The Nonlinear Study of Composite Plates Natural Frequency Using RNN-LSTM combined with Wavelet Transform Features

An efficient face representation is a vital step for a successful face recognition system. Gabor features are known to be effective for face recognition. The Gabor features extracted by Gabor filters have large dimensionality. The feature of wavelet transformation is feature reduction. Hence, the large dimensional Gabor features are reduced by wavelet transformation. The discriminative common vectors are obtained using the within-class scatter matrix method to get a feature representation of face images with enhanced discrimination and are classified using radial basis function network. The proposed system is validated using three face databases such as ORL, The Japanese Female Facial Expression (JAFFE) and Essex Face database. Experimental results show that the proposed method reduces the number of features, minimizes the computational complexity and yielded the better recognition rates.

Contourlet transform is a new two-dimensional extension of the wavelet transforms using multiscale and directional filter banks. In this paper, the effectiveness of the features obtained from the contourlet transform is investigated and is compared with the wavelet transform features for image texture classification. We specially focused on image acquisition conditions that an image from one scene may be acquired with different illumination, scale, direction, distance and slope. It is shown that the accuracy of the contourlet transform features in such conditions is more than that of the wavelet transform. However, wavelet transform is still applicable in many texture classification tasks.

Increasing suspicious instructions of various malware through a challenge to the malware analysts to identify and classify samples belongs to the malicious family. They have witnessed the very fast increase in both the number and complexity of malware set of instructions. Malware invest profoundly in technology and capability to reorganize the process of building and mutate existing malware set of instructions to avoid traditional protection. Classify malware variants by applying image processing techniques. The textures play an important role in many image processing applications. In this paper we proposed the Support Vector Machine (SVM) multi-class malware image classification challenge from an image processing perspective. The multi-resolution and wavelets are used to build effective texture feature vector using Gabor Wavelet, GIST and Discrete wavelet Transform and other features. The proposed algorithm experimented on Malimg Dataset of malware total 12,470 samples are used. In that 1610 samples are trained and 1710 samples are tested on 8 malware family which is randomly selected from the dataset. We compare this approach to existing malware classification approaches previously published research work. This is an efficient and more accurate malware detection algorithm using Wavelet Transform with machine learning classifiers techniques to detect malware samples more capably compare to existing work.

MRI Brain Image Segmentation is one of the difficult and complex techniques in the medical field. Normally the pathological tissues such as Tumor and Edema are easily segmented. In this paper, both the normal tissues such as WM (White Matter), GM (Gray Matter) and CSF (Cerebrospinal Fluid) and the pathological tissues such as Tumor, Edema and also Atrophy in the MRI Brain Images are segmented effectively. Initially, the Wavelet Transform features and the Semantic feature from the MRI Brain Images are extracted in two different ways. These extracted features are the input to the next process. Then the proposed segmentation technique performs classification process by utilizing a dual Artificial Neural Network. The ANN is helpful for classifying whether the image is normal or abnormal. Based on the results, the segmentation is carried out. In Segmentation, the normal tissues such as WM, GM and CSF are segmented from the normal MRI images and pathological tissues such as Tumor, Edema and Atrophy are segmented from the abnormal images. The implementation result shows the efficiency of proposed tissue segmentation technique in segmenting the tissues accurately from the MRI images. The performance of the segmentation technique is evaluated by performance measures such as accuracy, specificity and sensitivity.

Vortex flowmeter was chosen to measuring the oil-gas-water three-phases flow due to its various of characteristics, experiment has been reported in the conditions as follows: vertical upward pipeline with 50 mm diameter, the volumetric flow of water is 5–8 m3 h−1, oil account for 5–30 % of the total mixed flow and the gas proportion of the gas void fraction is from 1 to 5 %. The signals collected by vortex sensor are compressed and de-noised by ddencmp function. As PSD (power spectral density) peak values and variations in water flow have a very clear relationship, based on the feature of wavelet transformation, criteria may be found to distinguish different flow. The result shows that low-frequency energy is greatly influenced by the oil content and low-frequency energy is declined with the oil content is increased while low-frequency energy is increased with the gas void fraction is increased. There is a flow threshold, when the three-phase flow does not reach the threshold, the influence on Low-frequency energy by the gas void fraction can be ignored. The volume measured by the vortex flow-meter was less influenced by media with low gas void fraction.

Due the rapid growth in the area of Digital Image Processing the semantic based techniques are also been emerged for an efficient processing. In order to achieve an efficient result; this paper a proposed a technique for the generation of image content descriptor with three features viz., Color auto-Correlogram, Gabor Wavelet and Wavelet Transform. Color Auto-Correlogram Feature is associated with color information of an image which is derived from the RGB color space of an image. The Gabor Wavelet Feature is has the texture information to extract the textural features associated with the image and the Wavelet Transform Feature is linked with shape information in the extraction of edges in an image. The feature extraction process is accomplished based on the input query image from the IDB and the features are stored in a feature dataset. The Manhattan distance is applied on the user given query image and feature vector computed from database images for measuring similarity. Finally, the proposed technique retrieves the meaningful image from the image database which satisfies the user expectation. The performance of the retrieval system has been analyzed by the performance measures Precision and Recall. The efficiency of the proposed feature descriptor is tested for CBIR system using Corel image database, Li image database and Caltech-101 image database.

The dynamic video frame dataset’s automated feature analysis addresses the complexity of intensity mapping with normal and abnormal classes. Iterative modelling is needed to learn the component of a video frame in several patterns for various video frame data types for threshold-based data clustering and feature analysis. GWO optimises the Convoluted Pattern of Wavelet Transform (CPWT) feature vectors employed in this paper's CNN feature analysis technique. A median filter reduces noise and smooths the video frame before normalising it. Edge information represents the video frame's bright spot boundary. Neural network based video frame classification clusters pixels using feature recurrent learning with minimal dataset training. The filtered video frame's features were evaluated using complex wavelet transformation feature extraction algorithms. These features demonstrate video frame spatial and textural classifications. CNN classifiers help analyse video frame instances and classify action labels. Categorization improves with the fewest training datasets. This strategy may be beneficial if compared to optimal practises.

There are various methods to extract feature from EEG signals but the effective feature selection is an issue. In this paper, a novel effective feature selection based on Statistical-Principal Component Analysis (S-PCA) and wavelet transform (WT) features in medical and BCI application is proposed. In this method, we decompose the signals to six sub-bands by four mother wavelet (sym6, db5, bior1.5 and robio2.8). Then five features (such as the number of zero coefficients, the smallest and largest coefficients, the mean and standard deviation of coefficients) extract from each sub-band as feature vector. In this algorithm, S-PCA is used to select ten effective features from among WT features. Finally, we use KNN classifier and seven different signals of brain activities to evaluate the proposed method. The results indicate the improvement of the classification performance in comparison with current methods.

Face recognition is used to identity a person effectively and most effective physiological biometric trait. In this paper, we propose sorting pixels-based face recognition using Discrete Wavelet Transform (DWT) and statistical features. The novel concept of sorting pixel values in ascending order is introduced and segmented into two parts viz., Low Pixel Values (LPV) and High Pixel Values (HPV). The DWT is applied on LPV matrix to generate low and high frequency bands such as LL, LH, HL and HH. The low frequency LL band is considered for features as the coefficient values are enhanced compared to original image pixel values and also reduction in dimensionality. The statistical measure is applied on HPV to compute mean, median, mode, maximum and standard deviation features. The features of LL band and statistical features are concatenated to obtain final features. The Artificial Neural Network (ANN) is used as classifier to recognize human beings. It is perceived that the performance of the proposed method is enhanced compared with the existing methods.

In the digital age, image manipulation is common, often done before publication on social media. However, this can lead to negative impacts, including visual deception. This research aims to detect splicing type image manipulation using Dyadic Wavelet Transform (DyWT) and Scale Invariant Feature Transform (SIFT) methods. The process starts with image decomposition using DyWT to obtain LL sub-images, followed by local feature extraction using SIFT. An application built on desktop-based Matlab source was developed to detect splicing forgery in digital images. The test used 20 images, this image dataset was taken from canon 5d mark II camera and Vivo X80 mobile phone. Each 10 original images, and 10 edited images. These 10 original images are left as they are without making changes, editing or manipulation, while the other 10 images are changed, edited or manipulated using editing software, the results of this editing are uploaded to social media, such as Facebook and Instagram, which will later be used as datasets in testing. The results show that the splicing technique is detected accurately, and processing is faster on images with low pixel resolution. The DyWT and SIFT methods are effective in detecting post-processing attacks such as rotation and rescaling, although they have drawbacks. DyWT struggles in detecting subtle changes and noise, while SIFT is less effective on non-geometric manipulations. Overall, both methods face challenges in detecting complex manipulations and require significant computational resources, especially on high-resolution images.

Epilepsy is a common neurological disease worldwide, characterized by recurrent seizures. There is currently no cure for epilepsy. However, seizures can be controlled by drugs and surgeries in about 70% of epileptic patients. A timely and accurate prediction of seizures can prevent injuries during seizures and improve the patients’ quality of life. In this paper, we proposed an intelligent epileptic prediction system based on Synchrosqueezed Wavelet Transform (SWT) and Multi-Level Feature Convolutional Neural Network (MLF-CNN) for smart healthcare IoT network. In this system, we used SWT to map EEG signals to the frequency domain, which was able to measure the energy changes in EEG signals caused by seizures within a well-defined Time-Frequency (TF) plane. MLF-CNN was then applied to extract multi-level features from the processed EEG signals and classify the different seizure segments. The performance of our proposed system was evaluated with the publicly available CHB-MIT dataset and our private ZJU4H dataset. The system achieved an accuracy of 96.99% and 94.25%, a sensitivity of 96.48% and 97.76%, a specificity of 97.46% and 94.07% and a false prediction rate (FPR/h) of 0.031 and 0.049 FPR/h on the CHB-MIT dataset and the ZJU4H dataset, respectively.

Improving the quality of sleep is an important issue for many researches. A number of biomedical signals, such as EEG, EMG, and EOG were used to classify sleep stages. Based on those signals, one can detect and diagnose the sleep related disorders. There were many researches focused on automatic sleep stages classification. In this research, a new classification method is presented by applying Elman neuron network combined with fuzzy rules and features are extracted by wavelets packets. Nine subjects were recruited from Cheng-Ching General Hospital, Taichung, Taiwan. The sampling frequency is 250Hz and the single channel (C3-A1) EEG signal was acquired for each subject. Combined network was used to recognize the sleep stages in each epoch (a 10 second segment data). The classification results relied on the strong points of neural network and fuzzy logic with average sensitivity is 88.48%, average specificity achieves 95.96%, and average accuracy is 93.79%. The data samples and the length of sleep intervals will be increased for experiment in the future to improve the accuracy.

Nowadays, it becomes important to determine the chemical structure without damaging the samples. As a result of the use of infrared, the spectras are obtained both quickly and without any special sample preparation process, and they contain specific characteristics. In this study, features of Fourier Transform Infrared spectra acquired from olive oil samples are extracted by Wavelet Transform (WT) and Variational Mode Decomposition (VMD) that does not require a main function.Afterwards, these attributes are classified in comparison by using the powerful classifiers, support vector machines (SVM) and random forests (RF). Experimental studies have shown that the features obtained by two proposed methods increase the classification performance.

Mammograms are one of the most widely used techniques for preliminary screening of breast cancers. There is great demand for early detection and diagnosis of breast cancer using mammograms. Texture based feature extraction techniques are widely used for mammographic image analysis. In specific, wavelets are a popular choice for texture analysis of these images. Though discrete wavelets have been used extensively for this purpose, spherical wavelets have rarely been used for Computer-Aided Diagnosis (CAD) of breast cancer using mammograms. In this work, a comparison of the performance between the features of Discrete Wavelet Transform (DWT) and Spherical Wavelet Transform (SWT) based on the classification results of normal, benign and malignant stage was studied. Classification was performed using Linear Discriminant Classifier (LDC), Quadratic Discriminant Classifier (QDC), Nearest Mean Classifier (NMC), Support Vector Machines (SVM) and Parzen Classifier (ParzenC). We have obtained a maximum classification accuracy of 81.73% for DWT and 88.80% for SWT features using SVM classifier.

Facial Expression Recognition is necessary for designing any human-machine interface. The main issue of Facial Expression Recognition is to decide what features are required to represent a Facial Expression. In this paper, we propose the hybrid technique for facial expression recognition. In this paper we proposed a combined method of feature extraction using Discrete Cosine Transform, Gabor Filter, Wavelet Transform and Gaussian distribution to improve the recognition rate. Experimental are performed on seven expressions, (anger,disgust, fear, happiness, sadness, surprise, neutral ) of JAFFE dataset. The result of Proposed work is compared with result of individual Feature Extraction Techniques that show that Facial Expression Recognition Rate can be improved by combining best features of DCT, Gabor Filter, Wavelet Transform and Gaussian Distribution.