Adaptive Pseudo Dilation for Gestalt Edge Grouping and Contour Detection

Abstract

Abstract-We consider the problem of detecting object contours in natural images. In many cases, local luminance changes turn out to be stronger in textured areas than on object contours. Therefore, local edge features, which only look at a small neighborhood of each pixel, cannot be reliable indicators of the presence of a contour, and some global analysis is needed. We introduce a new morphological operator, called adaptive pseudo-dilation (APD), which uses context dependent structuring elements in order to identify long curvilinear structure in the edge map. We show that grouping edge pixels as the connected components of the output of APD results in a good agreement with the gestalt law of good continuation. The novelty of this operator is that dilation is limited to the Voronoi cell of each edge pixel. An efficient implementation of APD is presented. The grouping algorithm is then embedded in a multithreshold contour detector. At each threshold level, small groups of edges are removed, and contours are completed by means of a generalized reconstruction from markers. The use of different thresholds makes the algorithm much less sensitive to the values of the input parameters. Both qualitative and quantitative comparison with existing approaches prove the superiority of the proposed contour detector in terms of larger amount of suppressed texture and more effective detection of low-contrast contours.

Image Segmentation Using a Sparse Coding Model of Cortical Area V1

“…It has been used extensively for testing other biologically-inspired boundary detection algorithms, and the results shown in the last column of Fig. 5 can be compared directly to: Fig.15a in Papari and Petkov (2008) Tang et al (2007b). The sparsity of the representations found by PC/BC are not strongly influenced by the proposed extensions to this algorithm: the inclusion lateral connections, and subsequently the inclusion of two sets of prediction neurons to represent boundary edges and texture edges.…”

Edge and line oriented contour detection: State of the art

“…Much research has been carried out in order to provide fast algorithms. Examples are morphological approaches, which basically tend to fill the gaps between contour segments by directional dilation [183,184] …”

“…5 When this is not the case, it is sufficient to replace P and R with new quantities P′ = λP and R′ = μR, where the coefficients λ and μ are chosen so that, for that specific task, P′ and R′ are equally important. contour detectors [73,123,128,184] for each one of the 40 images of the RuG dataset [128]. For each algorithmic result, the similarity measures (Fig.…”

A novel multi-scale and multi-expert edge detector based on common vector approach

“…To compare the abilities of proposed methods for contour extraction, the results of the proposed method, Kunal [27], ED [28], EDISON [29] and APD [30] are presented in Fig. 3.…”