Nowadays, massive data appears in the form of high-speed data streams. It is an important and challenging problem to perform various mining tasks on data streams, such as finding heavy hitters, estimating frequencies,...
详细信息
ISBN:
(数字)9798350369007
ISBN:
(纸本)9798350369014
Nowadays, massive data appears in the form of high-speed data streams. It is an important and challenging problem to perform various mining tasks on data streams, such as finding heavy hitters, estimating frequencies, and etc. Current applications often need to handle several tasks at the same time. Traditional sketch solutions rely on different data structures and algorithms for different tasks. As a result, multiple data structures are needed in practice. In this paper, we propose a generic sketch algorithm, namely HeavyCache, which can quickly record each item and perform a broad spectrum of data mining tasks. The key idea is to leverage cache mechanism to separate heavy items from light items. Specifically, HeavyCache accurately records the detailed information of items, while simply approximately records the frequencies of light items. We show how HeavyCache is used to process the six typical tasks. Experimental results show that our HeavyCache significantly outperforms state-of-the-art solutions in terms of both accuracy and speed for each of the six tasks.
The use of Unmanned Aerial Vehicles (UAVs) in intelligent urban planning has become increasingly important, offering advanced imaging capabilities and facilitating the implementation of cutting-edge technologies such ...
详细信息
ISBN:
(数字)9798350373295
ISBN:
(纸本)9798350373301
The use of Unmanned Aerial Vehicles (UAVs) in intelligent urban planning has become increasingly important, offering advanced imaging capabilities and facilitating the implementation of cutting-edge technologies such as image classification and changedetection. This study explores the technological innovations and practical applications of UAVs in urban planning, focusing on the development of novel image analysis algorithms. By integrating advanced airborne imagers with drone platforms, the study proposes innovative approaches for image classification and changedetection using deep learning methods. The proposed algorithms are rigorously evaluated through extensive experiments, demonstrating their effectiveness in improving the accuracy and efficiency of urban planning processes. The results highlights the potential of UAV-based techniques to revolutionize urban planning practices, paving the way for sustainable and informed decision-making in complex urban environments.
Underwater object detection is a typical problem in underwater exploration, widely used in fields such as underwater archaeology, underwater rescue, and underwater facility maintenance. Due to the particularity of wat...
详细信息
ISBN:
(数字)9798350384185
ISBN:
(纸本)9798350384192
Underwater object detection is a typical problem in underwater exploration, widely used in fields such as underwater archaeology, underwater rescue, and underwater facility maintenance. Due to the particularity of water media, underwater object detection has brought challenges compared to land object detection. In order to overcome the limitations of water media, sonar is a commonly used tool. Ships are equipped with side scan sonar (SSS) to conduct large-scale surveys of underwater topography. Object detection is achieved by observing terrain changes. With the development of machine learning technology, the use of autonomous underwater vehicles (AUVs) equipped with side scan sonar for autonomous underwater object detection has become a trend. Usually, training machine learning algorithms requires extensive training samples. For situations where the object shape is unknown in advance, general autonomous object detection methods are difficult to be effective. It needs to develop unsupervised object detectionalgorithms to solve this problem, and we attempt to apply image change detection algorithms to SSS images. Therefore, in this paper, we investigate the alignment problem of SSS images. Specifically, we adopt three alignment algorithms, including RANSAC, GLU-Net and UDIS++, for underwater object detection in SSS images. To verify the feasibility and timeliness of these algorithms, we conduct extensive experiments on common multi beam SSS images and synthetic aperture sonar images.
Learning-based changedetection (CD) in water scenarios is a key functionality for unmanned aerial vehicle (UAV). However, computer vision algorithms require large number of labeled datasets. Inspired by parallel inte...
详细信息
Learning-based changedetection (CD) in water scenarios is a key functionality for unmanned aerial vehicle (UAV). However, computer vision algorithms require large number of labeled datasets. Inspired by parallel intelligence, we propose a systematic framework for data generation. In this work, the framework consists of simulated scene and image generation network. In simulated scene, simulated images with pixel-level annotations are automatically generated. Then, image generation network uses paired images (real and simulated) to generate synthetic images. We use simulated and synthetic images in combination with publicly available real-world images to conduct experiments. The experimental results indicate that: 1) simulated images can be used in changedetection research; 2) synthetic images effectively improve the performance of supervised changedetection model.
Beam tracking methods are instrumental for efficient use of the multi-gigahertz bandwidth available at mmWave frequencies. In this paper, we propose a Multi-armed Bandit (MAB) based Reinforcement Learning (RL) algorit...
详细信息
ISBN:
(数字)9798350303582
ISBN:
(纸本)9798350303599
Beam tracking methods are instrumental for efficient use of the multi-gigahertz bandwidth available at mmWave frequencies. In this paper, we propose a Multi-armed Bandit (MAB) based Reinforcement Learning (RL) algorithm to periodically select transmitter-receiver beam pairs so as to maximize the average spectral efficiency. Contrary to a traditional Bayesian MAB-based approach, the MAB algorithm proposed by us can track a user as it moves across multiple correlation distances. The algorithm keeps track of the received signal strength to detect a change in the channel correlation and adjusts its strategy to adapt to the new channel conditions. We derive an upper bound on the regret of the proposed algorithm. The proposed algorithm is evaluated on channel data generated using the open-source simulator NYUSIM and is observed to outperform existing algorithms, thus removing the requirement of repeated initial access procedures.
Within the realm of deep learning techniques, numerous remote sensing applications can be effectively addressed using deep learning algorithms. However, there is a scarcity of studies in synthetic Aperture radar (SAR)...
详细信息
ISBN:
(数字)9798350360325
ISBN:
(纸本)9798350360332
Within the realm of deep learning techniques, numerous remote sensing applications can be effectively addressed using deep learning algorithms. However, there is a scarcity of studies in synthetic Aperture radar (SAR)-based urban flood mapping involving deep learning techniques, primarily due to two reasons. First, SAR-based urban flood mapping is inherently rooted in changedetection, resulting in a complex multi-modality problem within the imbalance data. This complexity arises from the integration of SAR intensity, InSAR coherence, and even SAR phase information acquired from different polarizations (i.e., VV and VH polarization in Sentinel-1 data) both before and after the event. The second challenge is the absence of a benchmark dataset specifically designed for SAR-based urban flood mapping. In an effort to fill this gap, a benchmark dataset for large-scale flood mapping using Sentinel-1 data, which includes not only SAR intensity but also InSAR coherence, should be created. The SAR pre-processing should be carefully checked at the very beginning. With this aim, we tested the specking filter and kernel size selection for the SAR preprocessing for deep learning models. Through this initiative, we found that despeckling SAR intensity and selecting the kernel size in InSAR coherence calculation do not significantly affect the accuracy in deep learning-based urban flood mapping using Sentinel-1 data. The curated benchmark dataset will be presented in the final paper.
Today's large knowledge graphs are conceived mainly for supporting search and e-commerce within large companies such as Google or Amazon, with well-crafted knowledge creation rules. Our recent experience of the CO...
详细信息
ISBN:
(数字)9798350317152
ISBN:
(纸本)9798350317169
Today's large knowledge graphs are conceived mainly for supporting search and e-commerce within large companies such as Google or Amazon, with well-crafted knowledge creation rules. Our recent experience of the COVID-19 pandemic, when knowledge has grown at unprecedented rates and has been often contradictory, inspired us to capture a huge gap in existing concepts and technology: today's knowledge management does not adequately support such a disruptive process. In this article, we propose the design and prototyping of the next generation of knowledge management concepts and systems, which will support domain diversity and scientific evolution as foundational ingredients. change management is based on a reactive approach, well-established in database systems, but so far lacking in knowledge systems. We propose the reactive interaction of several knowledge hubs, each developed within a scientific domain and “owner” of a portion of a common knowledge representation. Knowledge is represented as graphs, with nodes and edges; edges may inter-connect nodes from different hubs. Most importantly, reactive rules cross the hub's borders and create the premises for a disciplined knowledge evolution, even under the pressure of crises. Similar challenges are not restricted to the recent pandemic and can address other crisis scenarios, including the catastrophic consequences of climate change or the recent (r)-evolution in artificial intelligence, studied by several scientific communities, whose management requires complex and controversial choices.
In this work, we consider the problem of transmission rate selection for a discrete time point-to-point block fading wire-less communication link. The wireless channel remains constant within the channel coherence tim...
详细信息
ISBN:
(纸本)9781728181059
In this work, we consider the problem of transmission rate selection for a discrete time point-to-point block fading wire-less communication link. The wireless channel remains constant within the channel coherence time but can change rapidly across blocks. The goal is to design a link rate selection strategy that can identify the best transmission rate quickly and adaptively in quasi-static channels. This problem can be cast into the stochastic bandit framework, and the unawareness of time-stamps where channel changes necessitates running change-point detection simultaneously with stochastic bandit algorithms to improve adaptivity. We present a joint channel change-point detection and link rate selection algorithm based on Thompson Sampling (CD-TS) and show it can achieve a sublinear regret with respect to the number of time steps $T$ when the channel coherence time is larger than a threshold. We then improve the CD-TS algorithm by considering the fact that higher transmission rate has higher packet-loss probability. Finally, we validate the performance of the proposed algorithms through numerical simulations.
changedetection in multiple-temporal Synthetic Aperture Radar images has been received great interests for recent decades. The basic principle of changedetection is to analyse the difference images generated from tw...
详细信息
ISBN:
(纸本)9781665410021
changedetection in multiple-temporal Synthetic Aperture Radar images has been received great interests for recent decades. The basic principle of changedetection is to analyse the difference images generated from two Synthetic Aperture Radar images captured in the same geographic area at two different times. The popular operators used to create difference images are traditional subtraction, ratio, logarithm based ones and modified versions of them, which can use pixel information in the local or global areas. A challenge in detecting changes is to reduce impacts of speckle noises inherently existing in Synthetic Aperture Radar images on the accuracy of the detection. This paper proposed a novel algorithm to create the difference images based on averaging heterogeneous factors of corresponding neighbourhood areas in the two images. The resultant difference image is then filtered by the average filter to reject remaining speckle noises.
The automatic detection of changes or anomalies between multispectral and hyperspectral images collected at different time instants is an active and challenging research topic. To effectively perform change-point dete...
详细信息
ISBN:
(数字)9789082797053
ISBN:
(纸本)9781728150017
The automatic detection of changes or anomalies between multispectral and hyperspectral images collected at different time instants is an active and challenging research topic. To effectively perform change-point detection in multitemporal images, it is important to devise techniques that are computationally efficient for processing large datasets, and that do not require knowledge about the nature of the changes. In this paper, we introduce a novel online framework for detecting changes in multitemporal remote sensing images. Acting on neighboring spectra as adjacent vertices in a graph, this algorithm focuses on anomalies concurrently activating groups of vertices corresponding to compact, well-connected and spectrally homogeneous image regions. It fully benefits from recent advances in graph signal processing to exploit the characteristics of the data that lie on irregular supports. Moreover, the graph is estimated directly from the images using superpixel decomposition algorithms. The learning algorithm is scalable in the sense that it is efficient and spatially distributed. Experiments illustrate the detection and localization performance of the method.
暂无评论