Bilateral filter

About: Bilateral filter is a research topic. Over the lifetime, 3500 publications have been published within this topic receiving 75582 citations.

Optimally stabilized PET image denoising using trilateral filtering.

Abstract: Low-resolution and signal-dependent noise distribution in positron emission tomography (PET) images makes denoising process an inevitable step prior to qualitative and quantitative image analysis tasks. Conventional PET denoising methods either over-smooth small-sized structures due to resolution limitation or make incorrect assumptions about the noise characteristics. Therefore, clinically important quantitative information may be corrupted. To address these challenges, we introduced a novel approach to remove signal-dependent noise in the PET images where the noise distribution was considered as Poisson-Gaussian mixed. Meanwhile, the generalized Anscombe’s transformation (GAT) was used to stabilize varying nature of the PET noise. Other than noise stabilization, it is also desirable for the noise removal filter to preserve the boundaries of the structures while smoothing the noisy regions. Indeed, it is important to avoid significant loss of quantitative information such as standard uptake value (SUV)-based metrics as well as metabolic lesion volume. To satisfy all these properties, we extended bilateral filtering method into trilateral filtering through multiscaling and optimal Gaussianization process. The proposed method was tested on more than 50 PET-CT images from various patients having different cancers and achieved the superior performance compared to the widely used denoising techniques in the literature.

Adaptive Image Decomposition by Improved Bilateral Filter

Abstract: This paper presents an adaptive image decomposition method based on the newly proposed texture descriptor, which combines contrast variation and structure correlation. By incorporating it into the classical edge-preserving bilateral filter, we construct an adaptive nonlinear filter capable of preserving edges and removing texture parts effectively. We derive the variational formulation of the proposed filter and explore its relationship with robust statistics. Comparative experiments against other state-of-the-art decomposition approaches show that better separation results of geometric structure from texture are obtained.

Systems and methods for adaptive control and dynamic range extension of image sensors

Abstract: Systems and methods are provided for obtaining adaptive exposure control and dynamic range extension of image sensors. In some embodiments, an image sensor of an image system can include a pixel array with one or more clear pixels. The image system can separately control the amount of time that pixels in different lines of the pixel array are exposed to light. As a result, the image system can adjust the exposure times to prevent over-saturation of the clear pixels, while also allowing color pixels of the pixel array to be exposed to light for a longer period of time. In some embodiments, the dynamic range of the image system can be extended through a reconstruction and interpolation process. For example, a signal reconstruction module can extend the dynamic range of one or more green pixels by combining signals associated with green pixels in different lines of the pixel array.

Variably fast and continuous bilateral approximation filtering using histogram manipulations

Abstract: A system and method for performing integral histogram convolution for filtering image data is disclosed. The method may include applying a filter window to a first portion of an image, wherein the filter window includes an interior region and a border region. The method may include generating a plurality of histograms for the pixels in the filter window. The method may include generating spatial weight coefficients for the pixels in the border of the filter window. The method may include generating a plurality of color weight coefficients for the pixels in the filter window. The method may include performing a filtering operation on the pixels in the filter window by applying a respective spatial weight coefficient and a respective color weight coefficient to the values in the plurality of histograms for each respective pixel in the filter window. The methods may be implemented by program instructions executing in parallel on CPU(s) or GPUs.