Tian Deng

Conventional photoelectric smoke detection relies on light scattering by smoke particles and is therefore sensitive to variations in optical properties such as refractive index, resulting in markedly different responses to smoldering white smoke and flaming black smoke. To address this limitation, a charge-based smoke sensing method combining needle-cylinder corona charging with Faraday cup current measurement is proposed. By integrating a particle-charging model with moment-based characterization of the particle size distribution, a quantitative relationship is established among particle charge, sensor output, and the first moment of the particle size distribution, M1. Experiments showed that the optimized charging module maintained a charging efficiency above 68.2% for 50–1000 nm DEHS particles. The sensor output exhibited strong linear relationships with M₁ for monodisperse DEHS aerosols and four standard test-fire smoke samples, with R2 values of 0.9983 and 0.9767, respectively. The retrieved M1 values agreed well with the SMPS measurements, with R2 of 0.958 and a mean absolute relative error of 4.7%. At comparable smoke extinction levels, the response of conventional photoelectric smoke detectors to n-heptane flaming black smoke was only approximately one tenth of that to cotton-wick smoldering white smoke, whereas the charge-based method exhibited substantially more balanced responses. These results demonstrate that the proposed method reduces the influence of particle optical properties on the measured response and provides an alternative electrical approach for quantitative detection of early-stage fire smoke.

Debris flows pose a significant global geohazard, causing a large number of deaths and infrastructure damage every year. Effective protection and land-use planning in the affected regions requires understanding susceptibility to these events. Although a global phenomenon, previous studies have focused extensively on local areas with specialized models and accordingly complex feature selections. In this study, we investigate whether a unified debris-flow susceptibility prediction paradigm can be achieved regardless of regional scale, using only very few global public remote sensing data sources. To this end, this work contributes in the following ways: (1) A novel two-step negative sample generation scheme is proposed, and two open debris-flow datasets are constructed based on global debris-flow locations and locations in Sichuan, China. (2) An open-source end-to-end machine learning platform using remote sensing features directly is proposed, which achieves state-of-the-art results with 0.947 and 0.957 AUC in both scales compared to 0.88 for previous methods on the same location data, while using far fewer features. (3) A comparative feature importance analysis shows that, given the significant feature distribution difference on global vs local datasets, alleviating the scale-level gap is possible by leveraging the advanced deep learning technologies. This allows our unified framework to be easily applied to any regional study of debris-flow susceptibility prediction.

Study of early Chinese ceramics is crucial for understanding cultural, economic, and technological developments in Chinese history. With the evolving deep learning techniques, one urgent question would be, whether we can identify early Chinese ceramics by a simple 2D image without further domain knowledge. This work collected a highly diverse dataset for ancient Chinese ceramics from 15 dynasties, with 4 representative glaze colors and 15 shape types. We studied the performance of five state-of-the-art neural networks on two identification tasks: ceramic visual feature recognition and early Chinese ceramic dating. A class-imbalance learning strategy is designed to improve the models’ performance on multi-label tasks. To the best of our knowledge, our work is the first to introduce deep learning into early Chinese ceramic recognition on a large scale. Experiments prove that deep learning can recognize visual features like glaze and most shape types with high accuracy, while ceramic dating is feasible for the main dynasties but remains challenging along the overall history. Further quantitative assessment shows that cultural inheritance and artistic continuity can lead to reasonable false dating by classifying ceramics into adjacent dynasties or periods. Moreover, although domain knowledge is required for interpretation, deep learning shows great potential in recognizing even unlabeled time-relevant features, which can help study the inheritance and evolution of early Chinese ceramic development.

Lithium-ion batteries (LIBs) are widely used in industrial applications for energy storage and power supply. The failure of LIBs has attracted significant attention since severe accidents caused by thermal runaway. The electrolyte leakage in LIBs represents a significant hidden risk to LIB systems and serves as an early indicator for potential thermal runaway. Current electrolyte leakage detection technologies often lack the sensitivity, cost-effectiveness, and longevity required for modern battery safety systems. To address this challenge, we investigate the phase transitions of leaked electrolytes and changes in particle size of ambient aerosols and propose a pack-level optical electrolyte leakage sensor for early safety warning of LIBs failures in Industrial Internet of Things (IIoTs) applications. Experiments demonstrate that the particle size of ambient aerosol increases rapidly due to the phase change condensation of the leaked electrolyte vapor. Correspondingly, the light scattering efficiency of ambient aerosol increases, providing measurable indicators for electrolyte leakage monitoring. Our optical sensor can promptly detect leakage in the overcharging test, slightly surpassing the parts-per-billion (PPB) gas sensor in response time, while reducing cost to 1/10 due to mature commercial optical components and increasing lifespan three times according to the lifespan of core elements. Utilizing readily available commercial components, this method offers a high sensitivity, cost-effective, and durable solution for improving safety in LIB systems.

The concentration of lung deposition surface area (LDSA) is a crucial metric for assessing potential health impacts associated with particle size distribution. Traditional LDSA sensors, which rely on particle charging, are limited to measuring particles up to 300 nm in size. However, aerosols in the submicron to micron range are also significant for evaluating exposure risks. In this study, we introduce an optical method for LDSA sensing that leverages light-scattering angular spectrum and deep learning techniques. Our approach utilizes a compact, lens-free sensor to sample angular spectrum signals, which are then processed by a ResNet-like neural network to adjust the conversion factor between LDSA concentration and scattering light intensity. Experimental results demonstrate that our prototype achieves a maximum relative error of 14.78% in measuring the angular spectrum. Additionally, the neural network compensates for variations in particle size distribution and refractive index, refining the conversion factor. When compared to reference LDSA concentrations provided by laboratory instruments, our prototype maintains relative errors within 17.66%. With its simple mechanical design, high accuracy, and cost-effectiveness, this method presents a promising solution for online and field measurements of LDSA concentration.

Due to its wide monitoring range and low cost, visual-based fire detection technology is commonly used for fire detection in open spaces. However, traditional fire detection algorithms have limitations in terms of accuracy and speed, making it challenging to detect fires in real time. These algorithms have poor anti-interference ability against fire-like objects, such as emissions from factory chimneys, clouds, etc. In this study, we developed a fire detection approach based on an improved YOLOv5 algorithm and a fire detection dataset with fire-like objects. We added three Convolutional Block Attention Modules (CBAMs) to the head network of YOLOv5 to improve its feature extraction ability. Meanwhile, we used the C2f module to replace the original C2 module to capture rich gradient flow information. Our experimental results show that the proposed algorithm achieved a mAP@50 of 82.36% for fire detection. In addition, we also conducted a comparison test between datasets with and without labeling information for fire-like objects. Our results show that labeling information significantly reduced the false-positive detection proportion of fire-like objects incorrectly detected as fire objects. Our experimental results show that the CBAM and C2f modules enhanced the network’s feature extraction ability to differentiate fire objects from fire-like objects. Hence, our approach has the potential to improve fire detection accuracy, reduce false alarms, and be more cost-effective than traditional fire detection methods. This method can be applied to camera monitoring systems for automatic fire detection with resistance to fire-like objects.

Current optical fire smoke detectors use scattering light intensity as an indicator of smoke concentration and trigger fire alarms when the intensity exceeds a threshold value. However, such detectors are prone to false alarms caused by non-fire aerosols since both fire smokes and non-fire aerosols scatter light. Thus, in order to reduce false alarms caused by non-fire aerosols such as dust and water vapor, fire detectors must be capable of distinguishing fire smoke from non-fire aerosols. Since the light scattering signals depend on the particle size information of aerosols, it is essential to study and characterize the particle size distribution of fire smoke and non-fire aerosols for differentiating them. In this paper, a comprehensive aerosol experimental platform is built to measure the particle size distribution of various typical fire smokes and non-fire aerosols. Through the conducted experiments, we note that there are significant differences in the particle size distributions of typical fire smokes and non-fire aerosols, with a boundary of about 1μm. The experimental results provide fundamental data support of the particle size distribution for developing a better fire detector that accurately identifies smoke.

The severe doxorubicin (DOXO) side effect of cardiomyopathy limits it clinical application as an effective anticancer drug. Although Ca²⁺ overload was postulated as one of the mechanisms for this toxicity, its role was, however, disputable in terms of the contractile dysfunction. In this work, the dynamics of the intracellular Ca²⁺ signal were optically mapped in a Langendorff guinea pig heart. We found that DOXO treatment: (1) Delayed the activation of the Ca²⁺ signal. With the reference time set at the peak of the action potential (AP), the time lag between the peak of the Ca²⁺ signal and AP (Ca-AP-Lag) was significantly prolonged. (2) Slowed down the intracellular Ca²⁺ releasing and sequestering process. Both the maximum rising (MRV) and falling (MFV) velocity of the Ca²⁺ signal were decreased. (3) Shortened the duration of the Ca²⁺ signal in one cycle of Ca²⁺ oscillation. The duration of the Ca²⁺ signal at 50% amplitude (CaD50) was significantly shortened. These results suggested a reduced level of intracellular Ca²⁺ after DOXO treatment. Furthermore, we found that the effect of tachypacing was similar to that of DOXO, and, interestingly, DOXO exerted contradictory effects on the tachypaced hearts: it shortened the Ca-AP-Lag, accelerated the MRV and MFV, and prolonged the CaD50. We, therefore, concluded that DOXO had a different effect on intracellular Ca²⁺. It caused Ca²⁺ underload in hearts with sinus rhythm; this might relate to the contractile dysfunction in DOXO cardiomyopathy. It led to Ca²⁺ overload in the tachypaced hearts, which might contribute to the Ca²⁺-overload-related toxicity.