Dark-frame subtraction

About: Dark-frame subtraction is a research topic. Over the lifetime, 1216 publications have been published within this topic receiving 20763 citations.

Noise Removal in Color Images on Image Gradient

Abstract: In this paper,we propose a novel noise removal algorithm in color imagesGradient is firstly extended for color imagesOn this extended gradient,an algorithm called Gradient Peer Group is used to detect the noise in color images and the Peer Group Averaging alorithm is applied to remove the noise detectedThe experiment shows that the proposed algorithm can detect and remove noise in color images effectivelyThe quality of images after noise removal is better than the ones reconstructed by other algorithms

A novel and fast algorithm for random valued impulse image noise cancellation using cubic B-splines

Abstract: A novel and faster method for cancellation of random valued impulse noise is proposed in this paper. Our proposed algorithm works in two stages. In stage one, the noisy pixels are located and eliminated and in second stage, these noisy pixels are used for edge restoration. The results thus obtained are quite better and comparable to previously proposed algorithms and methods even at the noise levels of 60%.

Research of image denoising based on high precision of star sensors

Abstract: Image denoising is the process of getting the location coordinates of the star point. The extent of denoising determines the measurement accuracy of star sensor attitude. As the star sensor accuracy increasing, the need for star chart-depth study of the noise reduction process, thus reducing the false rate of the target extraction. Noise sources are: Star background noise, molecular noise, electronic noise, etc. In order to reduce the noise, analyze noise source and characteristicness; compare three Laplacian noise reduction effect; and put forward the associative templates. The results show that Laplacian typically is unacceptably sensitive to noise, will produce double edges, and is unable to detect edge direction; LOG has a voluminous convolving mask so that the computation is complex; The improved mask has good edge detection and high antinoise performance.

Restoration of impulse noise corrupted images based on gradients

Abstract: A new filtering algorithm is presented which can remove impulse noise from corrupted images while preserving details. The algorithm is based on a new impulse detection technique that uses image gradients. The proposed impulse detector can effectively categorize all the pixels in an image into two classes -- noise pixels and noise-free pixels. The noise-free pixels are kept untouched while the noise pixels are filtered by a noise cancellor such as median filter. Experimental results show that the proposed algorithm provides significant improvement over many existing techniques in terms of both subjective and objective evaluations. It also has the advantage of computational simplicity over those algorithms.

Spatio-temporal filter with adaptive multiple outliers rejector

Abstract: Digital images captured in dim light conditions can be very noisy: mixtures of different noise types may damage the image signal, e.g. impulsive and Gaussian noises. A spatio-temporal filter is proposed, which is capable to eliminate multiple adjacent defective pixels and reduce Gaussian noise in the raw data acquired by the image sensor. The proposed solution is adaptive in that it automatically adjusts its sensitivity, depending on the estimated noise.