Statistical model-based algorithms for image analysis

Therrien, C.W.; Quatieri, Thomas F.; Dudgeon, Dan E.

doi:10.1109/proc.1986.13504

Cited by 89 publications

(34 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The tendency to produce a "noisy" result is not surprising since no attempt has been made to model spatial correlation in the labelled image. One method of encouraging the formation of regions in the segmented image to use a Bayesian approach with a Markov process describing the region prior [2], [12], [7]. In our case we simply apply a median filter to the segmented image, as shown in Fig.…”

Section: Discussionmentioning

confidence: 99%

“…In this work we investigate the use of a two dimensional AR model. AR models have been successful in modelling image texture in several image processing applications, for example optical aerial images [12]. We apply a two dimensional AR model to complex amplitude data in order to partition the image into regions eharaeterised by distinct second order statistics.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Segmentation of ultrasound image data by two dimensional autoregressive modelling

Abbott

Braun

1997

Image Analysis and Processing

View full text Add to dashboard Cite

Abstract. In this paper we treat ultrasound image data as a two dimensional autoregressive (AR) signal. The image is modelled as consisting of distinct regions each described by one of a small number of AR models. Segmentation is performed by maximising the image likelihood function, which takes on a convenient form due to the AR model. Image data is presented to the algorithm in complex amplitude form. Results from application of this method to a cardiac phantom data set are presented.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Segmentation of ultrasound image data by two dimensional autoregressive modelling

Abbott

Braun

1997

Image Analysis and Processing

View full text Add to dashboard Cite

show abstract

“…In simple image subtraction, the measure of change is based on the difference between corresponding pixel values. Adaptive techniques (Therrien et al 1986, Lee et al 1986, Ca rlotto et al 1992 use information over larger areas (e.g., within a sliding window, regions of similar material type, or the entire image) to model and predict the two images from each other. The difference between the actual and predicted image is used as a measure of change.…”

Section: Change Detectionmentioning

confidence: 99%

<title>Nonlinear mean-square estimation with applications in remote sensing</title>

Carlotto¹

1996

SPIE Proceedings

View full text Add to dashboard Cite

An approach to image modeling based on nonlinear mean-square estimation that does not assume a functio nal form for the model is described. The relationship between input and output images is represented in the form of a lookup table that can be efficiently computed from, and applied to images. Three applications are presented to illustrate the utility of the technique in remote sensing. The first illustrates how the method can be used to estimate the values of physical parameters from imagery. Specifically we estimate the topograph ic component (i.e., the variation in brightness caused by the shape of the surface) from multispectral imager y. The second application is a nonlinear change detection algorithm which predicts one image as a nonlinea r function of another. In cases where the frequency of change is large (e.g., due to atmospheric and environ mental differences), the algorithm is shown to be superior in performance to linear change detection. In the last application, a technique for removing wavelength-dependent space-varying haze from multispectral ima gery is presented. The technique uses the IR bands, which are not affected significantly by haze, to predict t he visible bands. Results show a significant reduction in haze over the area considered. Additional applicati on areas are also discussed.

show abstract

“…One of these, the Reed Xiaoli (RX) algorithm [2], detects objects of known shape/shape that are spectrally different from the immediate background. Autoregressive [3] and fractal models [4] have been also proposed for detecting manmade objects in textured backgrounds. Many of these algorithms use some form of sliding window to find objects of a given size.…”

Section: Introductionmentioning

confidence: 99%

Feature-based anomaly detection

Carlotto

2007

SPIE Proceedings

View full text Add to dashboard Cite

A feature-based approach for detecting anomalies in spectral, spatial, temporal, and other domains is described. When the frequency of occurrence is small relative to the background, anomalies such as man-made objects in natural image backgrounds do not form their own clusters, but are instead assigned the nearest background cluster, becoming an outlier (statistical anomaly) in that cluster. Our method clusters data, which may be spectral, spatial, or temporal in nature, into one or more background types and computes the Mahalanobis distance between the data and assigned model (background cluster). The detection of a variety of objects and phenomena in panchromatic and multispectral imagery, and video are illustrated.

show abstract

Statistical model-based algorithms for image analysis

Cited by 89 publications

References 32 publications

Segmentation of ultrasound image data by two dimensional autoregressive modelling

Segmentation of ultrasound image data by two dimensional autoregressive modelling

<title>Nonlinear mean-square estimation with applications in remote sensing</title>

Feature-based anomaly detection

Contact Info

Product

Resources

About