Abstract—The goal of this work is to accurately detect and localize boundaries in natural scenes using local image measurements. We formulate features that respond to characteristic changes in brightness, color, and texture associated with natural boundaries. In order to combine the information from these features in an optimal way, we train a classifier using human labeled images as ground truth. The output of this classifier provides the posterior probability of a boundary at each image location and orientation. We present precision-recall curves showing that the resulting detector significantly outperforms existing approaches. Our two main results are 1) that cue combination can be performed adequately with a simple linear model and 2) that a proper, explicit treatment of texture is required to detect boundaries in natural images. Index Terms—Texture, supervised learning, cue combination, natural images, ground truth segmentation data set, boundary detection, boundary localization. 1

Abstract—A new method for evaluating edge detection algorithms is presented and applied to measure the relative performance of algorithms by Canny, Nalwa-Binford, Iverson-Zucker, Bergholm, and Rothwell. The basic measure of performance is a visual rating score which indicates the perceived quality of the edges for identifying an object. The process of evaluating edge detection algorithms with this performance measure requires the collection of a set of gray-scale images, optimizing the input parameters for each algorithm, conducting visual evaluation experiments and applying statistical analysis methods. The novel aspect of this work is the use of a visual task and real images of complex scenes in evaluating edge detectors. The method is appealing because, by definition, the results agree with visual evaluations of the edge images. Index Terms—Experimental comparison of algorithms, edge detector comparison, low level processing, performance evaluation, analysis of variance, human rating. 1

Most visual features are parametric in nature, including, edges, lines, corners, and junctions. We propose an algorithm to automatically construct detectors for arbitrary parametric features. To maximize robustness we use realistic multi-parameter feature models and incorporate optical and sensing effects. Each feature is represented as a densely sampled parametric manifold in a low dimensional subspace of a Hilbert space. During detection, the vector of intensity values in a window about each pixel in the image is projected into the subspace. If the projection lies sufficiently close to the feature manifold, the feature is detected and the location of the closest manifold point yields the feature parameters. The concepts of parameter reduction by normalization, dimension reduction, pattern rejection, and heuristic search are all employed to achieve the required efficiency. Detectors have been constructed for five features, namely, step edge (five parameters), roof edge (five parameters), line (six parameters), corner (five parameters), and circular disc (six parameters). The results of detailed experiments are presented which demonstrate the robustness of feature detection and the accuracy of parameter estimation.

This paper studies the problem of combining region and boundary cues for natural image segmentation. We employ a large database of manually segmented images in order to learn an optimal affinity function between pairs of pixels. These pairwise affinities can then be used to cluster the pixels into visually coherent groups. Region cues are computed as the similarity in brightness, color, and texture between image patches. Boundary cues are incorporated by looking for the presence of an “intervening contour”, a large gradient along a straight line connecting two pixels. We first use the dataset of human segmentations to individually optimize parameters of the patch and gradient features for brightness, color, and texture cues. We then quantitatively measure the power of different feature combinations by computing the precision and recall of classifiers trained using those features. The mutual information between the output of the classifiers and the same-segment indicator function provides an alternative evaluation technique that yields identical conclusions. As expected, the best classifier makes use of brightness, color, and texture features, in both patch and gradient forms. We find that for brightness, the gradient cue outperforms the patch similarity. In contrast, using color patch similarity yields better results than using color gradients. Texture is the most powerful of the three channels, with both patches and gradients carrying significant independent information. Interestingly, the proximity of the two pixels does not add any information beyond that provided by the similarity cues. We also find that the convexity assumptions made by the intervening contour approach are supported by the ecological statistics of the dataset. 1.

Image segmentation, i.e., identification of homogeneous regions in the image, has been the subject of considerable research activity over the last three decades. Many algorithms have been elaborated for gray scale images. However, the problem of segmentation for colour images, which convey much more information about objects in scenes, has received much less attention of scientific community. While several surveys of monochrome image segmentation techniques were published, similar comprehensive surveys for colour images, to our knowledge, did not emerge. This report

A new performance evaluation paradigm for computer vision systems is proposed. In real situation, the complexity of the input data and/or of the computational procedure can make traditional error propagation methods infeasible. The new approach exploits a resampling technique recently introduced in statistics, the bootstrap. Distributions for the output variables are obtained by perturbing the nuisance properties of the input, i.e., properties with no relevance for the output under ideal conditions. From these bootstrap distributions, the confidence in the adequacy of the assumptions embedded into the computational procedure for the given input is derived. As an example, the new paradigm is applied to the task of edge detection. The performance of several edge detection methods is compared both for synthetic data and real images. The confidence in the output can be used to obtain an edgemap independent of the gradient magnitude.

We present a methodology for the quantitative performance evaluation of detection algorithms in computer vision. A common method is to generate a variety of input images by varying the image parameters and evaluate the performance of the algorithm as algorithm parameters vary. Operating curves that relate the probability of mis-detection and false alarm are generated for each parameter setting. Such an analysis does not integrate the performance of the numerous operating curves. In this paper we outline a methodology for summarizing many operating curves into a few performance curves. This methodology is adapted from the human psychophysics literature and is general to any detection algorithm. The central concept is to measure the effect of variables in terms of the equivalent effect of a critical signal variable; which in turn facilitates the determination of the breakdown point of the algorithm. We demonstrate the methodology by comparing the performance of two line detection algorit...

Edges are useful features for structural image analysis, but the output of standard edge detectors must be thresholded to remove the many spurious edges. This paper describes experiments with both new and old techniques for: 1) automatically determining appropriate edge threshold values, and 2) determining edge saliency (as alternatives to gradient magnitude).

We present a comparative study between a complex Wavelet Coefficient Shrinkage (WCS) filter and several standard speckle filters that are widely used in the radar imaging community (Lee, Kuan, Frost, Geometric, Kalman, Gamma, etc.). The WCS filter is based on the use of Symmetric Daubechies (SD) wavelets which share the same properties as the real Daubechies wavelets but with an additional symmetry property. The filtering operation is an elliptical soft-thresholding procedure with respect to the principal axes of the 2-D complex wavelet coefficient distributions. Both qualitative and quantitative results (signal to mean square error ratio, equivalent number of looks, edgemap figure of merit) are reported. Tests have been performed using simulated speckle noise as well as real radar images. It is found that the WCS filter performs equally well as the standard filters for low-level noise and slightly outperforms them for higher-level noise. Keywords: Image processing, Synthetic Aperture...

Accurate image segmentation is important for many image, video and computer vision applications. Over the last few decades, many image segmentation methods have been proposed. However, the results of these segmentation methods are usually evaluated only visually, qualitatively, or indirectly by the effectiveness of the segmentation on the subsequent processing steps. Such methods are either subjective or tied to particular applications. They do not judge the performance of a segmentation method objectively, and cannot be used as a means to compare the performance of different segmentation techniques. A few quantitative evaluation methods have been proposed, but these early methods have been based entirely on empirical analysis and have no theoretical grounding. In this paper, we propose a novel objective segmentation evaluation method based on information theory. The new method uses entropy as the basis for measuring the uniformity of pixel characteristics (luminance is used in this paper) within a segmentation region. The evaluation method provides a relative quality score that can be used to compare different segmentations of the same image. This method can be used to compare both various parameterizations of one particular segmentation method as well as fundamentally different segmentation techniques. The results from this preliminary study indicate that the proposed evaluation method is superior to the prior quantitative segmentation evaluation techniques, and identify areas for future research in objective segmentation evaluation.