John F. Haddon

Image segmentation by unifying region and boundary information

A two-stage method of image segmentation based on gray level cooccurrence matrices is described. An analysis of the distributions within a cooccurrence matrix defines an initial pixel classification into both region and interior or boundary designations. Local consistency of pixel classification is then implemented by minimizing the entropy of local information, where region information is expressed via conditional probabilities estimated from the cooccurrence matrices, and boundary information via conditional probabilities which are determined a priori. The method robustly segments an image into homogeneous areas and generates an edge map. The technique extends easily to general edge operators. An example is given for the Canny operator. Applications to synthetic and forward-looking infrared (FLIR) images are given. >

Generalised threshold selection for edge detection

Abstract This paper derives a generalised technique for selecting thresholds of edge strength maps from theoretical considerations of the known noise statistics of the image. A technique is described for estimating the mean and variance of the noise of an image using a pair of images. The threshold selection technique can be applied to any single digital edge operator and has been extended for use with combinations of edge operators. Two examples of thresholded edge maps using the techniques developed here are shown.

Grouping of directional features using an extended Hough transform

Directional features extracted from Gabor responses are used as primitives for perceptual grouping. In previous work, we extracted Gabor features in 8 directions and then applied two self-organising maps, thus classifying each pixel in the image within a neuron-map, each corner of which represents one of four main directions. In this work we group pixels with similar directional features to detect salient structures within an image. Results obtained from application to forward-looking infrared (FLIR) images are very promising.

Perceptual Grouping from Gabor Filter Responses

Perceptual organisation can be defined as the ability to impose structural organisation on sensory data, so as to group sensory primitives arising from a common underlying cause. Our organisational philosophy is hierarchical, with complex organisations being formed from simpler ones. In this paper, directional features extracted from Gabor responses are used as the primitives for perceptual grouping. In previous work, we extracted Gabor features in 8 directions and then applied two SOMs, thus classifying each pixel in the image within a 8x10 neuronmap, each corner of which represents one of four main directions, (horizontal, vertical, left diagonal and right diagonal). In the present work we group pixels with similar directional features, thereby detecting salient structures within an image. These detected-structures will be used as tokens from which to create the next level of abstraction in the hierarchy of the system. This approach is an alternative to the use of sets of edges as primary features: the directional features that Gabor filters provide are a potentially richer source of information. Preliminary results obtained from application to forward-looking infrared (FLIR) images are very promising. At present only four main directions are utilised, i.e. vertical, horizontal, right diagonal and left diagonal: the technique may be readily extended to the eight utilised in previous work. The next stage will be to group the tokens by the application of additional Gestalt-laws in order to detect objects.

Integrating Spatio-temporal Information in Image Sequence Analysis to Enforce Consistency of Interpretation

We present a technique of image-sequence analysis which ensures consistency of interpretation in terms of the classification of segmented regions, both spatially within an image and temporally through a sequence of forward-looking infrared images taken from a low-flying aircraft. The technique is based on classical relaxation labeling techniques but is novel in that it treats segmented regions assingle entitiesthat have both spatial and temporal neighbors. This enables classification probabilities both to be propagated through the sequence and to be used in the minimization of the Kullback entropy of information. Sample results are shown from the application of the method to a 12-s sequence of infrared images that have been segmented using co-occurrence-based techniques. For completeness, the paper also describes the whole analysis algorithm including segmentation, texture analysis using discrete Hermite functions, and classification using neural network techniques.© British Crown copyright 1998. Published with the permission of the Defence Evaluation and Research Agency on behalf of the Controller of HMSO.

A relaxation computation of optic flow from spatial and temporal cooccurrence matrices

Introduces the temporal cooccurrence matrix and illustrates its application to the initial estimate of the optic flow field. Previous work on spatial segmentation and relaxation is extended to the relaxation of this flow field. The algorithm is inherently parallel since, at each iteration, only local neighbourhood operations are involved.<<ETX>>

FLIR image segmentation and natural object classification

In this paper we compare four classification techniques for classifying texture data of various natural objects found in forward-looking infrared (FLIR) images. The techniques compared include linear discriminant analysis, mean classifier and two different models of k-nearest neighbour methods. Hermite functions are used for texture feature extraction from segmented regions of interest in natural scenes taken as a video sequence. A total of 2680 samples for a total of twelve different classes are used for object recognition. The results on correctly identifying twelve natural objects in scenes are compared across the four classifiers on both unnormalised and normalised data. On unnormalised data, the average best recognition rate obtained using a ten fold cross-validation is 96.5%, and on unnormalised data it is 86.1% with a single nearest neighbour technique.

Detecting periodic structure

We present a method for the detection of periodic structure in images. Finding a global threshold to detect a significant frequency component, and then associating it with responsible features, is fragile. The method extracts features from the original signal, and uses their autocorrelation as the basis for an adaptive filter, with which they are convolved. This convolution represents the conformity of a local neighbourhood to the dominant spectral frequencies: combined with the original feature signal, it effectively incorporates the property of periodicity into a local feature strength. A recursive implementation is suggested. The method avoids the fragility associated with global thresholds, and is shown to work on real data.

Ego motion from near-degenerate sequences

This paper discusses ego motion recovery from planar scenes with a camera view parallel to the direction of movement. Reasons for the failure of two published algorithms are considered, and a novel method a successful recovery is presented.

Simultaneous region and edge segmentation of infrared images using non-maximal suppression for edge thinning.

A cooccurrence space is defined by utilising the combinations of pixel strengths defined by a Canny edge operator. A region and boundary segmentation derived from this space is first edge thinned by non-maximal suppression and then hysteresis is used as a post-processing step to improve the edges. The distributions in cooccurrence space define the thresholds employed in the hysteresis post-processing.