基于脊波变换的一种红外图像增强技术

在研究红外图像特性的基础上,提出了一种基于多尺度脊波变换的红外图像增强方法。脊波变换特别适合于具有直线或超平面奇异性的高维信号的描述,在导引头红外图像处理中使用脊波变换来检测图像的边缘和跟踪是一种有效的方法,而且具有较高的逼近准确度。首先,利用脊波变换技术对红外图像进行处理。然后,对脊波变换系数进行增强。最后用脊波逆变换对红外图像重构。结果表明,该方法对导引头的红外自动目标识别是快速和有效的。

A Technique for Enhancing Infrared Image Based on Ridgelet Transforms

On the basis of analyzing the characteristics of infrared imaging,a method for infrared image enhance based on mutilscale ridgelet transforms is given.Ridgelet transform is especially suitable for the describing of those signals that have linear or super plane singularities and much better approximation.Ridgelet transform is an effective method for image edge detection and tracking,and has much higher accuracy for seeker infrared image processing.First,infrared image is decomposed by ridgelet transforms.Second,the ridgelet transform coefficients are enhanced.Finally,infrared enhancing image is reconstructed by inverse ridgelet transform.Experiments show that the method is a rapid and effective for infrared seeker automatic target recognition.