With the breakthroughs in deep learning, recent years have witnessed a booming of artificial intelligence (AI) applications and services, spanning from personal assistant to recommendation systems video/audio surveillance. More recently, with proliferation mobile computing Internet Things (IoT), billions IoT devices are connected Internet, generating zillions bytes data at network edge. Driving by this trend, there is an urgent need push AI frontiers edge so as fully unleash potential big...

With the wide penetration of smart robots in multifarious fields, simultaneous localization and mapping (SLAM) technique robotics has attracted growing attention community. Yet collaborating SLAM over multiple still remains challenging due to performance contradiction between intensive graphics computation limited computing capability robots. While traditional solutions resort powerful cloud servers acting as an external provider, we show by real-world measurements that significant...

The Yellow River Delta (YRD), known for its vast and diverse wetland ecosystem, is the largest estuarine delta in China. However, human activities climate change have significantly degraded ecosystem recent decades YRD. Therefore, an understanding of land use modifications essential efficient management preservation ecosystems this region. This study utilized time series remote sensing data extreme gradient boosting method to generate maps YRD from 2000 2020. Several methods, including...

Three-dimensional magnetic recording (3DMR) is a highly promising approach to achieving ultra-large data storage capacity in hard disk drives. One of the greatest challenges for 3DMR lies performing sequential and correct writing bits into multi-layer medium. In this work, we have proposed hierarchical architecture based on layered heat-assisted with multi-head array. The feasibility validated dual-layer system FePt-based thin films via micromagnetic simulation. Our results reveal...

Recently, the explosion of resource-hungry and delay-sensitive Internet-of-Things (IoT) applications as exemplified by wearable appliances, video surveillance, connected vehicles have posed great challenges on underlying IoT devices which typically limited computation resource. In response, offloading is envisioned a promising approach to augmenting capability devices. Toward real-time efficient offloading, in this paper we propose novel edge resource pooling framework, massive crowd at...

With the breakthroughs in deep learning, recent years have witnessed a booming of artificial intelligence (AI) applications and services, spanning from personal assistant to recommendation systems video/audio surveillance. More recently, with proliferation mobile computing Internet-of-Things (IoT), billions IoT devices are connected Internet, generating zillions Bytes data at network edge. Driving by this trend, there is an urgent need push AI frontiers edge so as fully unleash potential big...

In many delay-sensitive monitoring and surveillance applications, unmanned aerial vehicles (UAVs) can act as edge servers in the air to coordinate with base stations (BSs) for in-situ data collection processing achieve real-time situation awareness. order ensure long-term freshness requirements of awareness, a swarm UAVs need fly frequently among different sensing regions. However, nonstop flying may quickly drain batteries armed UAVs, hence an energy-efficient algorithm UAVs' dynamic...

Graph Neural Networks (GNNs) have gained growing interest in miscellaneous applications owing to their outstanding ability extracting latent representation on graph structures. To render GNN-based service for IoT-driven smart applications, the traditional model serving paradigm resorts cloud by fully uploading geo-distributed input data remote datacenter. However, our empirical measurements reveal significant communication overhead of such cloud-based and highlight profound potential...

Federated Learning (FL) has been a promising paradigm in distributed machine learning that enables in-situ model training and global aggregation. While it can well preserve private data for end users, to apply efficiently on IoT devices yet suffer from their inherent variants: available computing resources are typically constrained, heterogeneous, changing dynamically. Existing works deploy FL by pruning sparse or adopting tiny counterpart, which alleviates the workload but may have negative...

With wide penetration of smart robots in many application fields, Simultaneous Localization And Mapping (SLAM) has attracted great attention the community. Yet performance issue on multi-robot SLAM still remains challenging due to contradiction constrained on-device resources and intensive graphics computation. While traditional approaches resort powerful cloud servers accelerate computing, we show by real-world measurements that significant communication overhead prevents its practicability...

Graph Neural Networks (GNNs) have gained growing interest in miscellaneous applications owing to their outstanding ability extracting latent representation on graph structures. To render GNN-based service for IoT-driven smart applications, traditional model serving paradigms usually resort the cloud by fully uploading geo-distributed input data remote datacenters. However, our empirical measurements reveal significant communication overhead of such cloud-based and highlight profound...

The explosion of resource-hungry mobile applications has posed great challenges on the underlying devices which typically have limited computation resource. In response, device-to-device (D2D) offloading is envisioned as a promising approach to problem by gearing resource-rich and resource-poor devices. Towards real-time efficient offloading, in this paper, we proposed novel edge resource pooling framework called ERP, massive crowd at network exploit D2D collaboration for sharing with each...

To identify dense and small-size pedestrians in surveillance systems, high-resolution cameras are widely deployed, where images captured delivered to off-the-shelf pedestrian detection models. However, given the highly computation-intensive workload brought by high resolution, resource-constrained fail afford accurate inference real time. address that, we propose Hode, an offloaded video analytic framework that utilizes multiple edge nodes proximity expedite with inputs. Specifically, Hode...

In view of the fact that accuracy texture image classification is easily affected by changes in illumination and rotation, based on analysis geometric curvatures information microscopic surface completed local binary pattern (CLBP), this paper proposed a new descriptor, named as Geometry-based Completed Local Binary Pattern (GCLBP). Inspired continuous rotation invariance robustness curvature information, principal (PCs) all pixels are first calculated then used to represent gradient...

An algorithm is called stable at a training set S if any change of single point in yields only small the output. Stability learning necessary for learnability supervised classification and regression setting. In this paper, we give formal definitions strong weak stability randomized algorithms prove non-asymptotic bounds on difference between empirical expected error.

To reduce invalid rules in the mining of association rules, we have analyzed reasons and presented a relative confidence judgment criteria. Based on value confidence, classify strong into positive, negative rules. We offer an algorithm with new criterion make tests Visual FoxPro. The indicate that method stated this paper can obviously

Target tracking is currently a hot research topic in machine vision. The traditional target algorithm based on the generative model selects features manually, which has simple structure and fast running speed, but it cannot meet requirements of accuracy complex scenes. Compared with algorithms, due to good performance, method full convolutional network become one important methods tracking. However, RPN-based Siamese lacks positional reliability when predicting area. Aiming at low network,...