Toward A Location Based Service for Assessing and Recommending Landscape Views

Abstract

GIScience 2016 Short Paper Proceedings Toward A Location Based Service for Assessing and Recommending Landscape Views S. Soleimani 1 , M. R. Malek 1 , G. Sinha 2 Geodesy & Geomatics Department, K.N.Toosi University of Technology, Tehran, Iran Email: {ssoleimani; mrmalek}@kntu.ac.ir Department of Geography, Ohio University, Athens, Ohio, 45701, USA Email: sinhag@ohio.edu Abstract The rationale and initial design considerations are presented for a location based service (LBS) that will offer people personalized recommendations for aesthetically desirable places. This will be based on people’s personal characteristics and environmental factors known to affect landscape perception. A prototype user-interface has been designed for eliciting users’ perceptions of landscape elements and colors they perceive from photographs of landscape views. In the long-term, a VGI repository database will be maintained to make personalized location specific recommendations about suitable landscape views for user-specified locations. A prototype fuzzy logic recommendation system comprising 72 rules has also been designed to model how personal factors (gender, age, health status) and perceived colors likely determine the aesthetic suitability of landscape views. An initial trial revealed some limitations and stressed the need for further testing of the feasibility of fuzzy logic as the reasoning framework for this LBS for recommending suitable landscape views. 1. Introduction As urbanity imposes psychological demands that become excessive, people seek stress- reducing (especially natural) environments where they can recreate and relax. The investigation of the ecological and visual aspects of physical landscapes is critical for understanding and addressing many of the psychosomatic disorders of urban dwellers. In this paper, we focus on visual perception of people’s everyday environments. Human-beings engage in a wide variety of important decisions in their everyday lives, based on visual properties of the geographic environment and their geographical knowledge. Examples include coping with environmental visual pollution, finding one’s way in spatially extended areas, and finding a landscape view appropriate to one’s emotional status. Color, shape, composition, and configuration of objects are important components of visual stimuli that should be studied to understand our perceptions of and reactions to physical landscapes in urban environments. In this short paper, we present the rationale and initial design considerations for a prototype Location Based Service (LBS) for landscape view assessment. This LBS will ultimately offer people personalized recommendations for places (near their chosen location(s)) that are most likely to appeal to them aesthetically. The LBS is designed to make such personalized recommendations based on available knowledge of how people’s personal characteristics affects landscape perception. A prototype user-interface has been designed for eliciting users’ perceptions about the types of landscape elements and the colors they perceive in images showing landscape views. An important goal of this project is to maintain a large database of people’s perception of landscape views, based on volunteered geographic information (VGI) generated from people tagging images uploaded by them and/or others.