A New Ultrasound Speckle Reduction Algorithm Based on Superpixel Segmentation and Detail Compensation

Speckle is a kind of noise commonly found in ultrasound images (UIs). Although traditional local operation-based methods, such as bilateral filtering, perform well in de-noising normal natural images with suitable parameters, these methods may break local correlations and, hence, their performance will be highly degraded when applied to UIs with high levels of speckle noise. In this work, we propose a new method, based on superpixel segmentation and detail compensation, to reduce UI speckle noise. In particular, considering that superpixel segmentation has the advantage of adhering accurately to the boundaries of objects or local structures, we propose a superpixel version of bilateral filtering to better protect the local structure during de-noising. Additionally, a human visual system (HVS)-inspired strategy for spatial compensation is introduced, in order to recover sophisticated edges as much as possible while weakening the high-frequency noise. Experiments on synthetic images and real UIs of different organs show that, compared to other methods, the proposed strategy can reduce ultrasound speckle noise more effectively.

The Spectral Dimension of Modeled Relative Humidity Feedbacks in the CMIP5 Experiments

The spectral radiative kernel technique is used to derive spectrally resolved relative humidity (RH) feedbacks for 16 GCMs that participated in CMIP5. Examining spectral and spatial details of these RH feedbacks leads to three findings. First, while the global average of broadband RH feedbacks is close to zero for all the GCMs, there exist wide discrepancies in the spectral details among the GCMs. Second, the spatial pattern of the RH feedbacks varies with spectral frequency; for a given frequency, the spatial feedback pattern correlates well with the spatial pattern of RH changes over the vertical layer to which the top-of-atmosphere spectral flux at the same frequency is most sensitive. Third, the nearly zero global average of broadband RH feedback is a result of spatial compensation and spectral compensation. Since GCMs have produced consistent RH feedbacks and RH changes to a large extent in the high latitudes, the tropical Atlantic, and the deep tropical Pacific, radiative RH kernels are further used to infer mean changes in RH vertical profiles from spectral RH feedbacks averaged over the three broad regions. Good agreements are demonstrated by comparing such inferences against the actual RH changes in the GCMs. This study suggests that the spectral dimension of RH feedback could provide an additional constraint to climate models and the understanding of vertical features of RH changes when they are obtainable from future observations.