红外图像目标特征提取方法

针对实际红外空中图像中目标的大小,方向等发生变化等情况,基于奇异值分解的特征向量稳定性不好。为解决这一技术难题,提出了基于尺度奇异值变换的红外图像目标特征提取方法。该方法是一种基于尺度变换的矩阵方法。同类目标图像样本经过上述变换后所得到的矩阵都具有相似的能量,用文中提出的矩阵表示方法对红外图像进行描述后,对其应的奇异值进行特征提取,克服了对原始图像的矩阵进行奇异值提时存在的缺点,实验表明该方法提取目标矩阵的奇异值特征向量具有较好的稳定性,对图像的平移,旋转和比例变化具有良好的不变性,是一种实用的特征提取方法。

Target feature extraction method of infrared image

The stability of feature vectors based on SVD decomposition is very low for the variation of target′s size and orientation. On this problem mentioned above, a new target feature extracting method is presented. It is a matrix describing method based on scale transformation. These matrixes of all same kind of target samples getting through transformation mentioned above have the similar energy. The matrix method proposed extracting target′s features from SVD feature values can overcome shortcomings of extracting SVD feature directly from matrix of the images. The experimental result shows this method has favorable stability to target matrix′s SVD feature vectors, and high invariant property to displacement, rotation and scale. It is a practical feature extracting method.