『視覺譜記』基於空間型構法則視野量化工具之移動路徑與時間頻譜疊圖分析方法──以火場逃生行為研究為例

Visual Notation a Movement Route and Time Spectrum Mapping Approach of Visual Field Quantified Tool Based on Space Syntax-The Case Study of Evacuation Behavior in a Fire Scene

本文目的在於探究使用者視覺資訊變化與其逃生行為（如移動之快慢／停滯／繞道…等）兩者之關係。本研究以空間型構理論（Space Syntax）之視野量化工具Isovist為基礎，將受測者逃生模擬過程以固定時間段紀錄，使之成為具備時間因子的移動路徑再導入量化可見視野之分析參數-視覺範圍值（Visual Area）與可判讀是否繞道之分析參數－位移程度（Range），量化分析受測者之視野資訊變化與逃生移動行為模式，並佐以空間視覺資訊圖譜與行為移動戳記等記錄形式記錄分析之，得以檢視其逃生過程中移動速度變化、停滯和繞道行為。因而以『視覺譜記』為本文標題。本研究透過虛擬實境（Virtual Reality，VR）火場情境模擬實驗蒐集了54個有效樣本數據。初步推斷，受測者產生停滯與繞道地點之空間特徵為視覺範圍值（Virtual Area）與整體空間視覺相對互視程度（Relative Visibility, Visual Integration-HH）落差較大處，即雖位於視野搜尋量大之空間場域，但卻無法從局部區域判讀整體的空間組構，最終做出錯誤之逃生移動決策。

The purpose of this article is to explore the relationship between user's visual information difference and evacuation behavior, such as fast or slow movement, stagnation and detour. This research is based on the Space Syntax's visual field tool, Isovist. The simulation process is recorded in a fixed period of time, making it a route with time factor, and then we employ analytical parameter, Visual Area and the Range, quantitatively to analyze the subject's visual information and behavior patterns, to clarify stagnation and detour behavior during the process. This paper records and analyzes usage behaviors in the form of spatial visual information maps and behavioral movement stamps as well, so it is named as ''visual Notation''. In this study, 54 valid sample data were collected through the VR fire scene simulation experiment. It is preliminarily inferred that the spatial characteristics of the subject's stagnation and detour locations are the large gap between Virtual Area and Relative Visibility, Visual Integration-HH. Even the location is in the high visual search volume, it's hard to interpret the overall spatial configurations from local area, and ultimately make a wrong movement decision.