本文選題：去霧處理 + 無線視頻傳感系統(tǒng)�。� 參考：《南京郵電大學(xué)》2017年碩士論文

【摘要】：在缺乏電力和通信網(wǎng)絡(luò)基礎(chǔ)設(shè)施的野外環(huán)境下,傳統(tǒng)的視頻感知系統(tǒng)無法滿足應(yīng)用需求,無線視頻傳感器網(wǎng)絡(luò)自組織、低功耗等特點(diǎn)能很好的解決上述問題。然而,野外場景通�；\罩在霧霾天氣條件下,導(dǎo)致采集的視頻圖像模糊不清,難以從中獲取場景的細(xì)節(jié)信息。為此,本文設(shè)計(jì)實(shí)現(xiàn)了一個(gè)基于去霧處理的無線視頻傳感系統(tǒng)。首先,針對系統(tǒng)野外環(huán)境視頻采集的應(yīng)用場景要求,本文結(jié)合視頻圖像去霧和無線傳感網(wǎng)技術(shù)對系統(tǒng)進(jìn)行了需求分析,設(shè)計(jì)了系統(tǒng)整體架構(gòu)和節(jié)點(diǎn)體系結(jié)構(gòu),并詳細(xì)介紹了系統(tǒng)各功能模塊的軟硬件實(shí)現(xiàn)。其次,針對Tarel算法存在的缺陷,提出一種改進(jìn)的單幅圖像去霧算法。用全變分模型濾波代替原算法的中值濾波,在大幅度提高計(jì)算效率的基礎(chǔ)上保持了圖像邊緣特性;針對估計(jì)的大氣耗散函數(shù)在明亮區(qū)域失效的情況,提出了一種基于霧氣濃度特征的容錯(cuò)機(jī)制,提高了算法的普適性。實(shí)驗(yàn)對比結(jié)果表明,該算法能有效保持復(fù)原圖像的邊緣特性,恢復(fù)圖像清晰自然,較目前的主流去霧算法,去霧效果和效率都具有一定的優(yōu)勢。然后,在改進(jìn)的單幅圖像去霧算法基礎(chǔ)上,本文提出一種視頻快速去霧算法。根據(jù)視頻幀之間的關(guān)聯(lián)性,將有霧視頻序列按圖像組為單位分別進(jìn)行去霧,圖像組的關(guān)鍵幀復(fù)原采用上文所提的單幅圖像去霧算法,其余非關(guān)鍵幀基于運(yùn)動矢量進(jìn)行復(fù)原,這樣可以大大減少去霧冗余,保證了視頻去霧的實(shí)時(shí)性。實(shí)驗(yàn)結(jié)果表明,恢復(fù)的視頻圖像對比度和清晰度得到明顯改善,且視頻流能實(shí)時(shí)流暢輸出。最后,結(jié)合本文所提的視頻去霧算法,搭建和實(shí)現(xiàn)了原型系統(tǒng)。功能測試和性能測試的結(jié)果表明,本文設(shè)計(jì)的基于去霧處理的無線視頻傳感系統(tǒng)能滿足野外環(huán)境的應(yīng)用需求,并具有可移植性。
[Abstract]:In the absence of power and communication network infrastructure in the field environment, the traditional video sensing system can not meet the needs of the application, wireless video sensor network self-organization, low power consumption and other characteristics can solve the above problems. However, field scenes are usually enveloped in smog weather conditions, resulting in the blurred video images collected from which it is difficult to obtain the details of the scene. Therefore, a wireless video sensing system based on de-fogging is designed and implemented in this paper. First of all, according to the requirements of the application scene of video capture in the field environment of the system, this paper analyzes the requirements of the system by combining the video image de-fogging and wireless sensor network technology, and designs the overall architecture and node architecture of the system. The software and hardware implementation of each function module of the system is introduced in detail. Secondly, aiming at the defects of Tarel algorithm, an improved single image de-fogging algorithm is proposed. The median filter of the original algorithm is replaced by the total variational model filter, and the edge characteristic of the image is maintained on the basis of greatly improving the computational efficiency, and the estimated atmospheric dissipation function fails in the bright region. A fault tolerant mechanism based on fog concentration features is proposed to improve the universality of the algorithm. The experimental results show that the algorithm can effectively maintain the edge characteristics of the restored image and restore the image clear and natural. It has some advantages over the current mainstream de-fogging algorithm, the effect and efficiency of fog removal. Then, based on the improved single image de-fogging algorithm, a fast video de-fogging algorithm is proposed. According to the correlation between the video frames, the fogged video sequence is de-fogged according to the image group, the key frame of the image group is restored using the single image de-fogging algorithm mentioned above, and the rest of the non-key frames are restored based on the motion vector. In this way, the redundancy of de-fogging can be greatly reduced, and the real time of video de-fogging can be guaranteed. The experimental results show that the contrast and definition of the restored video images are improved obviously, and the video stream can be output smoothly in real time. Finally, the prototype system is built and implemented with the video de-fogging algorithm proposed in this paper. The results of function test and performance test show that the wireless video sensing system based on de-fogging in this paper can meet the requirements of field application and has portability.
【學(xué)位授予單位】：南京郵電大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP391.41;TP212.9;TN919.8

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 唐寧;趙鵬;吳紹啟;;基于FPGA的實(shí)時(shí)視頻圖像去霧系統(tǒng)[J];計(jì)算機(jī)工程與設(shè)計(jì);2016年11期

2 鞠銘燁;張登銀;紀(jì)應(yīng)天;;基于霧氣濃度估計(jì)的圖像去霧算法[J];自動化學(xué)報(bào);2016年09期

3 楊東瓊;;嵌入式Linux文件系統(tǒng)淺析[J];信息與電腦(理論版);2015年18期

4 李鵬;唐亮;凌虎;;基于大氣散射理論的視頻去霧算法的研究[J];山東工業(yè)技術(shù);2015年11期

5 甄海濤;聶洪淼;陳慶文;;基于嵌入式Linux的根文件系統(tǒng)制作與移植[J];自動化技術(shù)與應(yīng)用;2014年12期

6 張茁;;無線WiFi技術(shù)應(yīng)用現(xiàn)狀及發(fā)展分析[J];數(shù)字技術(shù)與應(yīng)用;2014年06期

7 沙超;張悅;王汝傳;;一種基于硬件編解碼的無線視頻節(jié)點(diǎn)設(shè)計(jì)與實(shí)現(xiàn)[J];信息化研究;2012年05期

8 張歡慶;高麗;宋承祥;;基于ARM的嵌入式Linux交叉編譯環(huán)境的研究與實(shí)現(xiàn)[J];計(jì)算機(jī)與數(shù)字工程;2012年02期

9 張怡;張加萬;楊國強(qiáng);李亮;;基于時(shí)空一致性優(yōu)化的視頻去霧算法[J];計(jì)算機(jī)應(yīng)用研究;2011年10期

10 熊丹;;常見的嵌入式操作系統(tǒng)[J];電子世界;2011年10期

，

本文編號：2021050