Rotation Angle Estimation Algorithms for Textures and Their Real-Time Implementation on the FU-SmartCam

Abstract

In this paper, we study rotation angle estimation algorithms for textures and discuss their real-time implementation on the FU-SmartCam. Rotation angle estimation for textures is closely related to the weft-straightening problem encountered in the textile industry. Earlier weft-straightening machines were based on simpler sensors and hardware, but with the increased complexity of fabric types, and demand for faster and more accurate machines, the whole industry started to switch to real-time machine vision based systems. For example, one of the leading companies in this area, Erhardt+Leimer, started to implement realtime machine vision systems based on the tattile smart camera. In this work, we both study rotation angle estimation algorithms for textures, and their real-time implementation on the locally developed Fatih University smart camera (FU-SmartCam). We also compare the FU-SmartCam with the tattile smart camera used in the Erhardt+Leimer system, both in terms of hardware and ease of prototyping.