應用於多干擾環境下之FPGA視覺化監控系統

在視覺化監控系統中，如何能快速且準確地偵測到移動目標物件是一個非常重要的步驟。不論是物件追蹤、物件辨識、事件偵測或行為分析等程序都高度仰賴移動物件偵測的精準程度。目前雖然已有許多方法可以偵測移動物件，但大多以個人電腦或嵌入式平台來實現。然而能夠做到精準的移動目標物偵測、追蹤與辨識，其演算法往往具備極高的運算複雜度，因此以軟體方式實現移動目標物偵測演算法往往難以達到即時（real time）運算處理的速度。為此，本論文提出以FPGA（場效可程式化邏輯閘陣列）來實現即時之視覺化監控系統。所設計系統使用Altera DE2-70發展板並以Verilog HDL來進行設計。此外本系統採用具備VGA（640 × 480）解析度的ALTERA D5M數位相機模組作為視訊擷取裝置。由於VGA解析度合於標準螢幕輸出規範，故可將處理結果直接將輸出於一般顯示裝置，極為便利。為了使所提出之視覺化監控系統及演算法能廣泛地適應於各種環境場合，我們將此系統應用在多干擾的天候環境中，排除外在環境，諸如下雨以及光源強度變化的干擾，以提高偵測之準確性。因此我們在本論文中提出了「多干擾偵測」、「陰影偵測」等演算法，來改善因為外在環境因素所導致系統發生誤判的窘境。經由實驗證明，本系統在多種干擾環境下仍能有效地偵測移動目標物件。

關鍵字

多干擾環境；雨水干擾移除演算法；視覺化監控系統；移動物件偵測；時序相減法；場效可程式化邏輯閘陣列

並列摘要

The detection of moving objects is the first step for object tracking, objects recognition, event detection as well as behavior analysis. Therefore, the capability of detecting moving objects in a quickly and accurately manner is very important in visual surveillance systems. So far, most of the moving object detection algorithms are implemented by using desktop computers or embedded platforms; however, it is hard to meet the real time requirement in visual surveillance system by using software-oriented approaches when a highly complex moving object detection algorithm is used. For this, we propose in this dissertation a real-time visual surveillance system based on ALTERA DE2-70 FPGA module. The hardware circuit, i.e., the proposed moving object detection algorithm, is developed with Verilog HDL. Besides, the ALTERA D5M module, a digital camera with VGA (640x480) resolution, is used for the purpose of video capture. As the VGA resolution is a standard format for monitor display, the system output, i.e., processed sequences, can be connected to any display very easily. In order that the proposed visual surveillance system can adapt itself to various occasions, we propose in this dissertation an algorithm that can exclude the interference of environment, such as rainfall interference, changes of light intensity, shadow interference, etc. As can be seen in our experiments, the proposed system can function very well under multi-interference environment, which justifies the usefulness of the proposed real-time moving object detection system.

並列關鍵字

Multi-interference environment ； Rainfall interference removal ； Visual surveillance system ； Moving object detection ； temporal difference ； FPGA

參考文獻

[2] A. Tripathi and S. Mukhopadhyay, “Video post processing: low-latency spatiotemporal approach for detection and removal of rain,” IET Image Process., no. 2, Mar. 2012, pp. 181-196.

[3] E. Arbel and H. Hel-Or “Shadow removal using intensity surfaces and texture anchor points”, IEEE Trans. Pattern Anal. Mach. Intell., vol. 33, no. 6, 2011, pp.1202 -1216.

[4] L. W. Kang, C. W. Lin, and Y. H. Fu, “Automatic single-image-based rain streaks removal via image decomposition,” IEEE Trans. Image Process., vol. 21, no. 4, Apr. 2012, pp. 1742–1755.

[5] S. Nadimi and B. Bhanu, “Physical models for moving shadow and object detection in video,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 26, Aug. 2004, pp. 1079-1087.

[6] N. Martel-Brisson and A. Zaccarin, “Learning and removing cast shadows through a multidistribution approach,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 29, no. 7, July 2007, pp. 1133-1146.

國際替代計量

應用於多干擾環境下之FPGA視覺化監控系統

全文下載

主題瀏覽