Image enhancement is a useful application to provide more explicit visual interpretation for the human eye, and a better understanding of digital imagery which is particularly useful in satellite images captured from remote sensing devices. In this correspondence, a recursive image enhancement technique is proposed; combining anisotropic and stationary wavelet transform. The estimated low-frequency and high-frequency subbands are interpolated and added recursively, to generate a new enhanced image with sharpened edge details. The proposed system is evaluated with famous wavelet families namely Haar, Daubechies-4, Symlet-4, and Coiflet-4; on randomly selected satellite images and noisy images. Subjective and quantitative methods are used to evaluate the proposed system.