A New Normalized Supervised Edge Detection Evaluation

Abstract

In digital images, edges characterize object boundaries, then their detection remains a crucial stage in numerous applications. To achieve this task, many edge detectors have been designed, producing different results, with different qualities. Evaluating the response obtained by these detectors has become a crucial task. In this paper, several referenced-based boundary detection evaluations are detailed, pointing their advantages and disadvantages through concrete examples of edge images. Then, a new supervised edge map quality measure is proposed, comparing a ground truth contour image, the candidate contour image and their associated spacial nearness. Compared to other boundary detection assessments, this new method has the advantage to be normalized and remains a more reliable edge map quality measure.

The Evaluation of Entropy-based Algorithm towards the Production of Closed-Loop Edge

“…The sample of experimental results is shown in Figure 8. Rather than employing parameters that appreciate the existence of open-loop edge [28]- [30], the analysis of experimental results is based on set parameters to measure the existence of closed-loop edge, namely the average number of detected objects (s1), the average number of detected edge pixels (s2), the average size of detected objects (s3), the ratio of the number of edge pixel per object (s4), and the average size of tenth biggest objects (s5). These parameters are to identify the requirement to become the desired algorithm, i.e., the method that delivered the meaningful result by producing the significant closed-loop edge while avoiding oversegmentation.…”

A New Objective Supervised Edge Detection Assessment Using Hysteresis Thresholds

“…A supervised evaluation criterion computes a dissimilarity measure between a segmentation result and a ground truth obtained from synthetic data or an expert judgment (i.e. manual segmentation) [11][12] [13] [14]. In this paper, the closer to 0 the score of the evaluation is, the more the segmentation is qualified as good.…”

“…Tolerating a distance from the true contour and integrating several TPs for one detected contour can penalize efficient edge detection methods, or, on the contrary, advantage poor ones (especially for corners or small objects). Thus, from the discussion below, the assessment should penalize a misplaced edge point proportionally to the distance from its true location (some examples in [14], and, as shown in Fig. 2).…”

“…Nevertheless, as pointed out in [11], an average distance from the edge pixels in the candidate image to those in the ground truth is more appropriate, like S k or Ψ . Eventually, Delta Metric (∆ k ) [27] intends to estimate the dissimilarity between each element of two binary images, but is highly sensitive to distances of misplaced points [8] [14].…”

“…A new objective edge detection assessment measure: In [14] a measure of the edge detection assessment is developed: it is denoted Ψ (Tab. 2) and improvemes the over-segmentation measure Γ , by combining both d Gt and d Dc , see Fig.…”