【摘要】：在視覺目標(biāo)跟蹤系統(tǒng)中,特征的表達(dá)和提取是重要的組成部分.本文提出基于多個自編碼機(jī)網(wǎng)絡(luò)相聯(lián)合的特征提取機(jī),通過對輸入數(shù)據(jù)進(jìn)行一定程度的重組,采用深度學(xué)習(xí)的理論對其局部特征進(jìn)行描述并對結(jié)果進(jìn)行聯(lián)合決策.結(jié)合該網(wǎng)絡(luò)結(jié)構(gòu),本文提出一種融合局部特征的深度信息進(jìn)行目標(biāo)跟蹤的算法.將輸入圖像分塊使得大量的乘法運(yùn)算轉(zhuǎn)化為加法和乘法的混合運(yùn)算,相對于全局的特征表達(dá),大幅降低了運(yùn)算復(fù)雜度.在跟蹤過程中,目標(biāo)候選區(qū)的各分塊權(quán)重能夠根據(jù)相應(yīng)網(wǎng)絡(luò)的置信度進(jìn)行自適應(yīng)的調(diào)整,提升了跟蹤器對光照變化、目標(biāo)姿態(tài)和遮擋的適應(yīng).實(shí)驗(yàn)表明,該跟蹤算法在魯棒性和跟蹤速度上表現(xiàn)優(yōu)秀.
[Abstract]:The representation and extraction of features is an important part of visual target tracking system. In this paper, a feature extraction machine based on multiple self-coding machine networks is proposed. By reorganizing the input data to a certain extent, the local features are described by using the theory of depth learning and the results are jointly determined. Combined with the network structure, this paper proposes an algorithm for target tracking based on the depth information of local features. When the input image is partitioned into blocks, a large number of multiplication operations are transformed into mixed operations of addition and multiplication. Compared with the global feature representation, the computational complexity is greatly reduced. In the tracking process, each block weight of the target candidate area can be adjusted adaptively according to the confidence degree of the corresponding network, and the adaption of the tracker to the illumination change, target attitude and occlusion is improved. The experimental results show that the proposed tracking algorithm performs well in robustness and tracking speed.
【作者單位】： 空軍工程大學(xué)信息與導(dǎo)航學(xué)院;解放軍第93716部隊(duì);
【基金】：Supported by National Natural Science Foundation of China(61473309) Natural Science Foundation of Shaanxi Province(2015JM6269,2016JM6050)
【分類號】：TP391.41

【相似文獻(xiàn)】

相關(guān)期刊論文 前5條

1 趙蘊(yùn)龍;王博識;張凱;董釗;張磊;;充分考慮節(jié)點(diǎn)編碼機(jī)會的編碼感知路由協(xié)議[J];應(yīng)用科學(xué)學(xué)報(bào);2014年01期

2 王曉鳴;TST700袖珍式話音編碼機(jī)[J];通信保密;1991年02期

3 馬義德,張宗念,余英林;H.263+部分可選模式的研究[J];蘭州大學(xué)學(xué)報(bào);2000年04期

4 姜波，，胡堅(jiān);一個有趣而全新的無線世界即將來臨[J];國外科技動態(tài);1994年12期

5 ;[J];;年期

相關(guān)會議論文 前2條

1 羅慶生;韓寶玲;吳瑞松;;液晶顯示技術(shù)在新型多功能噴墨編碼機(jī)中的應(yīng)用[A];'99全國包裝與食品加工技術(shù)研討會論文集[C];1999年

2 張小衛(wèi);;中國噴墨編碼機(jī)市場概況[A];'99全國包裝與食品加工技術(shù)研討會論文集[C];1999年

相關(guān)碩士學(xué)位論文 前2條

1 肖巍;無線網(wǎng)絡(luò)中基于即時譯碼網(wǎng)絡(luò)編碼的研究[D];南京郵電大學(xué);2016年

2 代思;無線網(wǎng)絡(luò)中編碼機(jī)會發(fā)現(xiàn)方法的研究[D];寧波大學(xué);2014年

本文編號：2204069