Multiscale Noise Research Articles

Multiscale Dynamically Parallel Shrinkage Network for Fault Diagnosis of Aviation Hydraulic Pump and Its Generalizable Applications

Aiming to address the multiscale characteristics and noise corruption problems in the vibration signals of aviation hydraulic pumps, this article develops a novel Multiscale Dynamically Parallel Shrinkage Network (MDPSN) to learn complementary and rich fault-related multiscale features, with the ultimate goal of yielding higher diagnostic accuracy. One significant property is the development of a novel dynamically parallel shrinking module (DPSM) that adaptively generates independent soft thresholds for different scales, effectively shrinking noise-related features to zeros. On one hand, DPSM aggregates and interacts with features at all scales to construct a global feature representation containing richer fault-related information, which is served as the foundation for soft thresholding generation, significantly improving the accuracy and rationality of the generated thresholds. On the other hand, DPSM can adaptively generate individual soft threshold for each scale, allowing each scale to use an independent threshold tailored to its own characteristic to eliminate noise-related information. This avoids the issues of over-denoising or under-denoising caused by the uniform application of thresholds across all scales. Finally, the effectiveness of MDPSN is validated by a series of experiment comparisons on an aviation hydraulic pump dataset and two bearing datasets with various types of noise. The experimental results demonstrate that MDPSN achieves superior diagnostic accuracy compared to five other comparison methods.

Dual channel and multi-scale adaptive morphological methods for infrared small targets

Infrared small target detection is a challenging task. Morphological operators with a single structural element size are easily affected by complex background noise, and the detection performance is easily affected by multi-scale background noise environments. In order to enhance the detection performance of infrared small targets, we propose a dual channel and multi-scale adaptive morphological method (DMAM), which consists of three stages. Stages 1 and 2 are mainly used to suppress background noise, while stage 3 is mainly used to enhance the small target area. The multi-scale adaptive morphological operator is used to enhance the algorithm’s adaptability to complex background environments, and in order to further eliminate background noise, we have set up a dual channel module. The experimental results indicate that this method has shown superiority in both quantitative and qualitative aspects in comparison methods, and the effectiveness of each stage and module has been demonstrated in ablation experiments. The code and data of the paper are placed in https://pan.baidu.com/s/19psdwJoh-0MpPD41g6N_rw.

Finite range decompositions of Gaussian fields with applications to level-set percolation

In a recent work [19], Muirhead has studied level-set percolation of (discrete or continuous) Gaussian fields, and has shown sharpness of the associated phase transition under the assumption that the field has a certain multiscale white noise decomposition, a variant of a finite-range decomposition. We show that a large class of Gaussian fields have such a white noise decomposition with optimal decay parameter. Examples include the discrete Gaussian free field, the discrete membrane model, and the mollified continuous Gaussian free field. This answers various questions from [19].Our construction of the white-noise decomposition is a refinement of Bauerschmidt's construction of a finite-range decomposition [3]. In the continuous setting our construction is very similar to Bauerschmidt's, while in the discrete setting several new ideas are needed, including the use of a result by Pólya and Szegő on polynomials that take positive values on the positive real line.

Coupled Hybrid Stochastic Resonance With Multiobjective Optimization for Machinery Dynamic Signature and Fault Diagnosis

Most traditional stochastic resonance methods pose three fundamental limitations in extracting fault signature: (1) simulated impulse signals are insufficient to match the complicated mechanical vibration signals; (2) one-dimensional signals are vulnerable to interference from multiscale noise, which potentially yields false detection; and (3) optimization function in the signal-to-noise ratio largely depends on the advanced requirements of the characteristic frequency, thereby restricting the development for actual operation. Accordingly, this study investigated the dynamic model of rolling bearings and explored the benefits of the coupled hybrid stochastic resonance (CHSR) method for identifying weak signal signatures. In addition, we proposed multi-objective optimization and established a dynamic model of a rolling bearing with four degrees of freedom under a rotor system with external excitation. Overall, the proposed system offers three main advantages: (1) the dynamic model facilitates correspondence with real vibration signals; (2) the CHSR method can process two-dimensional signals and overcome the saturation phenomenon, which enhances the weak useful signal signature describing machinery health; and (3) the cross-correlation spectral kurtosis and the critical amplitude are defined within the multi-objective function to adaptively optimize the parameter adjustment of the CHSR. Thereafter, the proposed method exhibited adequate enhancement ability integrated the study on dynamic simulation and machinery bearings of wind power generator. Moreover, the applied machine learning model further clarified the superior performance of CHSR method in identifying the weak fault signature.

Bus Cabin Noise Predictions of Large HVAC System using Total Noise Method

HVAC system design has an accountability towards acoustic comfort of passengers of a vehicle. Owing to larger cabin volume of a bus, multiple air blowers have to be installed to ensure comfort of passengers. Such multiple blowers produce significant flow induced noise inside the cabin. For commercial success, it becomes essential to predict intensity of such a flow induced noise at very early stages in product development. Conventionally sliding mesh based CFD approach is deployed to predict flow and turbulence noise around each blower. However due to complexity, this method becomes computationally intensive resulting in cost and time inefficiency. Hence it is desirable to innovate around an alternative rapid, reliable prediction method, which ensures quick turnaround of prediction. This paper describes a unique innovative approach developed around a multiscale method where flow induced noise generated by a single blower in motion is predicted using commercial Lattice Boltzmann CFD software with a digitally scaled down HVAC system in an anechoic digital wind tunnel. These CFD predictions are used to replace all blowers with virtual stationary speakers inside digital cabin to emulate noise emanated by a large HVAC unit. Authors named this method as total Noise Multiscale Approach, in the paper. With the total multiscale approach, overall sound pressure level predicted inside the bus cabin at rear passenger ear levels are comparable with the physical test measurements and has shown fair correlation. Using this multiscale total noise method, computational cost and turnaround time has significantly reduced compared to the flow conventional resolving approach for cabin with all the blowers. This predictive total noise method found useful in designing and planning countermeasures during product development.

Enhanced features in image manipulation detection

Image editing technology leads to tampering traces hidden in images hard to be identified, which results in great challenges for image manipulation detection. Therefore, it is necessary to enhance the characteristic response of tampering traces. In this paper, a two-stream network is exploited to extract tampering features from both semantic level and noise level through end-to-end training. At the input end of the network, a multi-scale high-frequency constrained convolution kernel module is proposed to extract multi-channel and multi-scale noise fingerprint information, which enhances the tampering traces. Moreover, we leverage multi-task learning to realize the classification of tampering techniques, the detection and segmentation of tampering regions. The tampered segmentation mask attention is introduced as a novel constraint to be fed back to the detection branch, which further enhances the tampering characteristics and improves the detection performance. Experiments on COVER, CASIA and NIST16 datasets prove that the proposed framework outperforms existing methods quantitatively and qualitatively. The proposed method effectively enhances the characteristics of tampered regions. Moreover, the proposed method has high robustness on JPEG recompression and image resize.

Identifying fetal status with fetal heart rate: Deep learning approach based on long convolution

CTG (Cardiotocography) is an effective tool for fetal status assessment. Clinically, doctors mainly evaluate the health of fetus by observing FHR (fetal heart rate). The rapid development of Artificial Intelligence has led realization of computer-aided CTG technology, Intelligent CTG classification based on FHR is a fundamental component of these technologies. Its implementation can provide doctors with auxiliary decisions. Most of existing FHR classification methods are based on combing different deep learning models, such as CNN (Convolutional Neural Network), LSTM (Long short-term memory) and Transformer. However, these studies ignore the balance of positive and negative samples in dataset and the matching degree between model and FHR classification task, which reduces the classification accuracy. In this paper, we mainly discuss two major problems in previous FHR classification studies: reduce class imbalance and select appropriate convolution kernel. To address above two problems, we propose a data augmentation method based on ECMN (Edge Clipping and Multiscale Noise) to resolve class imbalance. Subsequently, we introduce a one-dimensional long convolutional layer, which use trend area to calculate the appropriate convolution kernel. Based on appropriate convolution kernel, an improved residual structure with attention mechanism named TGLCN (Trend-Guided Long Convolution Network) is proposed to improve FHR classification accuracy. Finally, horizontal and longitudinal experiments show that the TGLCN obtains high classification accuracy and speed of parameter adjustment.

Three-Dimensional Point Cloud Data Pre-Processing for the Multi-Source Information Fusion in Aircraft Assembly

Wing-body assembly is a key part of aircraft manufacturing, and during the process of wing assembly, the 3D point cloud data of the components are an important basis for attitude adjustment. The large amount of measured point cloud data and the obvious noise affect the quality and efficiency of the final assembly. To address this problem, research on the pre-processing method of the component point cloud data is carried out. Firstly, a feature-enhanced point cloud resampling method is proposed to preserve key features such as part contours in the resampling process. Then, a multi-scale point cloud data noise filtering method is proposed, which can effectively filter out the outliers. The experimental results show that the proposed method improves the speed and accuracy of the subsequent point cloud analysis effectively and is successfully applied to the assembly process of a large passenger aircraft, laying the foundation for high-quality assembly.

Weak signal enhancement for rolling bearing fault diagnosis based on adaptive optimized VMD and SR under strong noise background

Owing to the nonlinearity and nonstationarity of the bearing fault signal, it is difficult to identify fault characteristics under the influence of a strong noise environment. The extraction of early weak fault features is critical for the reliability of bearing operations. Therefore, an urgent problem is reasonable noise reduction and feature enhancement in weak-signal processing. Traditional variational modal decomposition (VMD) and stochastic resonance (SR) are commonly applied to detect weak signals in fault diagnosis. The VMD method can decompose the signal into several intrinsic mode functions (IMFs) to effectively reduce the modal aliasing problem. However, uniform standards for the key parameters of decomposition and the selection of the optimal IMF after decomposition are lacking. Meanwhile, some disadvantages of SR still exist; for example, the interference of multiscale noise may lead to false detection by incorrect selection of high-pass filter parameters, and the system parameters are not adaptive to different signals to achieve the best response output. To better address the weak signal feature enhancement, a novel rolling bearing fault diagnosis method combining adaptive VMD and SR by improved differential search (IDS) optimization is proposed. First, the bearing fault signal is decomposed into several IMFs using the IDS-VMD. Second, the feature information of the fault signal is retained and reconstructed using the correlation kurtosis for sensitive modal extraction. Furthermore, the fault features of the reconstructed signal are effectively enhanced by the variable-step IDS-SR, which can reasonably transfer the noise energy of the input components to the fault characteristic frequency. Finally, the periodic pulse can be observed in the corresponding envelope spectrum. The simulated and experimental data show that the proposed method can not only effectively extract the signal feature information in the actual fault but also realize early weak fault diagnosis of rolling bearings more accurately.

Multiscale Noise Reduction Attention Network for Aeroengine Bearing Fault Diagnosis

Aero-engine has a complex mechanical structure and its working condition are varying widely. Its fault signals are thus modulated through complex nonlinear transfer paths and influenced by non-Gaussian noises. However, popular data-driven multi-scale diagnostic model does not sufficiently consider the embedded noises, and consequently keeps the noise content in the advanced discriminative features, which decreases the diagnostic accuracy. Therefore, a multi-scale attention network with adaptive noise reduction (MANANR) is proposed in this paper for aero-engine bearing fault diagnosis. The MANANR firstly divides the original vibration signal into different scales via the average method of adjacent points. Then, two stacked multi-scale noise reduction modules B1 and B2 is designed for noise reduction. The core strategy behind B1 and B2 is the threshold noise reduction (TNR) which removes the noises from multi-scale convolution features adaptively. Based on the physical principles that global average pooling (GAP) and max pooling (MAP) can extract periodic and impulsive characteristics of the fault signals respectively, the threshold value of TRN is thus constructed through fusing GAP and MAP outputs. Furthermore, attention mechanism is employed to enhance the discriminative capability of multi-scale features by globally capturing the relations among different scales and channels. Finally, one two-layer classifier is introduced to confirm bearing fault patterns. Experimental results demonstrate that the proposed method has the feature-level noise reduction property, and more importantly achieves satisfying intelligent diagnosis precision for aero-engine bearing with the minimum peeling area of 0.5 mm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> under the continuous acceleration condition from 12000 rpm to 12550 rpm, which outperforms state-of-the-art intelligent diagnostic methods.

Variable-scale evolutionary adaptive mode denoising in the application of gearbox early fault diagnosis

Due to the influence of gearbox structure complexity, vibration signals are always corrupted by heavy multiscale noise interference distributed in modulation frequency bands, which brings great challenge for early fault identification. However, conventional adaptive noise cancellation methods mainly focus on a comprehensive denoising effect while ignore improper reference induced by multiscale characteristics of noise interference. This will lead to two major problems. On one aspect, reference filtering at sensitive scales is not mutually independent, and it has a negative influence on each other. On the other aspect, improper reference may cause losses of fault-related information. These limitations greatly hinder the fault features identification, especially in an early fault stage of a fault. Aiming at these problems, this study proposes a novel variable-scale evolutionary adaptive mode denoising method (VEAMD) for weak feature enhancement of gearbox early fault diagnosis. First, to sense the multiscale distribution characteristics of a desired signal, an ergodic VMD process with a dual domain energy factor (DDEF) is presented, where a series of Wiener filters are adaptively designed. Secondly, the desired signal and a reference signal are input into these Wiener filters to generate multiple mode subsignals with wideband noise suppressed. Meanwhile, multiscale noise interferences are separated into different scales as reference bases for mode denoising. Third, via evolutionary digital filters (EDFs), a refined reference filtering is performed within mode scale to obtain multiple intrinsic mode subsignals (IMSs), namely mode denoising. Finally, the Pearson correlation coefficients between two parts of each subsignal pair is employed as a weight to synthesize a denoised signal. In this manner, VEAMD effectively tackles an improper reference at local scales by a series of adaptive mode denoising, achieving more significant noise cancellation capability and an effect of weak feature enhancement, which is conducive for a timely and accurate early gearbox fault diagnosis. Simulation and experiment prove the superiority of VEAMD over other four methods.

Domain-adaptive denoising network for low-dose CT via noise estimation and transfer learning.

In recent years, low-dose computed tomography (LDCT) has played an important role in the diagnosis CT to reduce the potential adverse effects of X-ray radiation on patients, while maintaining the same diagnostic image quality. Deep learning (DL)-based methods have played an increasingly important role in the field of LDCT imaging. However, its performance is highly dependent on the consistency of feature distributions between training data and test data. Due to patient's breathing movements during data acquisition, the paired LDCT and normal dose CT images are difficult to obtain from realistic imaging scenarios. Moreover, LDCT images from simulation or clinical CT examination often have different feature distributions due to the pollution by different amounts and types of image noises. If a network model trained with a simulated dataset is used to directly test clinical patients' LDCT data, its denoising performance may be degraded. Based on this, we propose a novel domain-adaptive denoising network (DADN) via noise estimation and transfer learning to resolve the out-of-distribution problem in LDCT imaging. To overcome the previous adaptation issue, a novel network model consisting of a reconstruction network and a noise estimation network was designed. The noise estimation network based on a double branch structure is used for image noise extraction and adaptation. Meanwhile, the U-Net-based reconstruction network uses several spatially adaptive normalization modules to fuse multi-scale noise input. Moreover, to facilitate the adaptation of the proposed DADN network to new imaging scenarios, we set a two-stage network training plan. In the first stage, the public simulated dataset is used for training. In the second transfer training stage, we will continue to fine-tune the network model with a torso phantom dataset, while some parameters are frozen. The main reason using the two-stage training scheme is based on the fact that the feature distribution of image content from the public dataset is complex and diverse, whereas the feature distribution of noise pattern from the torso phantom dataset is closer to realistic imaging scenarios. In an evaluation study, the trained DADN model is applied to both the public and clinical patient LDCT datasets. Through the comparison of visual inspection and quantitative results, it is shown that the proposed DADN network model can perform well in terms of noise and artifact suppression, while effectively preserving image contrast and details. In this paper, we have proposed a new DL network to overcome the domain adaptation problem in LDCT image denoising. Moreover, the results demonstrate the feasibility and effectiveness of the application of our proposed DADN network model as a new DL-based LDCT image denoising method.

Dynamic Multiscale Feature Fusion Method for Underwater Target Recognition

The feature information of small-scale targets is seriously missing under the interference of complex underwater terrain and light refraction. Moreover, the unbalanced distribution of underwater target samples can also affect the accuracy of spatial semantic feature extraction. Aiming at the above problems, this paper proposes a dynamic multiscale feature fusion method for underwater target recognition. Firstly, this paper uses multiscale info noise contrastive estimation (MS-InfoNCE) loss to extract the significant features of the target at 4 scales. Secondly, the method learns the spatial semantic features of the target through a dynamic conditional probability matrix. Finally, this paper designs different feature fusion mechanisms for different scale targets, dynamically fusing multiscale significant features and spatial semantic features to recognize underwater weak targets. The experimental results show that the recognition accuracy of the proposed algorithm is 1.38% higher than that of the existing algorithm when recognizing underwater distorted targets.

Multi-scale attention mechanism residual neural network for fault diagnosis of rolling bearings

Rolling bearing fault diagnosis is crucial to improve industrial safety and reliability. In recent years, intelligent fault diagnosis method represented by deep learning (DL) has been receiving increasing attention. In order to ameliorate the full training of the deep network, improve the model accuracy, and perfect the analysis of mechanical vibration signals with huge amount of information, a multi-scale attention mechanism residual network (MSA-ResNet) fault diagnosis method is proposed in this paper. First, an attention mechanism block is introduced to construct a new type of residual block combination. Second, a multi-scale structure is constructed by choosing an appropriate convolution kernel size. Finally, the overall framework of MSA-ResNet is constructed for efficient training and failure pattern recognition. The MSA-ResNet algorithm introduces an attention mechanism in each residual module of the residual network (ResNet), which improves the sensitivity to features. The features of different scales are obtained through the multi-scale convolution kernel, and the multi-scale feature extraction of complex nonlinear mechanical vibration signals is realized. The processing of original vibration signal rarely involves artificial interference, which is more conducive to industrial application of the proposed method. Diagnostic experiments are conducted on bearing datasets from the Case Western Reserve University (CWRU) and the Machinery Failure Prevention Technology (MFPT) to verify the effectiveness of the proposed method. The results illustrating the rolling bearing fault diagnosis method based on MSA-ResNet have advantages in multi-scale feature extraction, noise immunity, and fault classification accuracy.

Single-Stage Adaptive Multi-Scale Point Cloud Noise Filtering Algorithm Based on Feature Information

This paper proposes a single-stage adaptive multi-scale noise filtering algorithm for point clouds, based on feature information, which aims to mitigate the fact that the current laser point cloud noise filtering algorithm has difficulty quickly completing the single-stage adaptive filtering of multi-scale noise. The feature information from each point of the point cloud is obtained based on the efficient k-dimensional (k-d) tree data structure and amended normal vector estimation methods, and the adaptive threshold is used to divide the point cloud into large-scale noise, a feature-rich region, and a flat region to reduce the computational time. The large-scale noise is removed directly, the feature-rich and flat regions are filtered via improved bilateral filtering algorithm and weighted average filtering algorithm based on grey relational analysis, respectively. Simulation results show that the proposed algorithm performs better than the state-of-art comparison algorithms. It was, thus, verified that the algorithm proposed in this paper can quickly and adaptively (i) filter out large-scale noise, (ii) smooth small-scale noise, and (iii) effectively maintain the geometric features of the point cloud. The developed algorithm provides research thought for filtering pre-processing methods applicable in 3D measurements, remote sensing, and target recognition based on point clouds.

Computational Breast Anatomy Simulation Using Multi-Scale Perlin Noise.

Virtual clinical trials (VCTs) of medical imaging require realistic models of human anatomy. For VCTs in breast imaging, a multi-scale Perlin noise method is proposed to simulate anatomical structures of breast tissue in the context of an ongoing breast phantom development effort. Four Perlin noise distributions were used to replace voxels representing the tissue compartments and Cooper's ligaments in the breast phantoms. Digital mammography and tomosynthesis projections were simulated using a clinical DBT system configuration. Power-spectrum analyses and higher-order statistics properties using Laplacian fractional entropy (LFE) of the parenchymal texture are presented. These objective measures were calculated in phantom and patient images using a sample of 140 clinical mammograms and 500 phantom images. Power-law exponents were calculated using the slope of the curve fitted in the low frequency [0.1, 1.0] mm-1 region of the power spectrum. The results show that the images simulated with our prior and proposed Perlin method have similar power-law spectra when compared with clinical mammograms. The power-law exponents calculated are -3.10, -3.55, and -3.46, for the log-power spectra of patient, prior phantom and proposed phantom images, respectively. The results also indicate an improved agreement between the mean LFE estimates of Perlin-noise based phantoms and patients than our prior phantoms and patients. Thus, the proposed method improved the simulation of anatomic noise substantially compared to our prior method, showing close agreement with breast parenchyma measures.

Multi-scale noise transfer and feature frequency detection in SSVEP based on FitzHugh–Nagumo neuron system

Objective. The steady-state visual evoked potential (SSVEP) is one of the most commonly used control signals for brain–computer interfaces (BCIs) due to its excellent interactive potential, such as high tolerance to noises and robust performance across users. In addition, it has a stable cycle, obvious characteristics and minimal training requirements. However, the SSVEP is extremely weak and companied with strong and multi-scale noise, resulting in a poor signal-to-noise ratio in practice. Common algorithms for classification are based on the principle of template matching and spatial filtering, which cannot obtain satisfied performance of SSVEP detection under the multi-scale noise. Therefore, using linear methods to extract SSVEP with obvious nonlinear and non-stationary characteristics, the useful signal will be attenuated or lost. Approach. To address this issue, two novel frameworks based on a two-dimensional nonlinear FitzHugh–Nagumo (FHN) neuron system are proposed to extract feature frequency of SSVEP. Results. In order to evaluate the effectiveness of the proposed methods, this research recruit 22 subjects to participate the experiment. Experimental results show that nonlinear FHN neuron model can force the energy of noise to be transferred into SSVEP and hence amplifying the amplitude of the target frequency. Compared with the traditional methods, the FHN and FHNCCA methods can achieve higher classification accuracy and faster processing speed, which effectively improves the information transmission rate of SSVEP-based BCI.

Waveform feature extraction and signal recovery in single-channel TVEP based on Fitzhugh–Nagumo stochastic resonance

Objective. Transient visual evoked potential (TVEP) can reflect the condition of the visual pathway and has been widely used in brain–computer interface. TVEP signals are typically obtained by averaging the time-locked brain responses across dozens or even hundreds of stimulations, in order to remove different kinds of interferences. However, this procedure increases the time needed to detect the brain status in realistic applications. Meanwhile, long repeated stimuli can vary the evoked potentials and discomfort the subjects. Therefore, a novel unsupervised framework was developed in this study to realize the fast extraction of single-channel TVEP signals with a high signal-to-noise ratio. Approach. Using the principle of nonlinear aperiodic FitzHugh–Nagumo (FHN) model, a fast extraction and signal restoration technology of TVEP waveform based on FHN stochastic resonance is proposed to achieve high-quality acquisition of signal features with less average times. Results: A synergistic effect produced by noise, aperiodic signal and nonlinear system can force the energy of noise to be transferred into TVEP and hence amplifying the useful P100 feature while suppressing multi-scale noise. Significance. Compared with the conventional average and average-singular spectrum analysis-independent component analysis(average-SSA-ICA) method, the average-FHN method has a shorter stimulation time which can greatly improve the comfort of patients in clinical TVEP detection and a better performance of TVEP waveform i.e. a higher accuracy of P100 latency. The FHN recovery method is not only highly correlated with the original signal, but also can better highlight the P100 amplitude, which has high clinical application value.

Deep learning-driven opportunistic spectrum access (OSA) framework for cognitive 5G and beyond 5G (B5G) networks

The evolving 5G and beyond 5G (B5G) wireless technologies are envisioned to provide ubiquitous connectivity and great heterogeneity in communication infrastructure by connecting diverse devices and providing multifarious services. Recently, the Internet of Things (IoT) and unmanned aerial vehicles (UAVs) are realized as an essential component of the upcoming 5G/B5G networks, enabling enhanced communication capacity, high reliability, low latency, and massive connectivity. However, one limiting factor in the expansion of 5G/B5G technology is the finite radio spectrum, which necessitates managing the anticipated spectrum crunch for future wireless networks. One potential solution is to develop intelligent cognitive methods to dynamically optimize the use of spectrum in 5G/B5G networks to solve the imminent problem of spectrum congestion and improve radio efficiency. This paper addresses the opportunistic spectrum access (OSA) problem in the 5G/B5G cognitive radio (CR) network of IoTs and UAVs through the novel deep learning-based detector, dubbed as Deep-CRNet. The proposed detector employs residual connections with cascaded multi-kernel convolutions to identify the primary user (PU) spectrum usage by extracting the inherent multi-scale signal and noise features in the sensed transmission patterns. Thereby, Deep-CRNet intelligently learns and locates the spectrum holes so that secondary users (SUs) and PUs can dynamically share network spectrum resources. The efficacy of Deep-CRNet is validated through simulation results, where it achieved 99.74% accuracy with 99.65% precision and 99.83% recall in accurately classifying the PU status. In addition, the average correct detection probability of Deep-CRNet in the low signal-to-noise ratio (−20 dB to −15 dB) range is 38.21% higher than the second best-performing detector.

Forgery Detection in Digital Images by Multi-Scale Noise Estimation.

A complex processing chain is applied from the moment a raw image is acquired until the final image is obtained. This process transforms the originally Poisson-distributed noise into a complex noise model. Noise inconsistency analysis is a rich source for forgery detection, as forged regions have likely undergone a different processing pipeline or out-camera processing. We propose a multi-scale approach, which is shown to be suitable for analyzing the highly correlated noise present in JPEG-compressed images. We estimate a noise curve for each image block, in each color channel and at each scale. We then compare each noise curve to its corresponding noise curve obtained from the whole image by counting the percentage of bins of the local noise curve that are below the global one. This procedure yields crucial detection cues since many forgeries create a local noise deficit. Our method is shown to be competitive with the state of the art. It outperforms all other methods when evaluated using the MCC score, or on forged regions large enough and for colorization attacks, regardless of the evaluation metric.