本文選題：無線傳感器網(wǎng)絡 + Android��； 參考：《安徽理工大學》2014年碩士論文

【摘要】：無線傳感器網(wǎng)絡是目前最活躍的科技領域之一,具有由體積小,數(shù)據(jù)采集、數(shù)據(jù)傳輸、數(shù)據(jù)處理功能的的傳感器節(jié)點組成,在日常生活,工業(yè)生產,學校科研方面都有廣泛的應用前景。隨著計算機視覺和Android系統(tǒng)的應用發(fā)展,加上芯片設計、制造工藝的進步,將Android系統(tǒng)移植到嵌入式設備里已成為趨勢。Android系統(tǒng)具有強大的人機交互能力,優(yōu)良的顯示效果和炫麗的圖形界面,但自身圖像處理、目標跟蹤方面的能力相對有限。計算機OpenCV視覺庫含有很多通用的算法能高效地完成這一任務,而且也得到Android系統(tǒng)的支持,故在Android系統(tǒng)上使用OpenCV成為可能。 運動目標檢測是目標跟蹤的前提條件,目前,在圖像處理及計算機視覺研究領域已有許多目標跟蹤算法,如經(jīng)典的Meanshift,質心算法,基于模板匹配的跟蹤算法等,這些算法對目標識別有可靠性但是目標跟蹤方面的效果差強人意。TLD目標跟蹤算法有三部分組成,即跟蹤器、學習過程和檢測器,采用跟蹤和檢測相結合的策略,是一種自適應的、可靠的跟蹤技術。 本文主要設計了一種作為無線傳感器網(wǎng)絡節(jié)點的Android嵌入式設備,實現(xiàn)目標跟蹤識別的功能。主控芯片是三星公司的S3C6410,在其上搭建了Android OpenCV開發(fā)包,使用TLD目標跟蹤算法,使節(jié)點的數(shù)據(jù)采集的有效性,可交互性都得到了提高。節(jié)點之間的數(shù)據(jù)傳輸是采用Wi-Fi互聯(lián)技術,節(jié)省了布線費用,同時短距離傳輸可靠性也能保證。最后試驗了設備對目標識別跟蹤的效果,證明了本系統(tǒng)識別的有效性。
[Abstract]:Wireless sensor network is one of the most active fields of science and technology at present. It is composed of sensor nodes with small size, data acquisition, data transmission and data processing functions. The school scientific research has the extensive application prospect. With the development of computer vision and Android system, as well as the development of chip design and manufacturing technology, transplanting Android system to embedded devices has become a trend. Excellent display effect and dazzling graphical interface, but its own image processing, target tracking ability is relatively limited. Computer OpenCV vision library contains many common algorithms to accomplish this task efficiently, and it is also supported by Android system, so it is possible to use OpenCV on Android system. Moving target detection is the precondition of target tracking. At present, there are many target tracking algorithms in the field of image processing and computer vision research, such as classical Meanshift, centroid algorithm, template matching based tracking algorithm and so on. These algorithms have reliability for target recognition but the effect of target tracking is unsatisfactory. TLD target tracking algorithm consists of three parts: tracker, learning process and detector, and adopts the strategy of combining tracking and detection. Is an adaptive, reliable tracking technology. In this paper, a Android embedded device is designed as the node of wireless sensor network to realize the function of target tracking and recognition. The main control chip is Samsung S3C6410. the Android OpenCV development kit is built on the chip, and the TLD target tracking algorithm is used to improve the validity and interactivity of the node data acquisition. The data transmission between nodes is based on Wi-Fi technology, which saves the wiring cost and ensures the reliability of short distance transmission. Finally, the effect of the equipment on target recognition and tracking is tested, and the effectiveness of the system is proved.
【學位授予單位】：安徽理工大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：TP212.9;TN929.5

【參考文獻】

相關期刊論文 前10條

1 常丹華;楊冬冬;韓夏;;OpenCV在智能監(jiān)控方面的應用研究[J];電視技術;2009年09期

2 顏振亞;鄭寶玉;林志偉;;無線傳感器網(wǎng)絡中機會協(xié)作傳輸及其性能研究[J];電子與信息學報;2009年01期

3 趙鑫;朱一群;;基于Android智能手機的智能家居監(jiān)控系統(tǒng)設計與實現(xiàn)[J];硅谷;2013年16期

4 賴小波;朱世強;曹新星;;基于OpenCV的焊縫跟蹤系統(tǒng)中攝像機標定方法[J];焊接學報;2010年07期

5 賈小軍;喻擎蒼;;基于開源計算機視覺庫OpenCV的圖像處理[J];計算機應用與軟件;2008年04期

6 劉純勤,李安濤;IEEE802.11無線局域網(wǎng)標準[J];通信技術;1998年01期

7 吳永浩;華云松;尚亮;;基于OpenCV的運動目標跟蹤系統(tǒng)研究[J];電子測試;2010年08期

8 蔡葉菁;計算機視覺[J];中國包裝工業(yè);2002年05期

9 袁寶紅;張德祥;張玲君;;基于OpenCV的視頻運動目標檢測與跟蹤[J];計算機系統(tǒng)應用;2013年05期

10 周鑫;錢秋朦;葉永強;王從慶;;改進后的TLD視頻目標跟蹤方法[J];中國圖象圖形學報;2013年09期

，

本文編號：1789713