Let {~(n)}nEZbe a 2-color random scenery, that is a random coloration of in two colors, such that the ~(i)'s are i.i.d. Bernoulli variables with parameter~. Let {S(n)}nEN be a symmetric random walk starting at O. Our main result shows that a.s., ~ 0 S (the composition of ~ and S) determines ~ up to translation and reflection. In other words, by observing the scenery ~ along the random walk path...

This work focuses on characterizing scenery images. We semantically divide the objects in natural landscape scenes into background and foreground and show that the shapes of the regions associated with these two types are statistically different. We then focus on the background regions. We study statistical properties such as size and shape, location and relative location, the characteristics o...

This paper proposes an optimization procedure of a spatial filter in a water level detection system. The water level of running water of a river is detected by applying a temporal filter and a spatial filter to a video of various river scenery. Previously, we have proposed an algorithm based on “addition” of frames and “horizontal” detection of edges. We have confirmed its superiority to the ex...

Character detection in scenery images is a very difficult task. This paper describes a strategy of selecting character patterns for easier detection in scenery image. These character patterns, called conspicuous character patterns, are selected from character font sets according to a criterion that evaluates how the font has a larger distance from a non-character pattern distribution and, simul...

Consider a walker performing a random walk in an i.i.d. random environment, and assume that the walker tells us at each time the environment it sees at its present location. Given this history of the transition probabilities seen from the walker but not its trajectory can we reconstruct the law of the environment? We show that in a one-dimensional environment, the law of the environment can be ...