Coherent imaging sensors have been widely used in many fields, such as medical, transportation, etc. However, the captured images in the real-world often exist the interference of speckle noise, which affects the application of artificial intelligence to discover the target of interest. To eliminate the unwanted speckle noise, this paper proposes a deep learning-enabled despeckling network. The proposed network adopts the multi-scale attention module to extract the distribution features of speckle noise, which can fully remove the noise and guarantee the natural edge. The proposed approach is able to promote the performance of various vision-based high-level tasks.

Coherent imaging systems have been applied in the detection of target of interest, natural resource exploration, ailment diagnosis, etc. However, it is easy to generate speckle-degraded images due to the coherent interference of reflected echoes, restricting these practical applications. Speckle noise is a granular interference that affects the observed reflectivity. It is often modeled as multiplicative noise with a negative exponential distribution. This nonlinear property makes despeckling of imaging data an intractable problem. To enhance the despeckling performance, we propose to blindly remove speckle noise using an intelligent computing-enabled multiscale attention-guided neural network (termed MSANN). In particular, we first introduce the logarithmic transformation to convert the multiplicative speckle noise model to an additive version. Our MSANN, essentially a feature pyramid network, is then exploited to restore degraded images in the logarithmic domain. To enhance the gener...