基於Faster R-CNN演算法的行人偵測應用研究與分析
No Thumbnail Available
Date
2025
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
本論文的研究動機在於物件偵測與追蹤的運作探討與原理分析，研究目的主要在於行人的影像偵測與追蹤上，了解現有的物件分類的演算法及數據庫，同時改良出新演算法以達到的較高的物件匹配度。本文中所改良的物件辨識演算法主要以Faster R-CNN為主，對行人影像目標進行物件追蹤，過程中也會與現有的演算法做分析比較取得研究的可行性與可靠度。
The motivation of research of this paper lies in the discussion and principal analysis of the operation of object detection and tracking. The research purpose is mainly on pedestrian image detection and tracking, to understand the existing object classification algorithms and databases, and to improve new algorithms to Achieved higher object matching degree. The improved object recognition algorithm in this article is mainly based on Faster R-CNN, which performs object tracking on pedestrian image targets. In the process, it will also be analyzed and compared with existing algorithms to obtain the feasibility and reliability of the research.
The motivation of research of this paper lies in the discussion and principal analysis of the operation of object detection and tracking. The research purpose is mainly on pedestrian image detection and tracking, to understand the existing object classification algorithms and databases, and to improve new algorithms to Achieved higher object matching degree. The improved object recognition algorithm in this article is mainly based on Faster R-CNN, which performs object tracking on pedestrian image targets. In the process, it will also be analyzed and compared with existing algorithms to obtain the feasibility and reliability of the research.
Description
Keywords
電腦視覺, 行人偵測, Faster R-CNN, 深度學習, Computer Vision, Pedestrian Detection, Faster R-CNN, Deep Learning