Object detection based on unmanned aerial vehicle(UAV) platforms is essential for both engineering and research. Complex scale problems in UAV application scenarios require strong regression localization capabilities from target algorithms. Nonetheless, due to the constraints of platform, it difficult increase accuracy by deepening network. Therefore, this paper presents an improved YOLOv5 with...

Shot boundary detection (SBD) is widely used in scene segmentation, semantic analysis, and video retrieval. However, existing SBD algorithms have certain applications processing, but they the following three problems. First, these cannot effectively handle shot boundaries caused by sudden lighting changes. Second, when there are dimly frames video, perform well. Third, object or camera motion a...

Sentiment analysis of news headlines is an important factor that investors consider when making investing decisions. We claim the sentiment financial impacts stock market values. Hence headline data are collected along with investment for a period time. Using Valence Aware Dictionary and Reasoning (VADER) analysis, correlation between values sentiments in established. In our experiments, on pri...

In recent years, with the rapid development of deep learning, super-resolution methods based on convolutional neural networks (CNNs) have made great progress. However, parameters and required consumption computing resources these are also increasing to point that such difficult implement devices low power. To address this issue, we propose a lightweight single image network an expectation-maxim...

Aiming at the problems of low detection accuracy and large computing resource consumption existing Unmanned Aerial Vehicle (UAV) algorithms for anti-UAV, this paper proposes a lightweight UAV swarm method based on You Only Look Once Version X (YOLOX). This uses depthwise separable convolution to simplify optimize network, greatly simplifies total parameters, while is only partially reduced. Mea...

In recent years, convolutional neural networks (CNNs) have been widely used in hyperspectral image (HSI) classification. However, feature extraction on data still faces numerous challenges. Existing methods cannot extract spatial and spectral-channel contextual information a targeted manner. this paper, we propose an encoder–decoder network that fuses attention for HSI classification from three...