Semi-Fragile Zernike Moment-Based Image Watermarking for Authentication

Short-term load forecasting (STLF) is one of the main challenging problems in management of power systems. By emerging the new technologies and development of smart grids, STLF becomes a necessity. A STLF framework is proposed in this work which uses the long short-term memory (LSTM) network for forecasting. In the proposed framework, a feature vector containing the lagged load values is considered for each point of the load time series where this feature vector can be given to the LSTM network in three ways: 1) the original feature vector (LSTM), 2) the coefficients of discrete wavelet transform of the feature vector (DWT-LSTM), and 3) the stacked vector of wavelet coefficients and original load values (DWT-stack-LSTM). In the proposed DWT-stack-LSTM method, both of original values of load and decomposition values of load in different frequencies and scales are given to LSTM. Thus, LSTM is trained to learn more about load variables and so, provides more accurate forecasting results.

Abstract. The aim of our research was to examine whether simulated forest data can be utilized for training supervised classifiers. We included two classifiers namely the random forest classifier and the novel convolutional neural network classifier that utilizes feature images. We simulated tree parameters and created a feature vector for each tree. The original feature vector was utilised with random forest classifier. However, these feature vectors were also converted into feature images suitable for input into a YOLO (You Only Look Once) convolutional neural network classifier. The selected features were red colour, green colour, near-infrared colour, tree height divided by canopy diameter, and NDVI. The random forest classifier and convolutional neural network classifier performed similarly both with simulated data and field-measured reference data. As a result, both methods were able to identify correctly 97.5 % of the field-measured reference trees. Simulated data allows much larger training data than what could be feasible from field measurements.

This paper presents a technique that used to enhance the image resolution. The input low resolution (LR) image is decomposed into low frequency (LF) and high frequency (HF) sub-bands by using lifting wavelet transform (LWT) and discrete wavelet transform (DWT). The resolution enhancement is done by interpolating HF sub-bands that are generated using LWT from the LR image. The HF sub-bands of DWT are used to preserve the edges of the images. The operation is done on HF sub-bands of LWT and DWT. The HF sub-bands that are generated from the DWT are interpolated and perform addition to the respective interpolated HF sub-bands of LWT. The HF sub-bands that are modified and the LR image will again interpolate and undergoes the inverse lifting wavelet transform (ILWT) to reconstruct the high resolution (HR) image. This implemented method results efficient over conventional methods.

Color image encryption using MSVD, DWT and Arnold transform in fractional Fourier domain

Cardiac arrhythmia basically refers to abnormal activity of heart. Correct classification of cardiac arrhythmia is therefore crucial for appropriate treatment of heart diseases. In this paper, a novel approach is proposed for cardiac arrhythmia classification. Initially, the feature vectors extracted from raw electrocardiogram (ECG) signals are projected into a particular subspace obtained via the Common Vector Approach, which is an effective subspace method. The projected vectors are then fed into two distinct decision-tree-based classifiers—namely, C4.5 and random forest. The results obtained from the proposed approach are compared with those obtained with the original feature vectors using the same classifiers. For this purpose, the well-known MIT-BIH arrhythmia database was utilized. Six different sets of features based on QRS, time-domain, wavelet transform and power spectral density are derived from ECG signals in this database. The feature sets are then used in the classification of five main beat types including non-ectopic, ventricular ectopic, supraventricular ectopic, fusion and unknown. The experimental results reveal that the recognition performances achieved by most of the projected features are explicitly higher than those obtained with the original ones. In addition, the classification accuracy of the proposed approach climbs to 100% for the test set.

While manifold learning algorithms can discover intrinsic low-dimensional manifold embedded in the high-dimensional Euclidean space, the discriminant ability of the low-dimensional subspaces obtained by the algorithms is often lower than those obtained by the conventional dimensionality reduction approaches. Furthermore, the original feature vectors may include redundancy such as high-order correlation which cannot be removed by manifold learning algorithms. To address the two problems, we first employ Gabor wavelet to remove intrinsic redundancies of images and obtain a set of over-completed feature vectors. Then a supervised manifold learning algorithm (ULLELDA) is applied to project Gabor-based data and out-of-the-samples into a common low-dimensional subspace. Experiments in two FERET face databases indicate that Gabors indeed help supervised manifold learning to remarkably improve the discriminant ability of low-dimensional subspaces.KeywordsFace RecognitionLinear Discriminant AnalysisDiscriminant AbilityFace DatabaseGabor WaveletThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Power utility providers and power industry service providers face a significant challenge in identifying the type of Power Quality Disturbances (PQD) automatically. This paper discusses a method to classify PQD using signal decomposition, statistical analysis and machine learning. Firstly, Discrete Wavelet Transform (DWT) is applied on the generated PQD signals to decompose the signal to obtain its representation in time and frequency domain. Secondly, first and second order statistical parameters are computed on the selected sub-band of DWT. These parameters are used as features vector for the machine learning based classifier. Our database consists of 2400 generated signals of PQD, which were divided into train and test set. Another set of noise corrupted signal database was generated to evaluate the capability of the system. SVM using quadratic kernel was selected as the classifier of the Power Quality Disturbances feature vector. Comparisons were also made with other types of classifiers and other types of mother wavelet filter functions. The results show that the combination of DWT and SVM managed to classify Power Quality Disturbances with high accuracy and has a strong resistance towards noise.

In his paper a method of seizure detection has been proposed based on the Discrete Wavelet Transform (DWT) analysis of the dominant Intrinsic mode function(IMF) resulting from the Empirical Mode Decomposition(EMD) of the EEG signals. Considering the normalized energy, Fourier spectrum and cross-correlation coefficient analysis, only the 4th Level DWT coefficients of the dominant IMF is found reasonable for feature computation. In order to reduce the dimension of the feature vector, Higher order statistics of these coefficients are employed to form he feature vector. The reduced feature vector thus formed is found effective for distinguishing seizure and non-seizure EEG signals when fed to a k-nearest neighborhood (k-NN) classifier. Extensive simulations are carried out using a benchmark EEG dataset. It is shown that the proposed method is capable of producing greater sensitivity, specificity and accuracy in comparison to that obtained by a sate-of-the-art method using the same EEG dataset and classifier.

In order to analyze medical images zero watermarks, the contribution of this research used ResNet101-DCT (Discrete Cosine Transform) in combination with the discrete wavelet transform (DWT). A pre-training network called ResNet101 is also used for the extraction of deep medical image features the vector of feature set is constructed a perceptual hash (*) function after the feature rearrange utilizing (DWT) and (DCT). It is necessary to create and maintain a key vector, randomly generate the watermark, and include an X-OR gate in order to encrypt the medical image's watermark. The feature attributes approach is used to find the relevant properties of the medical image being analyzed before constructing a feature vector. We may create an encrypted watermark by XORing the feature vector with the logical-key-vector. After decoding the encrypted watermark, it may be used to determine who owns the watermark and ownership information for the medical image by computing the normalized correlation (NC) coefficient. In most cases, the NC value will be higher than 50%.

Wavelet-domain hidden Markov models (HMMs), in particular, hidden Markov tree (HMT), were recently proposed and applied to image processing, where it was usually assumed that three subbands of the two-dimensional discrete wavelet transform (DWT), i.e., HL, LH, and HH, are independent. In this paper, we study wavelet-based texture analysis and synthesis using HMMs. Particularly, we develop a new HMM, called HMT-3S, for statistical texture characterization in the wavelet domain. In addition to the joint statistics captured by HMT, the new HMT-3S can also exploit the cross correlation across DWT subbands. Meanwhile, HMT-3S can be characterized by using the graphical grouping technique, and has the same tree structure as HMT. The proposed HMT-3S is applied to texture analysis, including classification and segmentation, and texture synthesis with improved performance over HMT. Specifically, for texture classification, we study four wavelet-based methods, and experimental results show that HMT-3S provides the highest percentage of correct classification of over 95% upon a set of 55 Brodatz textures. For texture segmentation, we demonstrate that more accurate texture characterization from HMT-3S allows the significant improvements in terms of both classification accuracy and boundary localization. For texture synthesis, we develop an iterative maximum likelihood-based texture synthesis algorithm which adopts HMT or HMT-3S to impose the joint statistics of the texture DWT, and it is shown that the new RMT-3S enables more visually similar results than HMT does.

In the fingerprint recognition application utilizing more information other than minutiae is much helpful. We present here a fingerprint matching scheme based on transform features and their comparison. The technique described here obviates the need for extracting minutiae points to match fingerprint images. The proposed scheme uses Discrete Cosine Transform (DCT), Fast Fourier Transform (FFT) and Discrete Wavelet Transform (DWT) to create feature vector for fingerprints. After finding out the core point, fingerprint image of size 64times64 is cropped around the core point. The transform is applied on the cropped image without any pre-processing. The transform coefficients are arranged in specific manner and are used to obtain the feature vector in terms of standard deviation. The fingerprint matching is based on the minimum Euclidean distance between two feature vectors. Here database is formed by capturing 8 images per person using 500 dpi optical scanner. Training images used to form feature vector are 2, 4 or 6 per person. In the matching phase either all or remaining images are checked in identification mode to find out the percentage recognition rate. Comparison for all the three transform is presented here and it is observed that DCT and DFT gives better result as compared DWT.

In any multi-script environment, handwritten script classification is an unavoidable pre-requisite before the document images are fed to their respective Optical Character Recognition (OCR) engines. Over the years, this complex pattern classification problem has been solved by researchers proposing various feature vectors mostly having large dimensions, thereby increasing the computation complexity of the whole classification model. Feature Selection (FS) can serve as an intermediate step to reduce the size of the feature vectors by restricting them only to the essential and relevant features. In the present work, we have addressed this issue by introducing a new FS algorithm, called Hybrid Swarm and Gravitation-based FS (HSGFS). This algorithm has been applied over three feature vectors introduced in the literature recently—Distance-Hough Transform (DHT), Histogram of Oriented Gradients (HOG), and Modified log-Gabor (MLG) filter Transform. Three state-of-the-art classifiers, namely, Multi-Layer Perceptron (MLP), K-Nearest Neighbour (KNN), and Support Vector Machine (SVM), are used to evaluate the optimal subset of features generated by the proposed FS model. Handwritten datasets at block, text line, and word level, consisting of officially recognized 12 Indic scripts, are prepared for experimentation. An average improvement in the range of 2–5% is achieved in the classification accuracy by utilizing only about 75–80% of the original feature vectors on all three datasets. The proposed method also shows better performance when compared to some popularly used FS models. The codes used for implementing HSGFS can be found in the following Github link: https://github.com/Ritam-Guha/HSGFS.

IoMT-fog-cloud based architecture for Covid-19 detection

This paper introduces a novel automatic hybrid classifier to detect and classify the Power Quality(PQ) problems in power systems using Wavelet Packet Transform (WPT) and Artificial Neural Networks (ANN). Various PQ events like Normal, Sag, Swell and Interruptions are obtained by modeling three phase distribution system using MATLAB simulink. For classification, the selection of suitable features from the disturbance signal is extremely important. Recent literature survey deals with various signal processing techniques like Fast Fourier Transform (FFT), Discrete Wavelet Transform (DWT) and Wavelet Packet Transform (WPT) were used for feature extraction. From the above techniques, the main issues addressed are selection of optimal feature subset and the design of ANN architecture model. To build a competent and robust classifier, it is essential to extract the utilizable feature vectors from the disturbance signal that can optimize data size as well as incorporate the main characteristics of the signal. This paper addresses these issues, and the distinctive feature vectors are obtained with reduced number of coefficients by using Energy Entropy based WPT. From these, the best discriminative feature vectors with reduced number of coefficients are obtained and are used as input to ANN, so that the burden over classification can be reduced. The classification performance is compared with FFT based ANN. The simulation results obtained have significant improvement over existing methods. Thus the proposed hybrid classifier provides the most excellent detection among PQ problems that arises in real time by improving classification accuracy in terms of both computation time and means square error.

We present, in this paper, a wavelet-based acoustic signal analysis to remotely recognize military vehicles using their sound intercepted by acoustic sensors. Since expedited signal recognition is imperative in many military and industrial situations, we developed an algorithm that provides an automated, fast signal recognition once implemented in a real-time hardware system. This algorithm consists of wavelet preprocessing, feature extraction and compact signal representation, and a simple but effective statistical pattern matching. The current status of the algorithm does not require any training. The training is replaced by human selection of reference signals (e.g., squeak or engine exhaust sound) distinctive to each individual vehicle based on human perception. This allows a fast archiving of any new vehicle type in the database once the signal is collected. The wavelet preprocessing provides time-frequency multiresolution analysis using discrete wavelet transform (DWT). Within each resolution level, feature vectors are generated from statistical parameters and energy content of the wavelet coefficients. After applying our algorithm on the intercepted acoustic signals, the resultant feature vectors are compared with the reference vehicle feature vectors in the database using statistical pattern matching to determine the type of vehicle from where the signal originated. Certainly, statistical pattern matching can be replaced by an artificial neural network (ANN); however, the ANN would require training data sets and time to train the net. Unfortunately, this is not always possible for many real world situations, especially collecting data sets from unfriendly ground vehicles to train the ANN. Our methodology using wavelet preprocessing and statistical pattern matching provides robust acoustic signal recognition. We also present an example of vehicle recognition using acoustic signals collected from two different military ground vehicles. In this paper, we will not present the mathematics involved in this research. Instead, the focus of this paper is on the application of various techniques used to achieve our goal of successful recognition.© (1996) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.