Bilateral filter

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

Systolic realization of motion compensated interpolation filter

Abstract: A motion compensated interpolation filter calculates an interpolated frame between two input frames of video signals having moving objects therein. The interpolation filter includes two 2-dimensional filter which are identical in structure. Each of the 2-dimensional filters has a 2-dimensional systolic array structure for providing a weighted sum of 4 pixels of each of the input frames, respectively. Each of the 2-dimensional filter includes (2N+1)·(2N+1) identical processing elements, and each of a block of (2N+1)·(2N+1) pixels is inputted to each of the processing elements, and each of the processing elements generates filter coefficients, to be multiplied to a corresponding pixel values. Since the systolic array structure inherently incorporates modularity and regularity therein, the interpolation filter is suitable for real-time processing with easy hardware implementation.

Mixed Filter for Image Denoising

Abstract: The conventional average filter and median filter have different filtering characteristics to Gauss noise and impulse noise. In fact, image usually is corrupted by Gauss noise and impulse noise simultaneously, so good filtering effect cannot be obtained if we only use average filter or median filter. In this paper, a new mixture filter method is proposed. The proposed method first classsifies the pixels into two classes, one is the pixels which are corrupted by Gauss noise and the other is the pixels corrupted by impulse noise and then average filter is used for the pixels corrupted by Gauss noise and median filter is used for the pixels corrupted by impulse noise. The simulation result verifies that the proposed method is feasible and efficient.

An efficient edge-based bilateral filter for restoring real noisy image

Abstract: This paper presents an efficient edge-based bilateral filter for real noisy image restoration. By dividing all pixels of a noisy image into edge region or non-edge region, the different strategies and parameters are adopted in the edge-based bilateral filter to balance the conservation of image features and the reduction of noise level. Extensive experimental results are shown that this filter achieves very competitive performance in restoring real noisy images, compared with other state-of-the-art denoising algorithm.

Solid-State Imaging Device and Pixel Correction Method

Abstract: A solid-state imaging device capable of correcting defective pixel signals to improve image quality. A line memory provides a value of a pixel currently selected for correction, together with values of its surrounding pixels. The surrounding pixels include corrected pixels preceding the selected pixel and uncorrected pixels succeeding the selected pixel. An extreme value remover removes effectively a maximum and minimum pixel values from the values of the corrected pixels read out of the line memory. An average calculator calculates an average value of the remaining uncorrected pixels and the corrected pixels read out of the line memory. A comparison processor compares the value of the selected pixel with the average value. If their difference exceeds a predetermined threshold, the comparison processor replaces the value of the selected pixel with the average value.

Two-Colored Pixels

Abstract: In this paper we show how to use two-colored pixels as a generic tool for image processing. We apply two-colored pixels as a basic operator as well as a supporting data structure for several image processing applications. Traditionally, images are represented by a regular grid of square pixels with one constant color each. In the two-colored pixel representation, we reduce the image resolution and replace blocks of N × N pixels by one square that is split by a (feature) line into two regions with constant colors. We show how the conversion of standard mono-colored pixel images into two-colored pixel images can be computed efficiently by applying a hierarchical algorithm along with a CUDA-based implementation. Two-colored pixels overcome some of the limitations that classical pixel representations have, and their feature lines provide minimal geometric information about the underlying image region that can be effectively exploited for a number of applications. We show how to use two-colored pixels as an interactive brush tool, achieving realtime performance for image abstraction and non-photorealistic filtering. Additionally, we propose a realtime solution for image retargeting, defined as a linear minimization problem on a regular or even adaptive two-colored pixel image. The concept of two-colored pixels can be easily extended to a video volume, and we demonstrate this for the example of video retargeting.