Plasma-Induced Nanocrystalline Domain Engineering and Surface Passivation in Mesoporous Chalcogenide Semiconductor Thin Films.

The synthesis of highly crystalline mesoporous materials is key to realizing high-performance chemical and biological sensors and optoelectronics. However, minimizing surface oxidation and enhancing the domain size without affecting the porous nanoarchitecture are daunting challenges. Herein, we report a hybrid technique that combines bottom-up electrochemical growth with top-down plasma treatment to produce mesoporous semiconductors with large crystalline domain sizes and excellent surface passivation. By passivating unsaturated bonds without incorporating any chemical or physical layers, these films show better stability and enhancement in the optoelectronic properties of mesoporous copper telluride (CuTe) with different pore diameters. These results provide exciting opportunities for the development of long-term, stable, and high-performance mesoporous semiconductor materials for future technologies.

FFENet: frequency-spatial feature enhancement network for clothing classification.

Clothing analysis has garnered significant attention, and within this field, clothing classification plays a vital role as one of the fundamental technologies. Due to the inherent complexity of clothing scenes in real-world environments, the learning of clothing features in such complex scenes often encounters interference. Because clothing classification relies on the contour and texture information of clothing, clothing classification in real scenes may lead to poor classification results. Therefore, this paper proposes a clothing classification network based on frequency-spatial domain conversion. The proposed network combines frequency domain information with spatial information and does not compress channels. It aims to enhance the extraction of clothing features and improve the accuracy of clothing classification. In our work, (1) we combine the frequency domain information and spatial information to establish a clothing feature extraction clothing classification network without compressed feature map channels, (2) we use the frequency domain feature enhancement module to realize the preliminary extraction of clothing features, and (3) we introduce a clothing dataset in complex scenes (Clothing-8). Our network achieves a top-1 model accuracy of 93.4% on the Clothing-8 dataset and 94.62% on the Fashion-MNIST dataset. Additionally, it also achieves the best results in terms of top-3 and top-5 metrics on the DeepFashion dataset.

FDE-net: Frequency-domain enhancement network using dynamic-scale dilated convolution for thyroid nodule segmentation.

Thyroid nodules, a common disease of endocrine system, have a probability of nearly 10% to turn into malignant nodules and thus pose a serious threat to health. Automatic segmentation of thyroid nodules is of great importance for clinicopathological diagnosis. This work proposes FDE-Net, a combined segmental frequency domain enhancement and dynamic scale cavity convolutional network for thyroid nodule segmentation. In FDE-Net, traditional image omics method is introduced to enhance the feature image in the segmented frequency domain. Such an approach reduces the influence of noise and strengthens the detail and contour information of the image. The proposed method introduces a cascade cross-scale attention module, which addresses the insensitivity of the network to the change in target scale by fusing the features of different receptive fields and improves the ability of the network to identify multiscale target regions. It repeatedly uses the high-dimensional feature image to improve segmentation accuracy in accordance with the simple structure of thyroid nodules. In this study, 1355 ultrasound images are used for training and testing. Quantitative evaluation results showed that the Dice coefficient of FDE-Net in thyroid nodule segmentation was 83.54%, which is better than other methods. Therefore, FDE-Net can enable the accurate and rapid segmentation of thyroid nodules.

Self-attending RNN for Speech Enhancement to Improve Cross-corpus Generalization.

Deep neural networks (DNNs) represent the mainstream methodology for supervised speech enhancement, primarily due to their capability to model complex functions using hierarchical representations. However, a recent study revealed that DNNs trained on a single corpus fail to generalize to untrained corpora, especially in low signal-to-noise ratio (SNR) conditions. Developing a noise, speaker, and corpus independent speech enhancement algorithm is essential for real-world applications. In this study, we propose a self-attending recurrent neural network (SARNN) for time-domain speech enhancement to improve cross-corpus generalization. SARNN comprises of recurrent neural networks (RNNs) augmented with self-attention blocks and feedforward blocks. We evaluate SARNN on different corpora with nonstationary noises in low SNR conditions. Experimental results demonstrate that SARNN substantially outperforms competitive approaches to time-domain speech enhancement, such as RNNs and dual-path SARNNs. Additionally, we report an important finding that the two popular approaches to speech enhancement: complex spectral mapping and time-domain enhancement, obtain similar results for RNN and SARNN with large-scale training. We also provide a challenging subset of the test set used in this study for evaluating future algorithms and facilitating direct comparisons.

Dense CNN with Self-Attention for Time-Domain Speech Enhancement.

Speech enhancement in the time domain is becoming increasingly popular in recent years, due to its capability to jointly enhance both the magnitude and the phase of speech. In this work, we propose a dense convolutional network (DCN) with self-attention for speech enhancement in the time domain. DCN is an encoder and decoder based architecture with skip connections. Each layer in the encoder and the decoder comprises a dense block and an attention module. Dense blocks and attention modules help in feature extraction using a combination of feature reuse, increased network depth, and maximum context aggregation. Furthermore, we reveal previously unknown problems with a loss based on the spectral magnitude of enhanced speech. To alleviate these problems, we propose a novel loss based on magnitudes of enhanced speech and a predicted noise. Even though the proposed loss is based on magnitudes only, a constraint imposed by noise prediction ensures that the loss enhances both magnitude and phase. Experimental results demonstrate that DCN trained with the proposed loss substantially outperforms other state-of-the-art approaches to causal and non-causal speech enhancement.

A New Framework for CNN-Based Speech Enhancement in the Time Domain.

This paper proposes a new learning mechanism for a fully convolutional neural network (CNN) to address speech enhancement in the time domain. The CNN takes as input the time frames of noisy utterance and outputs the time frames of the enhanced utterance. At the training time, we add an extra operation that converts the time domain to the frequency domain. This conversion corresponds to simple matrix multiplication, and is hence differentiable implying that a frequency domain loss can be used for training in the time domain. We use mean absolute error loss between the enhanced short-time Fourier transform (STFT) magnitude and the clean STFT magnitude to train the CNN. This way, the model can exploit the domain knowledge of converting a signal to the frequency domain for analysis. Moreover, this approach avoids the well-known invalid STFT problem since the proposed CNN operates in the time domain. Experimental results demonstrate that the proposed method substantially outperforms the other methods of speech enhancement. The proposed method is easy to implement and applicable to related speech processing tasks that require time-frequency masking or spectral mapping.