Unmanned aerial vehicles (UAVs) play an important role in conducting automatic patrol inspections of cities, which can ensure the safety urban residents' life and property normal operation cities. However, during inspection process, problems may arise. For example, numerous small objects UAV images are difficult to detect, severely occluded, requirements for real-time performances posed. To address these issues, we first propose a object detection network (RTD-Net) images. Besides, deal with...

Sampling techniques are widely used for traffic measurements at high link speed to conserve router resources. Traditionally, sampled data is network management tasks such as matrix estimations, but recently it has also been in numerous anomaly detection algorithms, security analysis becomes increasingly critical providers. While the impact of sampling on engineering metrics flow size and mean rate well studied, its remains an open question.This paper presents a comprehensive study whether...

Railway shunting accidents, in which trains collide with obstacles, often occur because of human error or fatigue. It is therefore necessary to detect traffic objects front the and inform driver take timely action. To these railways, we proposed an object-detection method using a differential feature fusion convolutional neural network (DFF-Net). DFF-Net includes two modules: prior module module. The produces initial anchor boxes for subsequent detection Taking as input, applies sub-module...

Unmanned aerial vehicles (UAVs) have been widely used in post-disaster search and rescue operations, object tracking, other tasks. Therefore, the autonomous perception of UAVs based on computer vision has become a research hotspot recent years. However, UAV images include dense objects, small arbitrary directions, which bring about significant challenges to existing detection methods. To alleviate these issues, we propose global-local feature enhanced network (GLF-Net). Considering...

While immutability is an important property of blockchain and can be utilized to build a credible data publishing platform, the same also abused disseminate illicit content may violate requirements under General Data Protection Regulation (GDPR) as well other privacy regulations. Hence, there has been increased focus on designing redactable blockchain-based solutions, which this survey article. Specifically, we extant literature blockchain, discuss existing limitations challenges prior...

Parameter configuration is a common procedure used in large-scale network protocols to support multiple operational goals. It can be formulated as black-box optimization problem and solved with an efficient search algorithm. This paper proposes new heuristic algorithm, Recursive Random Search(RRS), for parameter optimization. The RRS algorithm based on the initial high-efficiency feature of random sampling it attempts maintain this high efficiency by constantly "restarting" adjusted sample...

Unauthorized operations referred to as "black flights" of unmanned aerial vehicles (UAVs) pose a significant danger public safety, and existing low-attitude object detection algorithms encounter difficulties in balancing precision speed. Additionally, their accuracy is insufficient, particularly for small objects complex environments. To solve these problems, we propose lightweight feature-enhanced convolutional neural network able perform with high flying real time provide guidance...

With the growing popularity of civilian unmanned aerial vehicles (UAVs), unauthorized flights are on rise accordingly. Therefore, it is critical to detect low-altitude UAVs for protecting personal privacy and public safety. Though substantial progress has been made in UAV detection, existing detection methods still have problems balancing accuracy, model size, speed. To address these limitations, this article proposes a novel deep learning method named convolution–transformer network...

Many accidents happen under shunting mode when the speed of a train is below 45 km/h. In this mode, attendants observe railway condition ahead using traditional manual method and tell observation results to driver in order avoid danger. To address problem, an automatic object detection system based on convolutional neural network (CNN) proposed detect objects which called Feature Fusion Refine (FR-Net). It consists three connected modules, i.e., depthwise-pointwise convolution, coarse...

The lack of floor plans is a critical reason behind the current sporadic availability indoor localization service. Service providers have to go through effort-intensive and time-consuming business negotiations with building operators, or hire dedicated personnel gather such data. In this paper, we propose Jigsaw, plan reconstruction system that leverages crowdsensed data from mobile users. It extracts position, size, orientation information individual landmark objects images taken by also...

Obstacles in front of a train pose significant threat to traffic safety, and many accidents happen under shunting mode when the speed is below 45 km/h. The existing track object–detection algorithms encounter difficulty balancing detection precision mode. Additionally, their accuracy insufficient, particularly for small objects complex environments. To address these problems, we propose stable lightweight feature extraction adaptive fusion network real-time obstacles railway scenarios ensure...

Images are important information carriers in our lives, and images should be secure when transmitted stored. Image encryption algorithms based on chaos theory emerge endlessly. Based previous various chaotic image fast algorithms, this paper proposes a color sector algorithm one-dimensional composite sinusoidal mapping. The main purpose of is to improve the decryption speed efficiency big data era. First, four basic maps combined pairs added with sine operations. Six (CSCM) were obtained....

Abstract Background This study compared the effectiveness of different ultrawide field fundus imaging systems (Clarus™ 500 and Optos ® ) in diagnosing diabetic retinopathy (DR). Methods was a prospective, multicentre study. Retinal photographs were captured at four eye centres utilizing both Clarus™ systems. The image quality effective retinal area compared. Then consistency severity DR by two distinct according to three separate grading criteria (the International Council Ophthalmology,...

Research on the detection and identification of anomalies in electric power systems is crucial for ensuring their secure stable operation. Anomaly models based Support Vector Machines (SVMs) effectively process high-dimensional data while maintaining strong generalization capabilities. However, performance SVMs significantly depends choice parameters, where improper parameter settings can lead to overfitting or underfitting, consequently decreasing accuracy anomaly detection. Furthermore,...

Coordinating various controllable distributed resources to reduce network losses is crucial the secure and economical operation of modern power systems. This paper proposes a bi-level optimization model for system loss reduction based on “source-grid-load-storage” coordinated optimization. The upper level aims minimize total annual planning cost system, determining location capacity photovoltaic systems, energy storage devices, electric vehicle charging stations. lower load curve smoothness...

Considerable research has been done on detecting and blocking portscan activities that are typically conducted by infected hosts to discover other vulnerable hosts. However, the focus enterprise gateway-level intrusion detection systems where traffic volume is low network configuration information readily available. This paper investigates effectiveness of existing algorithms in context a large transit backbone proposes new algorithm meets demands aggregated high speed traffic. Specifically,...

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> Packet sampling is commonly deployed in high-speed backbone routers to minimize resources used for network monitoring. It known that packet distorts traffic statistics and its impact has been extensively studied engineering metrics such as flow size mean rate. However, it unclear how impacts anomaly detection, which become increasingly critical providers. This paper the first attempt address this...

With the high growth rates of railway transportation, it is extremely important to detect obstacles ahead train ensure safety. Manual and traditional feature-extraction methods have been utilized in this scenario. There are also deep learning-based object detection approaches. However, case a complex scene, these approaches either inefficient or insufficient accuracy, particularly for small objects. To address issue, we propose feature-enhanced single-shot detector (FE-SSD). The proposed...

Detection of railway traffic objects is an important task during train driving and implemented to ensure safe driving. Although object detection has been investigated for years, many challenges exist in precisely detecting under complex scenes. These mainly include adverse weather states, various backgrounds, diverse objects, low-quality images. To address these issues, we introduce a novel deep learning method, called multi-mode feature enhanced convolutional neural network (MMFE-Net),...

As the Internet infrastructure grows to support a variety of services, its legacy protocols are being overloaded with new functions such as traffic engineering. Today, operators engineer capabilities through clever, but <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">manual</i> parameter tuning. In this paper, we propose back-end tool for large-scale configuration that is based on efficient state space search techniques and on-line simulation. The...