We introduce a new similarity measure that is insensitive to sub-pixel misregistration. The proposed measure is essential in some differences detection scenarios. For example, in a setting where a digital reference is compared to an image, where the imaging process introduces deformations that appear as non constant misregistration between the two images. Our goal is to ignore image differences that result from misregistration and detect only the true, albeit minute, defects. In order to define a misregistration insensitive similarity, we argue that a similarity measure must respect convex combinations. We show that the well known SSIM does not hold this property and propose a modified version of SSIM that respects convex combinations. We then use this measure to define Sub-Pixel misregistration aware SSIM (SPSSIM).

This paper presents a novel self-similarity based approach for the problem of vanishing point estimation in man-made scenes. A vanishing point (VP) is the convergence point of a pencil (a concurrent line set), that is a perspective projection of a corresponding parallel line set in the scene. Unlike traditional VP detection that relies on extraction and grouping of individual straight lines, our approach detects entire pencils based on a property of 1D affine-similarity between parallel cross-sections of a pencil. Our approach is not limited to real pencils. Under some conditions (normally met in man-made scenes), our method can detect pencils made of virtual lines passing through similar image features, and hence can detect VPs from repeating patterns that do not contain straight edges. We demonstrate that detecting entire pencils rather than individual lines improves the detection robustness in that it improves VP detection in challenging conditions, such as very-low resolution or weak edges, and simultaneously reduces VP false-detection rate when only a small number of lines are detectable.

In the past years, electronic routers have had trouble keeping up with the increase in optical fiber capacity. As their power consumption has grown exponentially and already exceeds standards, it seems that an alternative solution is mandatory. Many have suggested all-optical routers as an alternative. However, these are deemed too complex, especially given the need to implement both switching and buffering, even though their fundamental complexity has apparently never been analyzed. In this paper, we study the number of fundamental optical components (2 times 2 switches and fiber delay lines) needed to emulate ideal routers. We first demonstrate that an N times N router with a buffer size of B per port needs at least thetas(N log(N B)) components, and then build a construction that achieves this lower bound. On the way, we also present an optical buffer construction of size B that works with thetas(log(B)) components, which is also shown to be a lower bound. Finally, we generalize this result to different router architectures and scheduling disciplines.

It is often claimed that future systems will necessarily be all-optical, because electronic devices are not fast enough to keep up with the increase in fiber capacity. However, two objections are commonly raised: first, optical systems need many basic optical components, which are typically very expensive; and second, optical systems need many switch reconfigurations, which are typically very slow. In this paper, we examine whether these two costs can be fundamentally bounded. First, we develop the equivalence between coding theory and optical system design by introducing the concept of super switches. Then, we show how the minimal expected number of switch reconfigurations is almost equal to the state space entropy of the optical system. Finally, we point out the trade-off between the two types of costs.