Search published articles


Showing 2 results for Traffic
Row/Column-First: A Path-based Multicast Algorithm for 2D Mesh-based Network on Chips
S. Kashi, A. Patooghy,
Volume 14, Issue 2 (6-2018)
Abstract
In this paper, we propose a new path-based multicast algorithm that is called Row/Column-First algorithm. The proposed algorithm constructs a set of multicast paths to deliver a multicast message to all multicast destination nodes. The set of multicast paths are all of row-first or column-first subcategories to maximize the multicast performance. The selection of row-first or column-first approaches is done based on the location of multicast source node i.e., how the multicast source is far from right/left and top/bottom margins of the mesh network. In this way, the proposed algorithm improves two performance criteria i.e., traffic and communication latency as compared with the well-known Column-Path multicast algorithm. In order to evaluate the proposed algorithm, an analytical model is developed to estimate the mentioned performance criteria. The modeling and simulation results show improvement of 10 and 20 percent on traffic and communication latency respectively for Row/Column-First algorithm as compared to the Column-Path algorithm.

Efficient Analysis of Traffic Intersection Scenes by Employing Traffic Phase Information
P. Ahmadi, I. Gholampour,
Volume 15, Issue 2 (6-2019)
Abstract
Analyzing motion patterns in traffic videos can be employed directly to generate high-level descriptions of their content. For traffic videos captured from intersections, usually, we can easily provide additional information about traffic phases. Such information can be obtained directly from the traffic lights or through traffic lights controllers. In this paper, we focus on incorporating additional information to analyze the traffic videos more efficiently. Using side information on traffic phases, the semantic of motion patterns from traffic intersection scenes can be learned more effectively. The learning is performed based on optical flow features extracted from training video clips, and applying them to supervised topic models such as MedLDA and MedSTC. Based on such models, any video clip can be represented based on the learned patterns. Such representations can be further exploited in scene analysis, rule mining, abnormal event detection, etc. Our experiments show that employing side information in intersection video analysis leads to improvement in discovering scene pattern. Moreover, supervised topic models achieve about 4% improvement in abnormal event detection, compared to the unsupervised ones, in terms of area under ROC.

Page 1 from 1     
Creative Commons License
© 2022 by the authors. Licensee IUST, Tehran, Iran. This is an open access journal distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license.