Small UAV target detection plays an important role in maintaining the security of cities and citizens. UAV targets have the characteristics of low-altitude flights, slow speeds, and miniaturization. Taking these characteristics into account, we present a real-time UAV target detection algorithm called Fast-YOLOv4 based on edge computing. By adopting Fast-YOLOv4 in the edge computing platform NVIDIA Jetson Nano, intelligent analysis can be performed on the video to realize the fast detection of UAV targets. However, the current iteration of the edge-embedded detection algorithm has low accuracy and poor real-time performance. To solve these problems, this paper introduces the lightweight networks MobileNetV3, Multiscale-PANet, and soft-merge to improve YOLOv4, thus obtaining the Fast-YOLOv4 model. The backbone of the model uses depth-wise separable convolution and an inverse residual structure to simplify the network’s structure and to improve its detection speed. The neck of the model adds a scale fusion branch to improve the feature extraction ability and strengthen small-scale target detection. Then, the predicted boxes filtering algorithm uses the soft-merge function to replace the traditionally used NMS (non-maximum suppression). Soft-merge can improve the model’s detection accuracy by fusing the information of predicted boxes. Finally, the experimental results show that the mAP (mean average precision) and FPS (frames per second) of Fast-YOLOv4 reach 90.62% and 54 f/s, respectively, in the workstation. In the NVIDIA Jetson Nano platform, the FPS of Fast-YOLOv4 is 2.5 times that of YOLOv4. This improved model performance meets the requirements for real-time detection and thus has theoretical significance and application value.

Aiming at the problems of the large amount of model parameters and false and missing detections of multi-scale drone targets, we present a novel drone detection method, YOLOv4-MCA, based on the lightweight MobileViT and Coordinate Attention. The proposed approach is improved according to the framework of YOLOv4. Firstly, we use an improved lightweight MobileViT as the feature extraction backbone network, which can fully extract the local and global feature representations of the object and reduce the model’s complexity. Secondly, we adopt Coordinate Attention to improve PANet and to obtain a multi-scale attention called CA-PANet, which can obtain more positional information and promote the fusion of information with low- and high-dimensional features. Thirdly, we utilize the improved K-means++ method to optimize the object anchor box and improve the detection efficiency. At last, we construct a drone dataset and conduct a performance experiment based on the Mosaic data augmentation method. The experimental results show that the mAP of the proposed approach reaches 92.81%, the FPS reaches 40 f/s, and the number of parameters is only 13.47 M, which is better than mainstream algorithms and achieves a high detection accuracy for multi-scale drone targets using a low number of parameters.

A new generation of multipurpose measurement equipment is transforming the role of computers in instrumentation. The new features involve mixed devices, such as kinds of sensors, analog-to-digital and digital-to-analog converters, and digital signal processing techniques, that are able to substitute typical discrete instruments like multimeters and analyzers. Signal-processing applications frequently use least-squares (LS) sine-fitting algorithms. Periodic signals may be interpreted as a sum of sine waves with multiple frequencies: the Fourier series. This paper describes a new sine fitting algorithm that is able to fit a multiharmonic acquired periodic signal. By means of a “sinusoidal wave” whose amplitude and phase are both transient, the “triangular wave” can be reconstructed on the basis of Hilbert-Huang transform (HHT). This method can be used to test effective number of bits (ENOBs) of analog-to-digital converter (ADC), avoiding the trouble of selecting initial value of the parameters and working out the nonlinear equations. The simulation results show that the algorithm is precise and efficient. In the case of enough sampling points, even under the circumstances of low-resolution signal with the harmonic distortion existing, the root mean square (RMS) error between the sampling data of original “triangular wave” and the corresponding points of fitting “sinusoidal wave” is marvelously small. That maybe means, under the circumstances of any periodic signal, that ENOBs of high-resolution ADC can be tested accurately.

Radar emitter signal (RES) recognition is the important content in radar reconnaissance and signal processing. In order to study the problem of RES recognition, and to improve the RES recognition rate of the electronic warfare equipment, the empirical mode decomposition (EMD) theory and wavelet packet (WP) are introduced into RES feature extraction. A new RES recognition method is proposed based on WP, EMD and neural network (NN). It uses wavelet packet to finish decomposition, de-noising and reconstruction of the RES. Then obtain the intrinsic mode function (IMF) through EMD, which can embody the characteristics of the RES. The energy of each IMF are calculated and normalized, which would be regarded as the feature vector. By constructing back propagation neural network (BPNN) classifier and redial basis function neural network (RBFNN) classifier, it realizes the RES recognition finally. Experiment results show that the RES recognition method based on WP, EMD and NN is an effective recognition method, which can achieve satisfying correct recognition rate in a larger signal to noise ratio, and has certain reference value in follow-up in-depth study

The emergence of adversarial examples has revealed the inadequacies in the robustness of image classification models based on Convolutional Neural Networks (CNNs). Particularly in recent years, the discovery of natural adversarial examples has posed significant challenges, as traditional defense methods against adversarial attacks have proven to be largely ineffective against these natural adversarial examples. This paper explores defenses against these natural adversarial examples from three perspectives: adversarial examples, model architecture, and dataset. First, it employs Class Activation Mapping (CAM) to visualize how models classify natural adversarial examples, identifying several typical attack patterns. Next, various common CNN models are analyzed to evaluate their susceptibility to these attacks, revealing that different architectures exhibit varying defensive capabilities. The study finds that as the depth of a network increases, its defenses against natural adversarial examples strengthen. Lastly, Finally, the impact of dataset class distribution on the defense capability of models is examined, focusing on two aspects: the number of classes in the training set and the number of predicted classes. This study investigates how these factors influence the model’s ability to defend against natural adversarial examples. Results indicate that reducing the number of training classes enhances the model’s defense against natural adversarial examples. Additionally, under a fixed number of training classes, some CNN models show an optimal range of predicted classes for achieving the best defense performance against these adversarial examples.

Entity alignment (EA) aims to automatically determine whether an entity pair in different knowledge bases or knowledge graphs refer to the same entity in reality. Inspired by human cognitive mechanisms, we propose a coarse-to-fine entity alignment model (called CFEA) consisting of three stages: coarse-grained, middle-grained, and fine-grained. In the coarse-grained stage, a pruning strategy based on the restriction of entity types is adopted to reduce the number of candidate matching entities. The goal of this stage is to filter out pairs of entities that are clearly not the same entity. In the middle-grained stage, we calculate the similarity of entity pairs through some key attribute values and matched attribute values, the goal of which is to identify the entity pairs that are obviously not the same entity or are obviously the same entity. After this step, the number of candidate entity pairs is further reduced. In the fine-grained stage, contextual information, such as abstract and description text, is considered, and topic modeling is carried out to achieve more accurate matching. The basic idea of this stage is to use more information to help judge entity pairs that are difficult to distinguish using basic information from the first two stages. The experimental results on real-world datasets verify the effectiveness of our model compared with baselines.