本文選題：高光譜圖像　切入點：目標探測　出處：《紅外與毫米波學(xué)報》2017年04期

【摘要】：提出一種用于高光譜圖像小目標探測的目標光譜學(xué)習(xí)算法,目的是從圖像中學(xué)習(xí)得到一條更為準確的目標光譜,從而提高有監(jiān)督目標探測的效果.該算法由基于自適應(yīng)權(quán)重的目標光譜學(xué)習(xí)算法和自完備字典兩部分組成.前一部分內(nèi)容是在已有完備的背景字典的情況下,通過稀疏編碼和梯度下降算法來優(yōu)化學(xué)習(xí)目標光譜;后一部分內(nèi)容通過背景字典的不斷擴充來確保該字典的完備性,從而保證了學(xué)習(xí)算法的準確性.仿真和實際高光譜數(shù)據(jù)的實驗結(jié)果表明,所提出的方法能有效地提取出準確的目標光譜,從而顯著提高目標探測算法的精度.
[Abstract]:Put forward a target for the spectral learning algorithm for small target detection of hyperspectral image from the image, to learn a more accurate target spectrum, thereby improving the supervised target detection results. The algorithm consists of target spectral learning algorithm based on adaptive weight and complete dictionary is composed of two parts. The first part in the background there is complete dictionary of the case, to optimize the learning target spectrum by the sparse encoding and gradient descent algorithm; part after by expanding background to ensure the completeness of the dictionary from the dictionary, and ensure the accuracy of the learning algorithm. The simulation and experimental results on real hyperspectral data show that the proposed method can effectively extract the target spectrum accurately, so as to improve the accuracy of target detection algorithm.

【作者單位】： 復(fù)旦大學(xué)電磁波信息科學(xué)教育部重點實驗室;復(fù)旦大學(xué)信息學(xué)院智慧網(wǎng)絡(luò)與系統(tǒng)研究中心;
【基金】：國家自然科學(xué)基金(61572133) 北京師范大學(xué)地表過程與資源生態(tài)國家重點實驗室開放基金(2015-KF-01)~~
【分類號】：TP751

【相似文獻】

相關(guān)期刊論文 前10條

1 石大蓮;呂群波;崔燕;高靜;袁艷;黃e，

本文編號：1661289