Subband Decomposition Research Articles

Detection Technique for Defects on Steel Sheet Surface by Texture Analysis Using Gabor Filters with Sub-band Decomposition Considering Defect Orientation

Detection Technique for Defects on Steel Sheet Surface by Texture Analysis Using Gabor Filters with Sub-band Decomposition Considering Defect Orientation

A Novel Method for Breast Cancer Classification: A Signal Processing-Based Approach in Ultrasound Images

Breast cancer, a leading cause of mortality among women worldwide, the importance of accurate and efficient diagnostic methods is emphasized. This study contributes to the literature on breast cancer classification, particularly using breast ultrasound images, with a new method using a signal processing approach. It introduces a novel approach by combining features extracted from signals obtained from breast ultrasound images with signals from Variational Mode Decomposition (VMD) sub-bands. The results demonstrate that utilizing features from both preprocessed raw data and VMD sub-band signals can effectively distinguish benign and malignant breast ultrasound images. Classification performance varied depending on the algorithms and data used. According to the numerical results, the highest classification performance was achieved through the study with balanced data using the artificial neural network method, with an area under the curve value of 0.9971 and an accuracy value of 0.9821.

An image denoising method based on the nonlinear Schrödinger equation and spectral subband decomposition

Most denoising methods inevitably smooth the image while denoising, which makes it difficult to effectively maintain the structural information and detail information of the original noise-free image. Stochastic resonance can directly convert the noise energy into signal energy, and denoising by stochastic resonance retains the signal energy. Therefore, this paper presents an image denoising method based on the nonlinear Schrödinger equation. First, by analyzing the dynamic stochastic resonance effect in a nonlinear optical system, the nonlinear Schrödinger equation is derived to describe the signal propagation in the system, and the equation is extended to construct a two-dimensional image denoising model. Then, we propose an image denoising algorithm based on image spectral subband decomposition. The algorithm decomposes the image spectrum into different frequency domain subbands and uses the Schrödinger denoising model to denoise the high-frequency subbands containing the rich structure and detailed information of the image. A series of comparative experiments show that the proposed method can significantly improve the image signal-to-noise ratio while effectively restoring image details and has better performance than other typical methods in the case of high-intensity noise.

A new time–frequency representation based on the tight framelet packet for telephone-band speech coding

To improve the quality and intelligibility of telephone-band speech coding, a new time–frequency representation based on a tight framelet packet transform is proposed in this paper. In the context of speech coding, the effectiveness of this representation stems from its resilience to quantization noise, and reconstruction stability. Moreover, it offers a sub-band decomposition and good time–frequency localization according to the critical bands of the human ear. The coded signal is obtained using dynamic bit allocation and optimal quantization of normalized framelet coefficients. The performances of the corresponding method are compared to the critically sampled wavelet packet transform. Extensive simulation revealed that the proposed speech coding scheme, which incorporates the tight framelet packet transform performs better than that based on the critically sampled wavelet packet transform. Furthermore, it ensures a high bit-rate reduction with negligible degradation in speech quality. The proposed coder is found to outperform the standard telephone-band speech coders in term of objective measures and subjective evaluations including a formal listening test. The subjective quality of our codec at 4 kbps is almost identical to the reference G.711 codec operating at 64 kbps.

Massive Ultrasonic Data Compression Using Wavelet Packet Transformation Optimized by Convolutional Autoencoders.

Ultrasonic signal acquisition platforms generate considerable amounts of data to be stored and processed, especially when multichannel scanning or beamforming is employed. Reducing the mass storage and allowing high-speed data transmissions necessitate the compression of ultrasonic data into a representation with fewer bits. High compression accuracy is crucial in many applications, such as ultrasonic medical imaging and nondestructive testing (NDT). In this study, we present learning models for massive ultrasonic data compression on the order of megabytes. A common and highly efficient compression method for ultrasonic data is signal decomposition and subband elimination using wavelet packet transformation (WPT). We designed an algorithm for finding the wavelet kernel that provides maximum energy compaction and the optimal subband decomposition tree structure for a given ultrasonic signal. Furthermore, the WPT convolutional autoencoder (WPTCAE) compression algorithm is proposed based on the WPT compression tree structure and the use of machine learning for estimating the optimal kernel. To further improve the compression accuracy, an autoencoder (AE) is incorporated into the WPTCAE model to build a hybrid model. The performance of the WPTCAE compression model is examined and benchmarked against other compression algorithms using ultrasonic radio frequency (RF) datasets acquired in NDT and medical imaging applications. The experimental results clearly show that the WPTCAE compression model provides improved compression ratios while maintaining high signal fidelity. The proposed learning models can achieve a compression accuracy of 98% by using only 6% of the original data.

Another look at SSIM image quality metric

We review the design of the SSIM quality metric and offer an alternative model of SSIM computation, utilizing subband decomposition and identical distance measures in each subband. We show that this model performs very close to the original and offers many advantages from a methodological standpoint. It immediately brings several possible explanations of why SSIM is effective. It also suggests a simple strategy for band noise allocation optimizing SSIM scores. This strategy may aid the design of encoders or pre-processing filters for video coding. Finally, this model leads to more straightforward mathematical connections between SSIM, MSE, and SNR metrics, improving previously known results.

Attention Augmented Face Morph Detection

Morph detection is of paramount significance when the integrity of Automatic Face Recognition (AFR) systems is concerned. Considering the risks incurred by morphing attacks, a robust automated morph detector is required which can distinguish authentic bona fide samples from altered morphed images. We leverage the wavelet sub-band decomposition of an input image, yielding the fine-grained spatial-frequency content of the input image. To enhance detection of morphed images, our goal is to find the most discriminative information across frequency channels and spatial domain. To this end, we propose an end-to-end attention-based deep morph detector which assimilates the most discriminative wavelet sub-bands of a given image which are obtained by a group sparsity representation learning scheme. Specifically, our group sparsity-constrained Deep Neural Network (DNN) learns the most discriminative wavelet sub-bands (channels) of an input image while the attention mechanism captures the most discriminative spatial regions of input images for the downstream task of morph detection. To this end, we adopt three attention mechanisms to diversify our refined features for morph detection. As the first attention mechanism, we employ the Convolutional Block Attention Module (CBAM) which provides us with refined feature maps. As the second attention mechanism, compatibility scores across spatial locations and output of our DNN highlights the most discriminative regions, and lastly, the multiheaded self-attention augmented convolutions account for our third attention mechanism. We evaluate the efficiency of our proposed framework through extensive experiments using multiple morph datasets that are compiled using bona fide images available in the FERET, FRLL, FRGC, and WVU Twin datasets. Most importantly, our proposed methodology has resulted in reduction in detection error rates when compared with state-of-the-art results. Finally, to further assess our multi-attentional morph detection, we delve into different combinations of attention mechanisms via a comprehensive ablation study.

Subband Decomposition Based Output-Only Modal Analysis

AbstractOutput-only modal analysis (OMA) is an indispensable alternative to experimental modal analysis for engineering structures while in operation. Conventional OMA often fails to identify the underlying modal structure with insufficient modal energy contribution. Such low modal participation is expected when the sampled response is subjected to sensor nonlinearity or when specific modes are not directly excited. A novel subband decomposition (SBD) method that resolves modal parameters even with biased modal energy distribution is proposed. It isolates the system response within a narrow frequency subband through a finite impulse response analysis filter bank. Whenever the filter subband captures a resonance, the filtered system response is close-to-singular and contains mainly the resonant mode contribution. A modal cluster metric is defined to identify the resonant normal modes automatically. The modal parameters are also identified and extracted within the subband possessing the locally maximal clustering measure. The proposed method assumes no a priori knowledge of the structure under operation other than the system should have any repeated natural frequencies. Therefore, the SBD algorithm is entirely data-driven and requires minimal user intervention. To illustrate the concept and the accuracy of the proposed SBD, numerical experiments of a linear cantilevered beam with various stationary and non-stationary loading are conducted and compared to other OMA methods. Furthermore, physical experiments on an aluminum cantilever beam examine the method’s applicability in field modal testing. Compared to traditional OMA methods, the numerical and physical experiments show orders of magnitude improvement in modal identification error using the proposed SBD.

Jomodevi: A joint motion and depth visibility prediction algorithm for perceived stereoscopic 3D quality

In this paper, we introduce a completely blind and unsupervised no-reference model that performs the quality assessment of stereoscopic 3D videos by using joint motion and depth visibility. In order to achieve such objective metric, first we compute the correlation map between the motion vector map and the depth map, and then we study the natural scene statistics of the estimated correlation map. We empirically model these statistics with Univariate Generalized Gaussian Distribution (UGGD) and compute the UGGD parameters at multi-scale and multi-orient steerable subband decomposition. It is shown that the estimated UGGD features are capable of discriminating distortions. We then compute the mean vector and the covariance matrix from the estimated UGGD features of the correlation maps of motion vector and the depth maps of undistorted and distorted S3D videos. This is followed by the computation of the Wave Hedges distance between the mean vectors of the undistorted and distorted contents, and the Bhattacharyya distance is measured between the corresponding covariance matrices. We pool these distance measures to achieve the overall joint motion and depth quality of S3D videos. The efficiency of the proposed algorithm is evaluated on the well-known IRCCYN, LFOVIAS3DPh1 and LFOVIAS3DPh2 S3D video datasets. Our model demonstrates robust and consistent performance across all distortion types and shows competitive performance against other 2D and 3D image and video quality assessment algorithms. Furthermore, since the proposed algorithm is completely blind, it does not require any training and testing analysis on the content features and the subjective scores.

A Variable-Scale Coherent Integration Method for Moving Target Detection in Wideband Radar

Accurate integration of the extended target’s energy is one of the important challenges of moving target detection in wideband radar. In this paper, a coherent integration method for wideband radar, i.e., variable-scale moving target detection (VSMTD), is proposed to resist range migration and Doppler broadening. On the one hand, subband decomposition can effectively integrate the energy of the extended target in range using variable-scale transformation, accomplished by modulating the filter bank. On the other hand, it increases the coherent integration time by mitigating the range migration in a sufficiently narrow subband. The discrete Fourier transform (DFT) modulated filter bank and the fast Fourier transform (FFT) algorithm are also used to achieve fast VSMTD implementation. Finally, the simulation results demonstrate the superior performance of the proposed VSMTD method.

An Automatic Detection by Classification of Cracked Pixels or Noncracked Pixels in Road Surface

Automatic detection and monitoring of the condition of cracks in the road surface are essential elements to ensure road safety and quality of service. A crack detection method based on wavelet transforms (2D-DWT) and Jerman enhancement filter is used. This paper presents different contributions corresponding to the three phases of the proposed system. The first phase presents the contrast enhancement technique to improve the quality of roads surface image. The second phase proposes an effective detection algorithm using discrete wavelet (2D-DWT) with “db8” and two-level sub-band decomposition. Finally, in the third phase, the Jerman enhancement filter is usually used with different parameters of the control response uniformity “ τ ” to enhance for cracks detection. The experimental results in this article provide very powerful results and the comparisons with five existing methods show the effectiveness of the proposed technique to validate the recognition of surface cracks.

High-performance and ultra-compact spike-based architecture for real-time acoustic echo cancellation

In recent years, the implementation of advanced adaptive filters in embedded devices for acoustic echo cancellation has increased because most of them are used in portable devices, especially in Internet of Things systems, in which high-performance and low-area digital hardware implementations are required. In this work, we present a high-performance and ultra-compact spike-based hardware architecture to efficiently compute adaptive filters to be used as an acoustic echo canceller (AEC). To achieve this architecture, we address two factors. (1) We propose a new data-selective adaptive filter along with subband decomposition least mean square (LMS) method to reduce the computational cost by minimizing the number of operations required to efficiently update the filter coefficients. The proposed method requires approximately 40% fewer updates when compared with conventional subband adaptive filters. As a consequence, the spike-based hardware architecture simulates the proposed adaptive filter at high processing speeds. (2) We propose a compact and high-precision neural multiplier since acoustic echo cancellers require a large number of high-precision multiplications to efficiently identify the echo path. The proposed neural multiplier expends up to 15 times fewer synapses when compared with existing approaches, which represents a significant improvement in terms of area. In addition, this improvement avoids routing problems by implementing large-scale synapse connectivity in advanced FPGAs. Herein, we carry out exhaustive testing by simulating several acoustic echo scenarios. The results demonstrate that the features of the model and the hardware implementation techniques potentially allow easy integration into portable devices for use in real-world acoustic echo cancellation scenarios.

Multirate Audiometric Filter Bank for Hearing Aid Devices.

The frequency-dependent nature of hearing loss poses many challenges for hearing aid design. In order to compensate for a hearing aid user's unique hearing loss pattern, an input signal often needs to be separated into frequency bands, or channels, through a process called sub-band decomposition. In this paper, we present a real-time filter bank for hearing aids. Our filter bank features 10 channels uniformly distributed on the logarithmic scale, located at the standard audiometric frequencies used for the characterization and fitting of hearing aids. We obtained filters with very narrow passbands in the lower frequencies by employing multi-rate signal processing. Our filter bank offers a 9.1× reduction in complexity as compared to conventional signal processing. We implemented our filter bank on Open Speech Platform, an open-source hearing aid, and confirmed real-time operation.

Classification of Sleep Apnea Based on Sub-Band Decomposition of EEG Signals.

To classify between normal and sleep apnea subjects based on sub-band decomposition of electroencephalogram (EEG) signals. This study comprised 159 subjects obtained from the ISRUC (Institute of System and Robotics—University of Coimbra), Sleep-EDF (European Data Format), and CAP (Cyclic Alternating Pattern) Sleep database, which consists of normal and sleep apnea subjects. The wavelet packet decomposition method was incorporated to categorize the EEG signals into five frequency bands, namely, alpha, beta, delta, gamma, and theta. Entropy and energy (non-linear) for all bands was calculated and as a result, 10 features were obtained for each EEG signal. The ratio of EEG bands included four parameters, including heart rate, brain perfusion, neural activity, and synchronization. In this study, a support vector machine with kernels and random forest classifiers was used for classification. The performance measures demonstrated that the improved results were obtained from the support vector machine classifier with a kernel polynomial order 2. The accuracy (90%), sensitivity (100%), and specificity (83%) with 14 features were estimated using the data obtained from ISRUC database. The proposed study is feasible and seems to be accurate in classifying the subjects with sleep apnea based on the extracted features from EEG signals using a support vector machine classifier.

Multimodal Medical Image Fusion Techniques – A Review

The main objective of image fusion for multimodal medical images is to retrieve valuable information by combining multiple images obtained from various sources into a single image suitable for better diagnosis. In this paper, a detailed survey on various existing medical image fusion algorithms, with a comparative discussion is presented. Image fusion algorithms available in the current literature are categorized into various methods known as (1) morphological methods, (2) human value system operator based methods, (3) sub-band decomposition methods, (4) neural n++32w1etwork based methods, and (5) fuzzy logic based methods. This research concludes that even though there exist a few open-ended creative and logical difficulties, the fusion of medical images in many combinations assists in utilizing medical image fusion for medicinal diagnostics and examination. There is a tremendous progress in the fields of deep learning, artificial intelligence and bio-inspired optimization techniques. Effective utilization of these techniques can be used to further improve the efficiency of image fusion algorithms.

EEG based visual cognitive workload analysis using multirate IIR filters

In this study, the visual cognitive workload is classified using electroencephalogram (EEG) signal acquired with dry electrodes. The visual cognitive workload of four levels is introduced using color objects and their counting in visual space. The EEG data of 44 subjects is acquired for this work. The classification accuracy along with computational complexity involved in subband decomposition technique is discussed in the proposed approach. The EEG signal is decomposed in ten subbands using multirate infinite impulse response (IIR) filter bank. Band energy, signal length, non-linear energy and band entropy of decomposed subbands are used as features. K-nearest neighbors (KNN) and ensemble subspace KNN classifiers are used for classification of cognitive workload. The classification accuracy and computational complexity are compared with two other subband decomposition methods: Morlet wavelet and analytic wavelet transform. The classification accuracy obtained using proposed method is 83.65%, which is greater than the Morlet wavelet (80.96%) and is less than the analytic wavelet transform (86.31%). The proposed method's computational complexity is 70% less than Morlet wavelet and 45% less than analytic wavelet transform.

An effective arrhythmia classification via ECG signal subsampling and mutual information based subbands statistical features selection

The deployment of wireless health wearables is increasing in the framework of mobile health monitoring. The power and processing efficiencies with data compression are key aspects. To this end, an efficient automated arrhythmia recognition method is devised. The aim of this work is to contribute to the realisation of modern wireless electrocardiogram (ECG) gadgets. The proposed system utilizes an intelligent combination of subsampling, denoising and wavelet transform based subbands decomposition. Onward, the subband’s statistical features are extracted and mutual information (MI) based dimension reduction is attained for an effective realization of the ECG wearable processing chain. The amount of information to be processed is reduced in a real-time manner by using subsampling. It brings a remarkable reduction in the proposed system's computational complexity compared to the fixed-rate counterparts. MI based features selection improves the classification performance in terms of precision and processing delay. Moreover, it enhances the compression gain and aptitudes an effective diminishing in the transmission activity. Experimental results show that the designed method attains a 4 times computational gain while assuring an appropriate quality of signal reconstruction. A 7.2-fold compression gain compared to conventional counterparts is also attained. The best classification accuracies of 97% and 99% are secured respectively for cases of 5-class and 4-class arrhythmia datasets. It shows that the suggested method realizes an effective recognition of arrhythmia.

High resolution sub-band decomposition underdetermined blind signal separation using virtual sensor based ICA method for low latency applications

Underdetermined blind source separation (UBSS) objective is to recover the source signals from a number of mixtures without any information about the mixing system. Technologies such as teleconferencing, hearing aids and hands-free telephony require real time processing, as long delays are considered intolerable in interactive two-way communication. This is a major challenge in BSS, as algorithms generally require significant amounts of data (several seconds of data or more) to generate sufficient statistics for separation. In this paper, we propose a hybrid algorithm which combines sub-band decomposition and well-known Independent Component Analysis (ICA) based algorithm, Extended-Infomax. We first employ sub-band decomposition algorithm in sparse time domain to compensate low data efficiency of short time block lengths and estimate the mixing matrix. Subsequently, the proposed virtual sensor based underdetermined Extended-Infomax source model is used to estimate the source signals. As the separation evaluation results are so sensitive to the mixing matrix estimation outcome, achieving nearly optimum results in the first phase of the proposed algorithm significantly improves the results of source separation performance. A weighted version of k-plane clustering algorithm is derived to obtain the mixing coefficients. Experimental evaluations reveal the effectiveness of our proposed method over the state-of-the-art techniques.

Effective epileptic seizure detection by using level-crossing EEG sampling sub-bands statistical features selection and machine learning for mobile healthcare

Mobile healthcare is an emerging approach which can be realized by using cloud-connected biomedical implants. In this context, a level-crossing sampling and adaptive-rate processing based innovative method is suggested for an effective and automated epileptic seizures diagnosis. The suggested solution can achieve a significant real-time compression in computational complexity and transmission activity reduction. The proposed method acquires the electroencephalogram (EEG) signal by using the level-crossing analog-to-digital converter (LCADC) and selects its active segments by using the activity selection algorithm (ASA). This effectively pilots the post adaptive-rate modules such as denoising, wavelet based sub-bands decomposition, and dimension reduction. The University of Bonn and Hauz Khas epilepsy-detection databases are used to evaluate the proposed approach. Experiments show that the proposed system achieves a 4.1-fold and 3.7-fold decline, respectively, for University of Bonn and Hauz Khas datasets, in the number of samples obtained as opposed to traditional counterparts. This results in a reduction of the computational complexity of the proposed adaptive-rate processing approach by more than 14-fold. It promises a noticeable reduction in transmitter power, the use of bandwidth, and cloud-based classifier computational load. The overall accuracy of the method is also quantified in terms of the epilepsy classification performance. The proposed system achieves100% classification accuracy for most of the studied cases.

A High Frequency Vibration Compensation Approach For Terahertz SAR Based on Sinusoidal Frequency Modulation Fourier Transform

In this paper, a vibration estimation method for terahertz synthetic aperture radar (THz SAR) is proposed based on the sinusoidal frequency modulation Fourier transform (SFMFT), which directly operates on the phase term of the signal. Small high frequency vibration of the platform will introduce nonlinear phase modulation in radar echoes. To improve the potential of the method for vibration with multiple vibration frequencies, SFMFT is used to extract the vibration spectrum of the platform. When the vibration amplitude exceeds a certain threshold, the phase ambiguity will occur. To improve the range of estimable vibration amplitude, a method based on the conjugate cancellation of the phase errors with sub-band decomposition is introduced to reconstruct a signal with a longer equivalent wavelength. Then the accurate phase errors caused by the platform vibration can be estimated by performing SFMFT on the composite signal. The simulations prove that the proposed method is superior to the time-frequency (TF) analysis methods and can improve the range of estimable vibration amplitude.