Browsing by Author "Bandara, RMPNS"

Now showing 1 - 2 of 2

item: Conference-Abstract
Automatic road extraction form high resolution satellite images
(Department of Civil Engineering, University of Moratuwa., 2013-07) Bandara, RMPNS; De Silva, PCP; Bandara, JMSJ; Pasindu, HR
The presence of high resolution satellite images and their potentials to be used in many fields such as urban planning, transportation engineering etc ,especially in the meaning of preparing and updating maps, have made the automatic extraction of objects, a new challenge in remote sensing. Automatic road extraction, one of major uses of preparing and updating maps, provides means for creating, maintaining, and updating transportation network, which subsequently offers databases for all means of traffic management. Moreover, automatic road extraction is a critical feature for an efficient use of remote sensing imagery in most contexts, which has been an active research area in computer vision and digital photogrammetric for over past decades. Further, the pixel-oriented analysis of satellite data has a main limit: the acknowledgement of semantic low level information, as the amount of energy emitted from the pixel, where the context does not assume any role. Conversely, the application of object-oriented image analysis on very high resolution data allows obtaining, by an automatic or semi-automatic analysis – with a minimal manual participation – a good classification also in presence of high and very high resolution data of small cities, where higher is an error possibility. Object-oriented image classification involves identification of image objects, or segments, that are spatially contiguous pixels of similar texture, color, and tone. A simplified methodology using the object oriented image analysis for automatic road extraction for the Colombo City Area is presented in this paper. The proposed object-oriented image classification method comprises few fundamental and important steps towards content analysis and image understanding for instant image segmentation and classification. Few algorithms and techniques for the segmentation and classification in order to identify road features from satellite images were also supported to the proposed method.
item: Conference-Abstract
Space syntax and mobile GIS application: investigation of relationship between human walking pattern and spatial configuration
(Department of Civil Engineering, University of Moratuwa., 2014-08) Jayasinghe, AB; Bandara, RMPNS; Pasindu, HR
The process of how people recognize spatial configurations and use them is an important subject for understanding of human walking pattern. Understanding of human walking pattern play a vital role in pedestrian pathway design and city street network design. In this context this study is to investigate the relationship between Mobile GIS is a newly introduced method that facilitate to collect much reliable and accurate data instead of conventional methods and space syntax enables the advance analysis in relation to space. In this study first actual human walking pattern were captured by using Mobile GIS application. Then spatial configuration and visibility of the study area were analyzed by using UCL Depth map software. Then visual analysis and spatial correlation were employed to identify the relationship between human walking pattern and spatial configuration. Results indicates that human walking pattern has significant correlation with spatial configuration (78%) and visibility of the space (62%).