Abstract: A wavelet-based algorithm is proposed for image contrast enhancement, which distinguishes the HF coefficients due to signal from those due to noise with the correlation between wavelet planes, then, increases wavelet coefficients due to signal and suppresses those due to noise. The algorithm solves the problem that some conventional enhancement methods enhance noise visibility at the same time enhance signal visibility. Experience results show that the algorithm can achieve an excellent balance between the enhancement of subtle image detail and the avoidance of noise amplification, and enhance often weak signal features against a background of relatively high noise.