Elaborated Motion Detector based on Hassenstein-Reichardt correlator model MIHAI-EMANUEL BASCH DAVID-GEORGE CRISTEA VIRGIL TIPONUŢ TITUS SLAVICI Applied Electronics “POLITEHNICA” University of Timişoara Timişoara, Str. Vasile Parvan, Nr. 2 ROMANIA bash_mihai@yahoo.com cristea.david@yahoo.com virgil.tiponut@yahoo.com tslavici@yahoo.com Abstract: Many of the motion detectors are based on identifying and tracking spatial or temporal features. The main disadvantage of these circuits is the uncertain detection at low SNR conditions. In 1956 Hassenstein and Reichardt tried to explain the mechanism of the insect vision and proposed an alternative to motion detection with an intensity-based spatiotemporal correlation algorithm. Based on this type of algorithm we implemented an improved motion detector that not only detects a moving object but also gives some additional information such as the sense and direction from where the object is coming. This system is intended to be used as front-end processing for a more complex visual motion computation models, like the ones performed by insects such as flies or locusts. Key-Words: motion detection, motion computation, Reichardt correlator, EMD, insect vision, sense and direction detectors 1 Introduction The flying insects visual systems is a good inspiration in creating a robust motion detector, that works in a natural environment. Even if insects are considered simple organisms compared to vertebrates because of their nervous systems structure, they have this extraordinary ability to fly and adapt as a result of genetic inheritance characteristics of previous generations, that is way not in vain it has been given them the name of “perfect flying machines”. They can land or take off vertically, move backward or laterally, and also fly upside-down. The Reichardt or correlation motion detector has been inspired by the experiments that followed the behavioral of the optomotor response of insects. This type of motion detector has a parallel architecture that compute in very different ways from traditional digital computers. In practice two channels detects motion in a preferred direction by comparing a signal from one receptor with a delayed signal from the other receptor. The Reichardt detector, also known as Hassenstein- Reichardt detector or Elementary Motion Detector (EMD) seems to be the fundamental part in all insect motion processing. This paper present an architecture that combine this basic motion detectors inspired by the insect visual system and add new features like the localization in space of the moving object. 2 The “basic” Reichardt detector The simplified version of the Hassenstein and Reichardt correlator model is made from two symmetrical sub- units (Fig.1). Fig.1 The elementary motion detector (EMD) block diagram In each sub-unit the signals received from their neighboring entries are multiplied each other after one of LATEST TRENDS on SYSTEMS (Volume I) ISSN: 1792-4235 192 ISBN: 978-960-474-199-1