Finite sampling corrected 3D noise with confidence intervals

2009 First Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, 2009

This paper deals with an original method to estimate the noise introduced by optical imaging systems, such as CCD cameras, multispectral scanners and imaging spectrometers. The power of the signal-dependent photonic noise is decoupled from the power of the signal-independent noise generated by the electronic circuitry. The method relies on the multivariate regression of local sample statistics such as mean and variance, in which statistically homogeneous pixels produce scatter-points that are clustered along a straight line, whose slope and intercept measure the signal-dependent and signal-independent components of the noise power, respectively. Experimental results carried out on a simulated noisy image and on true data from a modern generation airborne imaging spectrometer highlight the accuracy of the proposed method and its robustness to image textures that may lead to a gross overestimation of the noise, especially for high SNR.

Physica medica : PM : an international journal devoted to the applications of physics to medicine and biology : official journal of the Italian Association of Biomedical Physics (AIFB), 2013

The goal of this study was to investigate the impact of computing parameters and the location of volumes of interest (VOI) on the calculation of 3D noise power spectrum (NPS) in order to determine an optimal set of computing parameters and propose a robust method for evaluating the noise properties of imaging systems. Noise stationarity in noise volumes acquired with a water phantom on a 128-MDCT and a 320-MDCT scanner were analyzed in the spatial domain in order to define locally stationary VOIs. The influence of the computing parameters in the 3D NPS measurement: the sampling distances bx,y,z and the VOI lengths Lx,y,z, the number of VOIs NVOI and the structured noise were investigated to minimize measurement errors. The effect of the VOI locations on the NPS was also investigated. Results showed that the noise (standard deviation) varies more in the r-direction (phantom radius) than z-direction plane. A 25 × 25 × 40 mm(3) VOI associated with DFOV = 200 mm (Lx,y,z = 64, bx,y = 0.3...

SPIE Proceedings, 2013

Noise is present in all image sensor data. Poisson distribution is said to model the stochastic nature of the photon arrival process, while it is common to approximate readout/thermal noise by additive white Gaussian noise (AWGN). Other sources of signal-dependent noise such as Fano and quantization also contribute to the overall noise profile. Question remains, however, about how best to model the combined sensor noise. Though additive Gaussian noise with signal-dependent noise variance (SD-AWGN) and Poisson corruption are two widely used models to approximate the actual sensor noise distribution, the justification given to these types of models are based on limited evidence. The goal of this paper is to provide a more comprehensive characterization of random noise. We concluded by presenting concrete evidence that Poisson model is a better approximation to real camera model than SD-AWGN. We suggest further modification to Poisson that may improve the noise model.

2007 International Symposium on Signals, Circuits and Systems, 2007

Time-of-flight (TOF) cameras are based on a new technology that delivers distance maps by the use of a modulated light source. In this paper we first describe a set of experiments that we performed with TOF cameras. We then propose a noise model which is able to explain some of the phenomena observed in the experiments. The model is based on assuming a noise source that is correlated with the light source (shot noise) and an additional additive noise source (dark current noise). The model predicts well the dependency of the distance errors on the image intensity and the true distance at an individual pixel.

IEEE Transactions on Medical Imaging, 2011

Any measurement of signal intensity obtained from an image will be corrupted by noise. If the measurement is from one voxel, an error bound associated with noise can be assigned if the standard deviation of noise in the image is known. If voxels are averaged together within a region of interest (ROI) and the image noise is uncorrelated, the error bound associated with noise will be reduced in proportion to the square root of the number of voxels in the ROI. However, when 3-D-radial images are created the image noise will be spatially correlated. In this paper, an equation is derived and verified with simulated noise for the computation of noise averaging when image noise is correlated, facilitating the assessment of noise characteristics for different 3-D-radial imaging methodologies. It is already known that if the radial evolution of projections are altered such that constant sampling density is produced in k-space, the signal-to-noise ratio (SNR) inefficiency of standard radial imaging (SR) can effectively be eliminated (assuming a uniform transfer function is desired). However, it is shown in this paper that the low-frequency noise power reduction of SR will produce beneficial (anti-) correlation of noise and enhanced noise averaging characteristics. If an ROI contains only one voxel a radial evolution altered uniform k-space sampling technique such as twisted projection imaging (TPI) will produce an error bound 35% less with respect to noise than SR, however, for an ROI containing 16 voxels the SR methodology will facilitate an error bound 20% less than TPI. If a filtering transfer function is desired, it is shown that designing sampling density to create the filter shape has both SNR and noise correlation advantages over sampling k-space uniformly. In this context SR is also beneficial. Two sets of 48 images produced from a saline phantom with sodium MRI at 4.7T are used to experimentally measure noise averaging characteristics of radial imaging and good agreement with theory is obtained.

2005 5th International Conference on Information Communications & Signal Processing

A noise model that is generated from pictures taken by differently illuminating the same subject matter is presented. The superposigram, a three dimensional structure based on three different illuminations of the subject matter is used to directly calculate the noise present in the images which is attributable to the noise originating in the sensor array. Though a similar method has been used to determine the dynamic range compression present in a particular camera, this paper focuses on using data in which no dynamic range compression has taken place (raw files from a Nikon D2h were used). The paper shows through a mathematical proof that the noise in the camera is well estimated by the superposigram and then shows empirically that this is true. Finally, the method is applied to four commercially available cameras, to evaluate the noise in each.

2008 IEEE International Conference on Shape Modeling and Applications, 2008

This paper discusses noise in range data measured by a Konica Minolta Vivid 910 scanner. Previous papers considering denoising 3D mesh data have often used artificial data comprising Gaussian noise, which is independently distributed at each mesh point. Measurements of an accurately machined, almost planar test surface indicate that real scanner data does not have such properties. An initial characterisation of real scanner noise for this test surface shows that the errors are not quite Gaussian, and more importantly, exhibit significant short range correlation. This analysis yields a simple model for generating noise with similar characteristics. We also examine the effect of two typical mesh denoising algorithms on the real noise present in the test data. The results show that new denoising algorithms are required to effectively remove real scanner noise.

Electronic Imaging

Fast track article for IS&T International Symposium on Electronic Imaging 2020: Image Quality and System Performance proceedings.

International journal of engineering research and technology, 2015

This paper presents Dobbin's method to estimate the noise power spectrum using a screen film system. The one-dimensional spectral estimate was obtained by extracting thick and thin slices from two-dimensional noise power. The slices were made parallel to the primary axis of ROI, but did not include the axis. We measured NPS using one slice, two slices, four slices, eight slices,upper eight slices (a) and eight slices (b) of data in the 128×128 two-dimensional NPS space which were extracted to generate the one- dimensional NPS curves in horizontal and vertical directions and they were compared with Dobbin's method. Very little was found in the NPS shape with regards to the two- dimensional space only and the slice which contained one row and one column was sufficient to study NPS in the two- dimensional space Keywords—X-ray, Noise power spectrum, screen film system, fast Fourier transform

2003

This paper presents a simulation tool for the rigorous analysis of the final uncertainty associated to different methodologies for noise figure characterization. The simulation tool permits the analysis of the combined effect of systematic errors and underlying uncertainties versus any significant characteristic of the DUT or measurement setup. Some application examples are presented showing the suitability of the proposed approach to determine the most efficient characterization methodology for a given DUT and measurement setup.