多組隨機特徵物件追蹤演算法

Abstract

本論文提出一個同時使用多組隨機影像特徵的追蹤演算法。主要改良壓縮追蹤中時好時壞的追蹤成果。在壓縮追蹤中，所使用的影像特徵透過隨機投影而產生。產生的影像特徵受隨機亂數影響以至於每次執行結果皆有所不同。倘若沒有擷取到目標明顯的特徵，追蹤器便有可能追蹤失敗。因此每次執行的追蹤結果並不一致。本論文所提出的演算法使用多個不同的影像特徵進行追蹤，並透過計算與目標模型的相似度選擇最佳的追蹤結果，減少由不佳的影像特徵決定目標位置的機會。在本論文中比較了各式不同判斷追蹤結果的方法，如：樸素貝氏分類器、相位相關法和巴氏係數。實驗結果顯示使用巴氏係數選擇最理想的追蹤結果，能夠大幅地降低追蹤誤差。與僅使用單一組特徵進行追蹤的演算法相比，中心位置誤差最多可由63.62像素減少至15.45像素，邊界盒重疊比例可由31.75%提升至64.48%，而追蹤成功率在最佳情況下，則可由38.51%增加至82.58%。

This thesis proposes an object-tracking algorithm with multiple randomly-generated image features. We mainly improve the tracking performance which is sometimes good and sometimes bad in compressive tracking. In compressive tracking, the image features are generated by random projection. The resulting image features are affected by the random numbers so that the results of each execution are different. If the obvious features of the target are not captured, the tracker is likely to fail. Therefore the tracking results are inconsistent for each execution. The proposed algorithm in this thesis uses a number of different image features to track, and chooses the best tracking result by measuring the similarity with the target model. It reduces the chances to determine the target location by the poor image features. In this thesis, we compare various methods of judging the tracking results, such as the Naive Bays classifier, the phase correlation method and the Bhattacharyya coefficient. The experimental results show that choosing the best tracking result by the Bhattacharyya coefficient can greatly reduce the tracking errors. The best performance improvements in terms of center location error, bounding box overlap ratio and success rate are from 63.62 pixels to 15.45 pixels, from 31.75% to 64.48% and from 38.51% to 82.58%, respectively.