Image entropy and empirical mode decomposition （EMD） are effective methods for target detection. EMD algorithm is a powerful tool for adaptive multiscale analysis of nonstationary signals. A new technique based on E...Image entropy and empirical mode decomposition （EMD） are effective methods for target detection. EMD algorithm is a powerful tool for adaptive multiscale analysis of nonstationary signals. A new technique based on EMD and modified local entropy is proposed in small target detection under sea-sky background. With the EMD algorithm, it is valid to estimate the background and get the target image by removing the background from the original image and segmenting the target based on the modified local entropy method. The data analysis and experiments show the validity of the proposed algorithm.展开更多
Dynamic hand gesture recognition is a desired alternative means for human-computer interactions.This paper presents a hand gesture recognition system that is designed for the control of flights of unmanned aerial vehi...Dynamic hand gesture recognition is a desired alternative means for human-computer interactions.This paper presents a hand gesture recognition system that is designed for the control of flights of unmanned aerial vehicles(UAV).A data representation model that represents a dynamic gesture sequence by converting the 4-D spatiotemporal data to 2-D matrix and a 1-D array is introduced.To train the system to recognize designed gestures,skeleton data collected from a Leap Motion Controller are converted to two different data models.As many as 9124 samples of the training dataset,1938 samples of the testing dataset are created to train and test the proposed three deep learning neural networks,which are a 2-layer fully connected neural network,a 5-layer fully connected neural network and an 8-layer convolutional neural network.The static testing results show that the 2-layer fully connected neural network achieves an average accuracy of 96.7%on scaled datasets and 12.3%on non-scaled datasets.The 5-layer fully connected neural network achieves an average accuracy of 98.0%on scaled datasets and 89.1%on non-scaled datasets.The 8-layer convolutional neural network achieves an average accuracy of 89.6%on scaled datasets and 96.9%on non-scaled datasets.Testing on a drone-kit simulator and a real drone shows that this system is feasible for drone flight controls.展开更多
Drones,also known as mini-unmanned aerial vehicles(UAVs),are enjoying great popularity in recent years due to their advantages of low cost,easy to pilot and small size,which also makes them hard to detect.They can pro...Drones,also known as mini-unmanned aerial vehicles(UAVs),are enjoying great popularity in recent years due to their advantages of low cost,easy to pilot and small size,which also makes them hard to detect.They can provide real time situational awareness information by live videos or high definition pictures and pose serious threats to public security.In this article,we combine collaborative spectrum sensing with deep learning to effectively detect potential illegal drones with states of high uncertainty.First,we formulate the detection of potential illegal drones under illegitimate access and rogue power emission as a quaternary hypothesis test problem.Then,we propose an algorithm of image classification based on convolutional neural network which converts the cooperative spectrum sensing data at a sensing slot into one image.Furthermore,to exploit more information and improve the detection performance,we develop a trajectory classification algorithm which converts theflight process of the drones in consecutive multiple sensing slots into trajectory images.In addition,simulations are provided to verify the proposed methods’performance under various parameter configurations.展开更多
The ability to hit a target with precision and from a great distance has been reserved for the world’s superpowers. However, this resource is increasingly being threatened as drones with this long-range and precision...The ability to hit a target with precision and from a great distance has been reserved for the world’s superpowers. However, this resource is increasingly being threatened as drones with this long-range and precision capability are becoming more accessible to those who don’t have this strategic ability. This article starts with an analysis of the Iranian HESA Shahed 136 drone to discuss the latest innovations in low-cost long-range precision weapons, specifically the use of kamikaze drones and loitering munitions. This is an exploratory study that starts by discussing the notion of a kamikaze drone and then analyses the design options for the Shahed 136, to reflect on the future of this new type of weapon and its implications for the economic and political relationship between weapon and cost. The conclusion is that the HESA Shahed 136 revolutionizes the concept of precise long-range strikes, a function that until now was reserved for expensive and technologically demanding tactical missiles and aircraft, and which can now be carried out with cheap drones. This creates an arms race not only in producing the most technological and precise weaponry but also the least expensive.展开更多
We propose an improved algorithm based on fractal dimension and third-order characterization to detect dim target with cluttered background in an infrared （IR） image. We also illustrate the performance and efficienc...We propose an improved algorithm based on fractal dimension and third-order characterization to detect dim target with cluttered background in an infrared （IR） image. We also illustrate the performance and efficiency comparisons between the presented algorithm and the traditional fractal detection method on real IR images. The experimental results show that the proposed algorithm is robust and efficient for IR dim target detection.展开更多
基金supported in part by the National Natural Science Foundation of China for Young Scholars under Grant No.40801164
文摘Image entropy and empirical mode decomposition （EMD） are effective methods for target detection. EMD algorithm is a powerful tool for adaptive multiscale analysis of nonstationary signals. A new technique based on EMD and modified local entropy is proposed in small target detection under sea-sky background. With the EMD algorithm, it is valid to estimate the background and get the target image by removing the background from the original image and segmenting the target based on the modified local entropy method. The data analysis and experiments show the validity of the proposed algorithm.
文摘Dynamic hand gesture recognition is a desired alternative means for human-computer interactions.This paper presents a hand gesture recognition system that is designed for the control of flights of unmanned aerial vehicles(UAV).A data representation model that represents a dynamic gesture sequence by converting the 4-D spatiotemporal data to 2-D matrix and a 1-D array is introduced.To train the system to recognize designed gestures,skeleton data collected from a Leap Motion Controller are converted to two different data models.As many as 9124 samples of the training dataset,1938 samples of the testing dataset are created to train and test the proposed three deep learning neural networks,which are a 2-layer fully connected neural network,a 5-layer fully connected neural network and an 8-layer convolutional neural network.The static testing results show that the 2-layer fully connected neural network achieves an average accuracy of 96.7%on scaled datasets and 12.3%on non-scaled datasets.The 5-layer fully connected neural network achieves an average accuracy of 98.0%on scaled datasets and 89.1%on non-scaled datasets.The 8-layer convolutional neural network achieves an average accuracy of 89.6%on scaled datasets and 96.9%on non-scaled datasets.Testing on a drone-kit simulator and a real drone shows that this system is feasible for drone flight controls.
基金supported by the Foundation of Graduate Innovation Center in NUAA under Grant No. kfjj20190414the open research fund of Key Laboratory of Dynamic Cognitive System of Electromagnetic Spectrum Space (Nanjing Univ. Aeronaut. Astronaut.), Ministry of Industry and Information Technology, Nanjing, 211106, China (No. KF20181913)+2 种基金National Natural Science Foundation of China (No. 61631020, No. 61871398, No. 61931011 and No. 61801216)the Natural Science Foundation for Distinguished Young Scholars of Jiangsu Province (No. BK20190030)the Natural Science Foundation of Jiangsu Province (No. BK20180420)
文摘Drones,also known as mini-unmanned aerial vehicles(UAVs),are enjoying great popularity in recent years due to their advantages of low cost,easy to pilot and small size,which also makes them hard to detect.They can provide real time situational awareness information by live videos or high definition pictures and pose serious threats to public security.In this article,we combine collaborative spectrum sensing with deep learning to effectively detect potential illegal drones with states of high uncertainty.First,we formulate the detection of potential illegal drones under illegitimate access and rogue power emission as a quaternary hypothesis test problem.Then,we propose an algorithm of image classification based on convolutional neural network which converts the cooperative spectrum sensing data at a sensing slot into one image.Furthermore,to exploit more information and improve the detection performance,we develop a trajectory classification algorithm which converts theflight process of the drones in consecutive multiple sensing slots into trajectory images.In addition,simulations are provided to verify the proposed methods’performance under various parameter configurations.
文摘The ability to hit a target with precision and from a great distance has been reserved for the world’s superpowers. However, this resource is increasingly being threatened as drones with this long-range and precision capability are becoming more accessible to those who don’t have this strategic ability. This article starts with an analysis of the Iranian HESA Shahed 136 drone to discuss the latest innovations in low-cost long-range precision weapons, specifically the use of kamikaze drones and loitering munitions. This is an exploratory study that starts by discussing the notion of a kamikaze drone and then analyses the design options for the Shahed 136, to reflect on the future of this new type of weapon and its implications for the economic and political relationship between weapon and cost. The conclusion is that the HESA Shahed 136 revolutionizes the concept of precise long-range strikes, a function that until now was reserved for expensive and technologically demanding tactical missiles and aircraft, and which can now be carried out with cheap drones. This creates an arms race not only in producing the most technological and precise weaponry but also the least expensive.
文摘We propose an improved algorithm based on fractal dimension and third-order characterization to detect dim target with cluttered background in an infrared （IR） image. We also illustrate the performance and efficiency comparisons between the presented algorithm and the traditional fractal detection method on real IR images. The experimental results show that the proposed algorithm is robust and efficient for IR dim target detection.
