A Multi-Scale Natural Scene Text Detection Method Based on Attention Feature Extraction and Cascade Feature Fusion.

Scene text detection is an important research field in computer vision, playing a crucial role in various application scenarios. However, existing scene text detection methods often fail to achieve satisfactory results when faced with text instances of different sizes, shapes, and complex backgrounds. To address the challenge of detecting diverse texts in natural scenes, this paper proposes a multi-scale natural scene text detection method based on attention feature extraction and cascaded feature fusion. This method combines global and local attention through an improved attention feature fusion module (DSAF) to capture text features of different scales, enhancing the network's perception of text regions and improving its feature extraction capabilities. Simultaneously, an improved cascaded feature fusion module (PFFM) is used to fully integrate the extracted feature maps, expanding the receptive field of features and enriching the expressive ability of the feature maps. Finally, to address the cascaded feature maps, a lightweight subspace attention module (SAM) is introduced to partition the concatenated feature maps into several sub-space feature maps, facilitating spatial information interaction among features of different scales. In this paper, comparative experiments are conducted on the ICDAR2015, Total-Text, and MSRA-TD500 datasets, and comparisons are made with some existing scene text detection methods. The results show that the proposed method achieves good performance in terms of accuracy, recall, and F-score, thus verifying its effectiveness and practicality.

Enhanced Hybrid Vision Transformer with Multi-Scale Feature Integration and Patch Dropping for Facial Expression Recognition.

Convolutional neural networks (CNNs) have made significant progress in the field of facial expression recognition (FER). However, due to challenges such as occlusion, lighting variations, and changes in head pose, facial expression recognition in real-world environments remains highly challenging. At the same time, methods solely based on CNN heavily rely on local spatial features, lack global information, and struggle to balance the relationship between computational complexity and recognition accuracy. Consequently, the CNN-based models still fall short in their ability to address FER adequately. To address these issues, we propose a lightweight facial expression recognition method based on a hybrid vision transformer. This method captures multi-scale facial features through an improved attention module, achieving richer feature integration, enhancing the network's perception of key facial expression regions, and improving feature extraction capabilities. Additionally, to further enhance the model's performance, we have designed the patch dropping (PD) module. This module aims to emulate the attention allocation mechanism of the human visual system for local features, guiding the network to focus on the most discriminative features, reducing the influence of irrelevant features, and intuitively lowering computational costs. Extensive experiments demonstrate that our approach significantly outperforms other methods, achieving an accuracy of 86.51% on RAF-DB and nearly 70% on FER2013, with a model size of only 3.64 MB. These results demonstrate that our method provides a new perspective for the field of facial expression recognition.

Correction to: High sensation seeking is associated with behavioral and neural insensitivity to increased negative outcomes during decision-making under uncertainty.

The original article unfortunately had "?" in place of "훥 "on few lines and those are corrected below.

High sensation seeking is associated with behavioral and neural insensitivity to increased negative outcomes during decision-making under uncertainty.

High-level sensation seeking (HSS) has been linked to a range of risky and unhealthy behavior; however, the neural mechanisms underlying such linkage remain unclear. In the present study, we used event-related potential (ERP) with a Balloon Analogue Risk Task (BART) to investigate how sensation seeking modulates brain responses to sequential decision-making with variable reward/loss outcome magnitudes. Behavior data showed that decision-making behavior was significantly affected by the large compared with the small magnitude of monetary outcome in the BART for individuals with low-level sensation seeking (LSS), but not for individuals with HSS. Similarly, HSS individuals displayed no changes in feedback-related negativity (FRN) in response to negative outcomes from decision-making with large or small reward/loss magnitudes, whereas LSS individuals showed greater FRN in response to decision-making with large loss magnitude than those with small loss magnitude. In addition, FRN amplitude differences correlated with decision-making behavior changes from small to large outcome magnitude for LSS, while such correlation was not observed for HSS. These findings suggest that a high-level of sensation seeking is associated with behavioral and neural insensitivity to increased negative outcome during decision-making under uncertainty, which may lead to greater risky behavior in these individuals when facing potential loss.

Modification of nanocrystalline TiO2 coatings with molecularly imprinted TiO2 for uric acid recognition.

Combining the surface modification and molecular imprinting technique, a novel piezoelectric sensing platform with excellent molecular recognition capability was established for the detection of uric acid (UA) based on the immobilization of TiO2 nanoparticles onto quartz crystal microbalance (QCM) electrode and modification of molecularly imprinted TiO2 (MIT) layer on TiO2 nanoparticles. The performance of the fabricated biosensor was evaluated, and the results indicated that the biosensor exhibited high sensitivity in UA detection, with a linear range from 0.04 to 45 µM and a limit of detection of 0.01 µM. Moreover, the biosensor presented high selectivity towards UA in comparison with other interferents. The analytical application of the UA biosensor confirmed the feasibility of UA detection in urine sample.

Influence of High-Pressure Water Injection on the Pore Structure of Anthracite Coal in Xinjing Coal Mine.

High-pressure water injection, as an important measure for coal and gas outburst prevention, is still under-researched, especially its mechanism on the coal pore structure. The anthracite samples taken from no. 3 coal seam in Xinjing coal mine were dried and injected with high-pressure water, after which their pore characteristics were studied by using mercury porosimetry (MP) and low-pressure N2 gas adsorption (LP-N2GA). The results of MP showed that after the water was injected into the coal samples, the pore volume and the pore size of samples increased, but the specific surface area (SSA) remained almost unchanged. It could be concluded from LP-N2GA experiments that after the high-pressure water injection, the SSA of coal samples reduced greatly, but their pore size increased significantly. Through detailed analysis, the mechanism of high-pressure water injection on the coal pore structure is described as follows: the pores within the samples fracture after high-pressure water injection and the diameter of pores becomes bigger, resulting in increases in both the pore volume and the pore size. In addition, water molecules injected will stay at the end of micropores, so there is almost no change in the SSA, as indicated by MP testing results. However, the SSA of coal samples decreased significantly in the LP-N2GA testing. This is because it is really difficult to evaporate water molecules staying in the micropores by heating because of the strong interaction between water and coal. This study is helpful to further understand the mechanism of high-pressure water injection on preventing coal and gas outburst at the microlevel.

Hypothetical Versus Real Monetary Reward Decrease the Behavioral and Affective Effects in the Balloon Analogue Risk Task.

A critical question is whether the same decision-making processes underlie task performance with hypothetical and real money as rewards. Across two studies, we administered the Balloon Analogue Risk Task to healthy young adults under these two reward conditions. We found that participants displayed greater risk aversion during trials immediately after the balloon exploded in the previous trial in case the reward was real money, than if the reward was hypothetical money and exhibited greater subjective ratings of regret following losing trials. Moreover, subjective regret ratings after the balloon exploded in the previous trial with real money correlated with risk-taking behavior changes in the current trial, whereas we did not observe this correlation with hypothetical monetary rewards. In addition, when we manipulated the real money amounts to be large or small, participants were more risk averse in the large real money condition compared to the real money amount, whereas we did not observe these differences with varying amounts of hypothetical money.

[Broad excitation band alkaline-earth silicate luminescent materials activated by rare earth and its applications].

Series of novel broad excitation band phosphors M2 MgSis O7 : Eu, Dy(M = Ca, Sr) were prepared by a high temperature solid-state reaction method. The crystal structure of compound was characterized. And the effects of part substitution of alkaline-earth on crystal structure, photoluminescence spectra and luminescence properties were also investigated. It is found that the excitation band of silicate luminescent materials extend to visible region and they exhibit yellow, green and blue long after-glow luminescence after excited by ultraviolet or visible light. Ca MgSi O7 : Eu, Dy luminescent materials can be excited effectively under the 450-480 nm range and exhibit a strong emission at 536 nm, nicely combining with blue light emitted by InGaN chips to produce white light. This promises the silicate luminescent materials a potential yellow phosphor for white LED.