编辑: 我不是阿L | 2019-07-16 |
Innovation Ordnance Survey Romsey Road Southampton SO16 4GU United Kingdom clare.
[email protected] Abstract Cognitive models of orientation using aerial-view 2D diagrams such as maps could help us to predict where their contents and design help, or hinder, cognitive processes, resulting in more usable map designs for situations where orientation is a common problem. However, the complex realities of geographic space, and the large scale transformations required with maps, make it less easy to model even basic processes such as orientation. This paper sets the context for ongoing work at Ordnance Survey, Great Britain'
s national mapping agency, to attempt to apply and extend cognitive modeling in order to enable (and ultimately perhaps to automate) decisions about geographic information provision for assisting real-world orientation tasks. Introduction A very common event , for example when traveling, occurs where we find ourselves viewing a real-world scene - either in an unfamiliar place, or by viewing a photograph or other image - and comparing this scene with a map or other information representation. It is often the case in this situation that we already know approximately where we are on the map. We may already know the name of the location - perhaps it is supplied with the photograph, e.g. in a guidebook or on a website - or we may have just emerged from a subway or rail station, or alighted from a bus or tram. We may even have walked there ourselves, without needing to refer to a map until this point. In any case, the next question is Which way am I - or is the camera - facing? In this paper, this task of matching the reality to the map will be referred to as the orientation task. Why This Problem? Although in many cases the map user in this situation may be trying to navigate through the landscape, this problem does not only arise during wayfinding. It may also occur when: ? Crown copyright 2005. Reproduced by permission of Ordnance Survey. o trying to identify a specific building or object which is not explicitly labeled on the map, e.g. to visit or study it, or in an emergency scenario;
o trying to match a historic image (e.g. of an old street scene) to a modern-day map;
o trying to match a historic map to a modern-day scene or image;
o making planning decisions based partly on viewing the current visual landscape;
o trying to orientate with respect to cardinal directions in the absence of local cues, e.g. to establish sun and wind direction;
o trying to understand the natural features of a landscape, with respect to the '
bigger picture'
given in a regional map;
o trying to decide on relative distances and directions to unseen distant locations (whether or not one intends to navigate to them). Therefore, from urban and rural planning to risk management scenarios and recreational tourism, people need to be able to match 2D and 3D representations. As a national mapping agency, and along with the rest of the geographic information industry, Ordnance Survey is increasingly moving towards a future where 2.5D and 3D representations of the world around us will be commonplace. So should we or anyone worry anymore about conventional 2D cartographic design? We believe that the orientation problem outlined above will still be relevant into the future because: o 2D mapping is inherently more portable (e.g. it can be printed, scanned and copied more easily). o The aerial view of a 2D map gives the '