Robust automatic pavement crack detection is critical to automated road condition evaluation. However, research on still limited and pixel-level remains a challenging problem, due heterogeneous pixel intensity, complex topology, poor illumination condition, noisy texture background. In this paper, we propose novel approach for automatically detecting cracks at level, leveraging multi-scale neighborhood information, intensity. Using intensity probabilistic generative model (PGM) based method...

Recent breakthrough in wireless energy transfer technology has enabled sensor networks (WSNs) to operate with zero-downtime through the use of mobile chargers (MCs), that periodically replenish supply nodes. Due limited battery capacity MCs, a significant number MCs and charging depots are required guarantee perpetual operations large scale networks. Existing methods for reducing treat tour planning depot positioning problems separately even though they inter-dependent. This paper is first...

Accurate travel-time prediction of public transport is essential for reliable journey planning in urban transportation systems. However, existing studies on bus travel-/arrival-time often focus only improving the accuracy a single trip. This inadequate modern systems, where usually consists multiple trips. In this paper, we investigate problem journeys that takes into account passenger's riding time trips, and also his/her waiting at transfer points (interchange stations or stops). A novel...

Pedestrian trajectory prediction is essential for collision avoidance in autonomous driving and robot navigation. However, predicting a pedestrian's crowded environments non-trivial as it influenced by other pedestrians' motion static structures that are present the scene. Such human-human human-space interactions lead to non-linearities trajectories. In this paper, we new spatio-temporal graph based Long Short-Term Memory (LSTM) network pedestrian environments, which takes into account...

Industry 4.0 has transformed manufacturing with the integration of cutting-edge technology, posing crucial issues in efficient task assignment to multi-tasking robots within smart factories. The paper outlines a unique method decentralizing auctions handle basic tasks. It also introduces an improved variant Binary Particle Swarm Optimization (IBPSO) algorithm manage complicated tasks that require multi-robot collaboration. main contributions we make are: design auction decentralization...

Pedestrian detection plays an important role in many applications such as autonomous driving. We propose a method that explores semantic segmentation results self-attention cues to significantly improve the pedestrian performance. Specifically, multi-task network is designed jointly learn and from image datasets with weak box-wise annotations. The feature maps are concatenated corresponding convolution features provide more discriminative for classification. By learning detection, our...

Mobile edge computing (MEC) is a promising technology to provide high bandwidth and low latency shared services resources mobile users. However, the MEC infrastructure raises major security concerns when involve sensitive private data of This paper proposes novel blockchain-based key management scheme for that essential ensuring secure group communication among devices as they dynamically move from one subnetwork another. In proposed scheme, device joins subnetwork, it first generates...

Network pruning for deep convolutional neural networks (CNNs) has recently achieved notable research progress on image-level classification. However, most existing methods are not catered to or evaluated semantic segmentation networks. In this paper, we advocate the importance of contextual information during channel by presenting a novel Context-aware Pruning framework. Concretely, formulate embedded leveraging layer-wise channels interdependency via Guiding Module (CAGM) and introduce...

Detection of malicious programs using hardware-based features has gained prominence recently. The tamper-resistant hardware metrics prove to be a better security feature than the high-level software metrics, which can easily obfuscated. Hardware Performance Counters (HPC), are inbuilt in most recent processors, often choice researchers amongst metrics. However, lack determinism their counts, thereby affecting malware detection rate, minimizes advantages HPCs. To overcome this problem, our...

It has been well recognized that detecting road surface in a realistic environment is challenging problem also computationally intensive. Existing detection methods attempt to fit the into rigid models (e.g., planar, clothoid, or B-Spline), thereby restricting surfaces match specific models. In addition, curve-fitting strategies employed such techniques incur high computational complexity, making them unsuitable for in-vehicle deployments. this paper, we propose an efficient nonparametric...

The first-mile transportation provides a transit service using ridesharing-based vehicles, e.g., feeder buses, for passengers to travel from their homes, workplaces, or public institutions the nearest depots (rapid-transit metro appropriated bus stations) which are located beyond comfortable walking distance. This paper studies vehicle routing problem (VRP) transportation, aims at finding optimal routes fleet deliver doorstep depots, where can continue journeys fixed-route buses trains. We...

Graph Neural Networks (GNNs) have become the backbone for a myriad of tasks pertaining to graphs and similar topological data structures. While many works been established in domains related node graph classification/regression tasks, they mostly deal with single task. Continual learning on is largely unexplored existing continual approaches are limited task-incremental scenarios. This paper proposes strategy that combines architecture-based memory-based approaches. The structural driven by...

In vehicular edge computing, efficient strategies for model deployment and task offloading offer tremendous potential to reduce response time machine learning inference. However, existing works do not pay much attention that there are shared structures among different types of inference tasks. This limits the improvement in time. paper aims fill this gap by investigating a share-aware joint problem multi-task computing. We formulate with an objective minimize total all requests, under...

The use of smart indoor robotics services is gradually increasing in real-time scenarios. This paper presents a versatile approach to multi-robot backing crash prevention environments, using hardware schemes achieve greater competence. Here, sensor fusion was initially used analyze the state multi-robots and their orientation within static or dynamic scenario. proposed novel scheme-based framework integrates both scenarios for execution prevention. A round-robin (RR) scheduling algorithm...