Floods are one of the most often occurring and damaging natural hazards. They impact the society on a massive scale and result in significant damages. To reduce the impact of floods, society needs to keep benefiting from the latest technological innovations. Drones equipped with sensors and latest algorithms (e.g., computer vision and deep learning) have emerged as a potential platform which may be useful for flood monitoring, mapping and detection activities in a more efficient way than current practice. To better understand the scope and recent trends in the domain of drones for flood management, we performed a detailed bibliometric analysis. The intent of performing the bibliometric analysis waws to highlight the important research trends, co-occurrence relationships and patterns to inform the new researchers in this domain. The bibliometric analysis was performed in terms of performance analysis (i.e., publication statistics, citations statistics, top publishing countries, top publishing journals, top publishing institutions, top publishers and top Web of Science (WoS) categories) and science mapping (i.e., citations by country, citations by journals, keyword co-occurrences, co-authorship, co-citations and bibliographic coupling) for a total of 569 records extracted from WoS for the duration 2000–2022. The VOSviewer open source tool has been used for generating the bibliographic network maps. Subjective discussions of the results explain the obtained trends from the bibliometric analysis. In the end, a detailed review of top 28 most recent publications was performed and subjected to process-driven analysis in the context of flood management. The potential active areas of research were also identified for future research in regard to the use of drones for flood monitoring, mapping and detection activities.

This paper proposes a framework for unmixing of hyperspectral data that is based on utilizing the scattering transform to extract deep features that are then used within a neural network. Previous research has shown that using the scattering transform combined with a traditional K-nearest neighbors classifier (STFHU) is able to achieve more accurate unmixing results compared to a convolutional neural network (CNN) applied directly to the hyperspectral images. This paper further explores hyperspectral unmixing in limited training data scenarios, which are likely to occur in practical applications where the access to large amounts of labeled training data is not possible. Here, it is proposed to combine the scattering transform with the attention-based residual neural network (ResNet). Experimental results on three HSI datasets demonstrate that this approach provides at least 40% higher unmixing accuracy compared to the previous STFHU and CNN algorithms when using limited training data, ranging from 5% to 30%, are available. The use of the scattering transform for deriving features within the ResNet unmixing system also leads more than 25% improvement when unmixing hyperspectral data contaminated by additive noise.

Virtual reality (VR) is increasingly important for exploring the real world, which has partially moved to virtual workplaces. In order to create immersive presence in a simulated scene for humans, VR needs to reproduce spatial audio that describes three-dimensional acoustic characteristics in the counterpart physical environment. When the user moves, this reproduction should be dynamically updated, which provides practical challenges because the bandwidth for continuously transmitting audio and video scene data may be limited. This paper proposes an interpolation approach for dynamic spatial audio reproduction using acoustic characteristics of direction and reverberation at limited numbers of positions, which are represented using a first order Ambisonics encoding of the room impulse response (RIR), called the directional RIR (DRIR). We decompose two known DRIRs into reflection components, before interpolating early dominant components for DRIR synthesis and utilizing DRIR recordings for accuracy evaluation. Results indicate that the most accurate interpolation is obtained by the proposed method over two comparative approaches, particularly in a simulated small room where most direction of arrival estimation errors of early components are below five degrees. These findings suggest precise interpolated DRIRs with limited data using the proposed approach, which is vital for dynamic spatial audio reproduction for VR applications.

Campus security incidents occur from time to time, which seriously affect the public security. In recent years, the rapid development of artificial intelligence has brought technical support for campus intelligent security. In order to quickly recognize and locate dangerous targets on campus, an improved YOLOv3‐Tiny model is proposed for dangerous target detection. Since the biggest advantage of this model is that it can achieve higher precision with very fewer parameters than YOLOv3‐Tiny, it is one of the Tinier‐YOLO models. In this paper, the dangerous targets include dangerous objects and dangerous actions. The main contributions of this work include the following: firstly, the detection of dangerous objects and dangerous actions is integrated into one model, and the model can achieve higher accuracy with fewer parameters. Secondly, to solve the problem of insufficient YOLOv3‐Tiny target detection, a jump‐join repetitious learning (JRL) structure is proposed, combined with the spatial pyramid pooling (SPP), which serves as the new backbone network of YOLOv3‐Tiny and can accelerate the speed of feature extraction while integrating features of different scales. Finally, the soft‐NMS and DIoU‐NMS algorithm are combined to effectively reduce the missing detection when two targets are too close. Experimental tests on self‐made datasets of dangerous targets show that the average MAP value of the JRL‐YOLO algorithm is 85.03%, which increases by 3.22 percent compared with YOLOv3‐Tiny. On the VOC2007 dataset, the proposed method has a 9.29 percent increase in detection accuracy compared to that using YOLOv3‐Tiny and a 2.38 percent increase compared to that employing YOLOv4‐Tiny, respectively. These results all evidence the great improvement in detection accuracy brought by the proposed method. Moreover, when testing the dataset of dangerous targets, the model size of JRL‐YOLO is 5.84 M, which is about one‐fifth of the size of YOLOv3‐Tiny (33.1 M) and one‐third of the size of YOLOv4‐Tiny (22.4 M), separately.

Understanding technology-enhanced learning (TEL) in schools continues to be difficult. A key reason is the complexity of digital technology use in the classroom to support learning. One of the reasons for this is the difficulty observing classrooms for extended periods of time to capture learning and the relevance to teacher's practice. This paper demonstrates how new technologies, data mining and machine learning approaches can be employed to explore the natural TEL classroom over time and meaningfully visualize results for teachers. To do this, a low-disturbance classroom observation kit is used to collect data in a secondary Science classroom for two months. One learning topic is identified, multimodal data are analysed and presented to the teacher for reflection. Three audio patterns relating to classroom activities and two teacher behaviours are identified, which bear relation to pedagogy, digital technology use and teaching strategies. Implications future research are discussed.

This paper proposes the co-prime circular microphone array (CPCMA), which applies co-prime array theory to circular microphone arrays. Compared with a conventional uniform circular array (UCA) with the same number of microphones and radius, the CPCMA achieves beampatterns with a narrower main lobe and fewer side lobes, whilst also avoiding spatial aliasing, thus having no grating lobes above the Nyquist frequency of the UCA. Array gain (AG) results indicate that the CPCMA is better than the UCA at amplifying the desired target signal while suppressing noise from other directions for typical speech signal frequencies. Compared with a UCA with similar performance and the same spatial Nyquist frequency, the CPCMA significantly reduces the required number of microphones. Simulations also indicate advantages of the CPCMA in speech source DOA estimation under high noise and low reverberation, especially when there are simultaneous multiple sources.