Noise Interruptions Research Articles

DTT-CGINet: A Dual Temporal Transformer Network with Multi-Scale Contour-Guided Graph Interaction for Change Detection

Deep learning has dramatically enhanced remote sensing change detection. However, existing neural network models often face challenges like false positives and missed detections due to factors like lighting changes, scale differences, and noise interruptions. Additionally, change detection results often fail to capture target contours accurately. To address these issues, we propose a novel transformer-based hybrid network. In this study, we analyze the structural relationship in bi-temporal images and introduce a cross-attention-based transformer to model this relationship. First, we use a tokenizer to express the high-level features of the bi-temporal image into several semantic tokens. Then, we use a dual temporal transformer (DTT) encoder to capture dense spatiotemporal contextual relationships among the tokens. The features extracted at the coarse scale are refined into finer details through the DTT decoder. Concurrently, we input the backbone’s low-level features into a contour-guided graph interaction module (CGIM) that utilizes joint attention to capture semantic relationships between object regions and the contour. Then, we use the feature pyramid decoder to integrate the multi-scale outputs of the CGIM. The convolutional block attention modules (CBAMs) employ channel and spatial attention to reweight feature maps. Finally, the classifier discriminates change pixels and generates the final change map of the difference feature map. Several experiments have demonstrated that our model shows significant advantages over other methods in terms of efficiency, accuracy, and visual effects.

Computational design of ultra-robust strain sensors for soft robot perception and autonomy

Compliant strain sensors are crucial for soft robots’ perception and autonomy. However, their deformable bodies and dynamic actuation pose challenges in predictive sensor manufacturing and long-term robustness. This necessitates accurate sensor modelling and well-controlled sensor structural changes under strain. Here, we present a computational sensor design featuring a programmed crack array within micro-crumples strategy. By controlling the user-defined structure, the sensing performance becomes highly tunable and can be accurately modelled by physical models. Moreover, they maintain robust responsiveness under various demanding conditions including noise interruptions (50% strain), intermittent cyclic loadings (100,000 cycles), and dynamic frequencies (0–23 Hz), satisfying soft robots of diverse scaling from macro to micro. Finally, machine intelligence is applied to a sensor-integrated origami robot, enabling robotic trajectory prediction (<4% error) and topographical altitude awareness (<10% error). This strategy holds promise for advancing soft robotic capabilities in exploration, rescue operations, and swarming behaviors in complex environments.

Synthetic thermal convolutional‐memory network for the lithium‐ion battery behaviour diagnosis against noise interruptions

AbstractIn order to meet the two global challenges of energy shortage and environmental pollution, various countries have begun to advocate the application of new energy equipment such as electric vehicles. This has also promoted the development of energy storage equipment and energy storage systems. With their high performance, lithium‐ion batteries are used in a wide range of electrical equipment. But the safety of lithium‐ion batteries depends on effective behaviour diagnosis. In order to better realise behaviour diagnosis, this paper combined the long and short‐term memory network (LSTM) with the temporal convolution network (TCN) for the first time and established a synthetic thermal convolutional‐memory network (STCMN) for lithium‐ion battery behaviour diagnosis against noise interruptions. In addition, a TCN‐LSTM alliance network structure is designed. The TCN‐LSTM alliance network is an effective architecture applied not only to the temperature prediction of Li‐ion batteries but also to the thermal diagnosis part. And these two parts finally constitute the thermal convolutional‐memory network. The experimental results show the network designed in this paper was able to improve Li‐ion battery behaviour detection.

WaveCNet: Wavelet Integrated CNNs to Suppress Aliasing Effect for Noise-Robust Image Classification.

Though widely used in image classification, convolutional neural networks (CNNs) are prone to noise interruptions, i.e. the CNN output can be drastically changed by small image noise. To improve the noise robustness, we try to integrate CNNs with wavelet by replacing the common down-sampling (max-pooling, strided-convolution, and average pooling) with discrete wavelet transform (DWT). We firstly propose general DWT and inverse DWT (IDWT) layers applicable to various orthogonal and biorthogonal discrete wavelets like Haar, Daubechies, and Cohen, etc., and then design wavelet integrated CNNs (WaveCNets) by integrating DWT into the commonly used CNNs (VGG, ResNets, and DenseNet). During the down-sampling, WaveCNets apply DWT to decompose the feature maps into the low-frequency and high-frequency components. Containing the main information including the basic object structures, the low-frequency component is transmitted into the following layers to generate robust high-level features. The high-frequency components are dropped to remove most of the data noises. The experimental results show that WaveCNets achieve higher accuracy on ImageNet than various vanilla CNNs. We have also tested the performance of WaveCNets on the noisy version of ImageNet, ImageNet-C and six adversarial attacks, the results suggest that the proposed DWT/IDWT layers could provide better noise-robustness and adversarial robustness. When applying WaveCNets as backbones, the performance of object detectors (i.e., faster R-CNN and RetinaNet) on COCO detection dataset are consistently improved. We believe that suppression of aliasing effect, i.e. separation of low frequency and high frequency information, is the main advantages of our approach. The code of our DWT/IDWT layer and different WaveCNets are available at https://github.com/CVI-SZU/WaveCNet.

RETRACTED ARTICLE: Huffman quantization approach for optimized EEG signal compression with transformation technique

The significance of the electroencephalography (EEG) signal is used to read the brain activity in the form of electrical patterns. EEG signals help to diagnose anomalies in the brain at the time of head injuries, epilepsy, seizures, brain tumor, dizziness and sleep deprivation. So such types of crucial signals should be transported in a secure method to avoid any data loss or to prevent noise interruptions which can lead to the misdetection of diseases. As the EEG signals are in higher-dimensional size, it should be compressed for effective transportation. In this research, a lossless compression method named as Huffman-based discrete cosine transform is implemented to transmit the EEG data efficiently. The discrete cosine transform and inverse discrete cosine transform are proposed here to increase the privacy of the data and reduce the complexity of the data. This paper mainly focuses on to get a high accuracy ratio in reconstructing the original data after compression and transportation without any losses in minimum computational time. The preprocessing and sampling are made at the initial stages to remove the noises and transmit the original data. The Huffman quantization method based on discrete cosine transform achieves high-performance metrics in terms of peak signal-to-noise ratio, quality score and compression ratio when compared with existing methods in various transformations of data.

Noise metrics analysis in schools near airports: A Brazilian case study

Abstract This study compares metrics for environmental noise diagnosis in schools at airport vicinity. The goal is to analyze and identify the most suitable criteria for scaling aircraft noise impact over schools, during landing and take-off operations. A Brazilian case study is conducted, based on the noise mapping and sound level verification. The day-night average noise level (DNL) and the time above limit (TA) are investigated using acoustic simulation and noise mapping and in order to identify the critical receivers. Results of DNL and TA for two schools at airport surroundings show that the criteria adopted by the municipal and airport authorities to describe the airport noise are unsatisfactory and do not reflect the intermittent behavior of this type of noise. It was verified that individual receiver analysis, based on noise interruptions thought TA parameter is more suitable for evaluation of noise impact over schools at airport vicinity.

Exploring Vertical Wilderness in the Acoustic Environment

Hearing sounds of nature is an important motivation for visitors to National Parks, such as Grand Teton National Park (GRTE; Newman et al. 2015). Furthermore, managers are required to provide park visitors with an enjoyable soundscape experience. In 2006, Pilcher and Newman conducted a study on visitor perceptions of soundscapes in highly trafficked locations in GRTE, the Jenny Lake boat dock and Inspiration Point. While this study used similar methods, it aimed to better understand the influence of soundscapes to a unique visitor group -- climbers on the Grand Teton. This iconic climbing destination is located in an area that is potentially susceptible to anthropogenic or human caused noise interruptions because of its proximity to an airport and heavily used highways. In the summer of 2015 researchers from Penn State University used a combination of qualitative interviews and listening exercises with climbers to identify sounds that were being heard during their climbing experience and their emotions related to those sounds. These data provide managers with information about sounds that could be prioritized when managing for an optimal soundscape experience.

Effects of the Level of Noise Interruptions on Temporary Threshold Shift

Intermittency is a noise exposure variable which must be considered in proposing standard limits. The Committee on Healing, Bioacoustics, and Biomechanics (CHABA) defined a noise interruption as a period in the noise exposure when the level falls below 80 dBA for more than five minutes or for “20% of the duration of the preceding noise burst.” Our previous study indicated that interruption levels below 80 dBA could have variable effects on resultant temporary threshold shifts. Thirty subjects, ten for each of the three conditions, were exposed to 45 minutes of broadband noise (103 dBA). The noise bursts were interrupted twice for five minutes by levels of 57 dBA, 67 dBA and 77 dBA. Corrected TTS2 measurements and uncorrected TTS30 measurements were taken. Subjects afforded the 67 dBA and 77 dBA interruptions (1) had significantly more TTS2 and TTS30 at 4kHz and 6kHz than those subjects afforded the 57 dBA interruptions, (2) demonstrated a slower recovery rate than the 57 dBA group, and (3) experienced sev...