Application of blind deconvolution approach with image quality metric in underwater image restoration

The degradation of underwater image obtained by range gated pulsed laser imaging system is mainly caused by scattering and absorption of in-water particles. Although the use of blind deconvolution of image restoration is a widely known concept, little literatures have discussed in detail its application in the problem of restoration of underwater images. In order to extend the imaging distance, we use blind deconvolution approach together with Wells' small angle approximation theory to restore the underwater image to a better level. Due to underwater no-reference imaging environment, image quality judgment based on blur metric method is incorporated in our approach to determine the appropriate deconvolution iteration time, which objectively evaluate the image restoration result.

I. Introduction

Images are obtained in areas ranging from everyday photography to astronomy, remote sensing, medical imaging, and microscopy. In each case, there is an underlying object or scene we wish to observe. Yet the observation process is not perfect: there is uncertainty in the measurements, occurring as blur, noise, and other degradations in the recorded images. The situation is more severe in underwater imaging using laser as light source: the degradation is caused both by light propagation in water and the optical imaging system design including lenses and recording devices. Rapid decay of signals due to absorption leads to poor SNR, strong backscatter by the water itself together with underwater particles dominantly induces the blurring of the image. Moreover, the inherent characteristic of imaging system brings about the additional noise on the underwater image. Image enhancement and restoration in order to extend the imaging range is important for many civilian and military applications, including target detection, search and rescue, and diver visibility [1].