基于协作稀疏编码的红外小目标检测

为了实现复杂背景下的红外小目标检测, 提出了一种基于协作稀疏编码(CSC)的红外小目标检测算法。首先通过滑动窗口法提取待测 试图像的图像块,并将 其转化为列向量作为超完备字典；然后采用CSC模型计算每一个图像块在超完 备字典中的系数矩 阵以及误差矩阵,其中系数矩阵的L_2,1范数代表图像的背景信息,而误 差矩阵的L_1,2范数代表红外小目标信 息；进而利用ADMM(alternating directional method of multiplier)算法解 算,得到系数矩阵和误差矩阵；最后通 过误差矩阵重建,得到红外小目标的位置。仿真及公开数据实验结果,证实了本文方法的有 效性。

Infrared small target detection based on collaborative sparse coding

In order to realize the detection of i nfrared small target in complicated background,this paper proposes an infrared small target detection algorithm based on collaborative spa rse coding (CSC).The sub-image blocks are extracted from the testing image through sliding window from top and left to bottom and right,and then each sub-iamge block is transformed to a column vector,which is used as over-compl ete dictionary.The corresponding coefficients matrix and error matrix are calculated with the const ructed over-complete dictionary using the proposed collaborative coding model. Then,the coefficients matrix and error matrix are obtained using alternating di rectional method of multiplier (ADMM).And finally the location of the infrared small target is obtai ned through reconstructing the error matrix.Experiments are executed on both the synthetic and public data sets ,and comparisons with the state-of-the-art methods show that the proposed met hod is more effective and robust.