CONTOUR DETECTION BY MULTIRESOLUTION SURROUND INHIBITION Giuseppe Papari*, Patrizio Campisi**, Nicolai Petkov*, Alessandro Neri** ABSTRACT In natural images, luminance changes occur both on object con- tours and on textures. Often, the latter are stronger than the former, thus standard edge detectors fail in isolating object contours from texture. To overcome this problem, we propose a multiresolution contour detector motivated by biological principles. At each scale, texture is suppressed by using a bipolar surround inhibition proc- ess. The binary contour map is obtained by a contour selection cri- terion that is more effective than the classical hysteresis threshold- ing. Robustness to noise is achieved by Bayesian gradient estima- tion. Keywords: edge, context, contour, surround suppression, texture 1. INTRODUCTION Edge and contour detection, an important task in computer vision, is a fertile field of ongoing research (see [1] for a survey). Standard edge detectors react to all non-negligible luminance changes in an image, irrespective whether they are originated by object contours or by texture (grass, foliage, waves, etc.). Moreover, luminance changes due to texture are often stronger than ones due to contours. Our goal is to isolate objects in a scene; therefore, some further process is needed with respect to general purpose edge detectors. Specifically, we use some principles deployed in the Human Visual System (HVS). Psychophysical studies show that the perception of an oriented stimulus can be influenced by other similar stimuli in the surroundings [2]. Neurophysiological researches show that sur- round modulation is due to a specific neural mechanism. In [3] it has been suggested that surround suppression effectively enhances contours in natural images rich in textures. Other psychological experiments show that the retinal image is decomposed through band-pass filters. Low-pass filters are responsible for the so called pre-attentive stage of vision, corresponding to the first 0.1 0.3 s of the image persistency on the retina, where only the general mor- phology is perceived [4]. High-pass filters deliver information for the subsequent attentive stage where details are recognized. A mul- tiresolution approach to contour detection has been proposed in [5]. In the current work we extend our previous studies [6] in combin- ing a multiresolution approach and surround inhibition. We pro- pose a method that detects contours at different resolutions and combines them by a contour-oriented selection algorithm. At each scale, noise is reduced by optimal Bayesian Minimum Mean Square Estimation (MMSE) of the gradient in additive noise and texture is suppressed by a biologically motivated surround inhibi- tion process. 2. SCALE DEPENDANT CONTOUR DETECTOR The proposed single scale contour detector is depicted in Fig. 1, where I w (x,y) is a noisy version of a given image I(x,y) corrupted by additive independent noise w(x,y). First, the gradient of the in- put image I w is evaluated by convolution with the gradient of a Gaussian kernel g (x,y) [8]. The gradient estimation depends on the parameter , which we will call scale, or resolution. 2 2 2 2 2 1 , , 2 x y w w w I I g I g g xy e (1) Bayesian denoising Gradient computation I w I w b I Surround inhibition Fig. 1 Scale dependant contour detector. Then, Bayesian denoising, described in Section 2.1, is applied on the noisy image gradient. Surround inhibition is performed as de- tailed in Section 2.2. 2.1 Bayesian denoising Our goal is to find the optimal estimator ˆ I az of the un- known vector a = I, when a noisy version z = I w = I + w is observed. As well known from the Bayesian estimation theory, the optimal MMSE estimator is given by: ˆ p p p p d a za a za za a az za a a (2) According to recent statistical studies on natural images [7], both p a a and p za za are assumed Gaussian Scale Mixture (GSM), with covariance matrices A i and N k respectively: 1 1 2 2 2 1 1 1 2 1 ,, , 1 ,, K i K K i k K i k k k p p a za a a0A za zaN i i k N N (3) where: 1 2 1 1 , , exp 2 det 2 T μR -μ R -μ R N (4) By substituting eq. (3) in eq. (2), we can find the following closed expression for the optimal MMSE estimator: 1 -1 , , ,0, ,0, i k i k i i k ik i k ik z A N A A N z az z A N 2 2 N N . (5) The nonlinearity defined by eq. (5), applied to each pixel of the gradient w I , gives the best estimation of I . 2.2 Surround inhibition Next, a surround inhibition operator taking into account the context *Institute of Mathematics and Computing Science University of Groningen P. O. Box 800, 9700 AV Groningen, The Netherlands G.Papari@cs.rug.nl, petkov@cs.rug.nl **Dipartimento di Elettronica Applicata Università degli Studi di Roma “Roma Tre”, Via della Vasca Navale 84, 00146 Roma, Italy campisi@uniroma3.it, neri@uniroma3.it 749 1424404819/06/$20.00 ©2006 IEEE ICIP 2006