結合生物反饋之新世代腦機介面及其在移動載具控制之應用---子計畫四：以機器視覺為基礎之人意向移動載具獲知與追蹤(I)

Abstract

本計畫以機器視覺為基礎之人意向移動載具獲知與追蹤課題。腦波生理 訊號，通常受試驗者的視覺、聽覺、感受等影響，本子計畫的目的，在於利 用視覺狀態回饋給使用者，探討其對的人腦波訊號之影響。利用本計畫結 果，子計畫三可據此鑑別岀重要視覺腦波生理訊號的位置，即頻道，及訊號 強弱變化，一起與子計畫五方向感腦波訊號、子計畫二注意力腦波訊號，整 體訓練學習後，能提昇在移動載具的控制效率。本子計畫執行所需之移動載 具，擬以電動代步車為實驗平台。第一年將探索在不同環境 (包括影子干 擾，高低照度情況，車顏色、人衣服、背景相近)等情況下，移動載具車之 視訊獲知方法。第二年將研究移動載具車之視訊空間定位與移動辨識方法， 提供受試驗者自然地交互作用，本計畫之輸出，可以語意式描述，如右前方 快速前進、慢速左轉、慢速右轉後退等文字型式表達，或以運動向量(向量 方向代表載具車方向向量長度代表載具車速度)於面板顯示。第三年將推廣 至任何視角均可適用之移動載具車的視訊空間定位與移動辨識方法。本子計 畫屬無所不在計算領域 (Ubiquitous Computing)，典型計畫如聞名於全世界 MIT 活氧計畫。

This project is concerned with machine vision-based moving vehicle capture and tracking. It is known that brain signals are dependent on the excitations of vision, audio, and feeling of a tester. The purpose of this project is to provide various feedback to a user and then investigate its effect on brain signaling. With an excitation, subproject III will identify the responsive brain channel and its signaling variation. The brain signaling variations of a person invoking a multi-task distraction and spatial disorder are studied by subprojects II and V, respectively. With these identified responsive brain signaling, multi-mode bio-feedback embedded in brain-computer interface formulation can reach a new era. The test-bed of moving vehicle to be used in this subproject is an electrical mini-car with rechargeable battery. The first year will dedicate to moving vehicle capture under various environment conditions. Good and robust image segmentation of vehicle will be focused on under casting shadow, lighting variation, and color closeness between object and background. The second year will investigate vision-based positioning and movement recognition of moving vehicle for BCI sensing and validation. Through car’s key shape and a learned fuzzy rule base for movement classification, either linguistic or numerical statement can be outputted to bio-sensing BCI loop. Finally, view-invariant movement recognition of vehicle will be studied in the third year. View-invariance classification adds more flexibility to application exploitation, so do to this project. This subproject belongs to the area of ubiquitous computing, a famous example like the “oxygen” project of MIT.