms that are included in each of the researches. Besides, the adopted environment, database collection and the tool for each of the research are portrayed. It provides the details of the performance analysis of the various diabetic detection models and reveals the best value in the case of each performance measure. Finally, it widens the research issues that can be accomplished by future researchers in the detection of diabetic retinopathy.

A non-extra pixel interpolation NPI is introduced for efficient image upscaling purposes. The NPI algorithm uses extended-triangular and linear scaling functions to match the pixel coordinates. The triangular function uses a modulo-operator with only two variables representing image pixels and scaling ratio. Every two variables of the linear scaling function represent the source/destination image pixels and scaling ratio. The traditional ceil function is used to round off non-integer pixel coordinates. The circshift and padarray functions are used to circularly shift the elements in array output by $k$-amount in each dimension and pad elements of the $d$th columns/rows by g-padsize in the shifted array, respectively. The $k$, $d$ and $g$ values are determined with respect to integer scaling ratios by a vector of $n$-elements. The Exactness, Peak Signal-to-Noise Ratio, Signal-to-Noise Ratio and Discrete Fourier Transform techniques were used for objective evaluation purposes. Experiments demonstrated comparable results as well as the need for further researches.

Mammography imaging is one of the most widely used techniques for breast cancer screening and analysis of abnormalities. However, due to some technical difficulties during the time of acquisition and digital storage of mammogram images, impulse noise may be present. Therefore, detection and removals of impulse noise from the mammogram images are very essential for early detection and further diagnosis of breast cancer. In this paper, a novel adaptive trimmed median filter (ATMF) is proposed for impulse noise (salt & pepper (SNP)) detection and removal with an application to mammogram image denoising. Automatic switching mechanism for updating the Window of Interest (WoI) size from ($3\times 3$) to ($5\times 5$) or ($7\times 7$) is performed. The proposed method is applied on publicly available mammogram images corrupted with varying SNP noise densities in the range 5%–90%. The performance of the proposed method is measured by various quantitative indices like peak signal to noise ratio (PSNR), mean square error (MSE), image enhancement factor (IEF) and structural similarity index measure (SSIM). The comparative analysis of the proposed method is done with respect to other state-of-the-art noise removal methods viz., standard median filter (SMF), decision based median filter (DMF), decision based unsymmetric trimmed median filter (DUTMF), modified decision based unsymmetric trimmed median filter (MDUTMF) and decision based unsymmetric trimmed winsorized mean filter (DUTWMF). The superiority of the proposed method over other compared methods is well evident from the experimental results in terms of the quantitative indices (viz., PSNR, IEF and SSIM) and also from the visual quality of the denoised images. Paired t-test confirms the statistical significance of the higher PSNR values achieved by the proposed method as compared to the other counterpart techniques. The proposed method turned out to be very effective in denoising both high and low density noises present in (mammogram) images.

Image creation and retention are growing at an exponential rate. Individuals produce more images today than ever in history and often these images contain family. In this paper, we develop a framework to detect or identify family in a face image dataset. The ability to identify family in a dataset of images could have a critical impact on finding lost and vulnerable children, identifying terror suspects, social media interactions, and other practical applications. We evaluated our framework by performing experiments on two facial image datasets, the Y-Face and KinFaceW, comprising 37 and 920 images, respectively. We tested two feature extraction techniques, namely PCA and HOG, and three machine learning algorithms, namely K-Means, agglomerative hierarchical clustering, and K nearest neighbors. We achieved promising results with a maximum detection rate of 94.59% using K-Means, 89.18% with agglomerative clustering, and 77.42% using K-nearest neighbors.

The upcoming generation is at high risk of developing many health issues like heart diseases, metabolic diseases and other life-threatening problems with high mortality as a consequence of obesity due to intake of unhealthy food which is totally deviated from a normal balanced diet with appropriate calories, proteins, vitamins and carbohydrates. In this work, the nutrient intake is calculated using food image. Our system provides efficient segmentation algorithms for separating food items from the plate. The given 2D image of food is converted into 3D image by generating its depth map for volume generation and color, texture and shape features are extracted. These features are fed as input into multi-class support vector machine classifier for learning. The learning phase involves training of various mixed and non mixed food items. The testing phase includes query image segmentation and classification for identifying the type of food and then finding calories using the nutrition data table. We have also estimated the ingredient and decay of food items. Our result shows accurate calorie estimation for various kinds of food items.

The content of massive image changing the brightest brightness is an impasse between most tests of sorted image realizations with low-resolution representation. I have done this research through image security, which will help curb crime in the coming days, and we propose a novel receipt for their strong and effective counterpart. Image classification using low levels of the image is a difficult method, so for this, I have adopted the method of automating the semantic image classification of this research and used it with different SVM classifiers, based on the normalized weighted feature support vector machine for semantic image classification. This is a novel approach given that weighted feature or normalized biased feature is applied and it is found that the normalized method is the best. It also uses normalized weighted features to compute kernel functions and train SVM. The trained SVM is then used to classify new images. During training and generalization, we displayed a decrease of identification error rate and there have been many benefits of using SVM with better performance in normalized image-cataloging systems. The importance of this technique and its role will be highlighted in the years to come.

The repolarization and depolarization in heart generate electrical signals in the form of an ECG wave. The condition of the heart can be indicated by using Heart Rate Variability (HRV) features. In this work, FIR filter is used at the pre-processing phase for denoising, and then statistical analysis is applied for time-domain HRV feature extraction and selection. This algorithm is evaluated on different records of MIT/BIH Normal Sinus Rhythm and Arrhythmia database. The $t$-test implementation in both databases shows that there are significant variations in HRV features, where meanRR and HR have suggestive significant ($0.05<p\le 0.10$) changes, while maxRR, minRR, maxminRR, and SDNN have strongly significant ($p\le 0.01$) changes. To validate the statistical analysis of HRV, feature classification has been done using SVM and kNN classifiers. A significant improvement of 2% and 14.02% has been observed in the overall accuracy of SVM and kNN classifiers after feature selection, respectively. These HRV features can be used for the early prediction of various Cardio-Vascular Diseases (CVD).