Image Analysis For Deformation Behavior Of Optical Based Silicone Tactile Sensor Abdul Halim Esa, Muhammad Azmi Ayub Faculty of Mechanical Engineering Universiti Teknologi MARA (UiTM) 40450 Shah Alam, West Malaysia abdhalim85@gmail.com muhammadayub@salam.uitm.edu.my Bakri Ali Faculty of Mechanical Engineering Universiti Teknologi MARA (UiTM), 13500 Permatang Pauh, West Malaysia bakali65@yahoo.co.uk Abstract – This paper presents a tactile sensor which is made of a silicone based material. The tactile sensor is designed similarly like a human finger. The hollow space inside the tactile sensor is marked at the ‘finger tip’ inner side to sense any deformation when force is apply outside the ‘finger tip’ . The force is apply by using micrometer tip . A CCD camera is connected to a fiberscope and the fiberscope is connected to the tactile sensor to ‘see’ the deformed image inside the silicone dome. Several images is captured. The captured image is analyzed by a WiT image processing algorithm. Based on the processed images, the result is analyzed and the deformation behavior is studied. Keywords – Optical based sensor, tactile sensor, image analysis, deformation behavior. I. INTRODUCTION Tactile sensor is a type of sensor that sense any physical properties and characteristics through contact with objects at any parts or section. This sensory device is an essential device to integrate with robot control system specifically in object manipulation task. [1] To date, common several basic tactile sensing principle are using capacitive sensor, piezoelectric sensor, inductive sensor, opto-mechatronic and piezoelectric sensor as their basic sensing principle. Massachusetts Institute of Technology (MIT) research team has developed sensor with a protruding shape that allows them to easily make contact with the world in a similar way to ridges of human fingerprint. The sensor can estimate the magnitude and direction of applied forces with great sensitivity by measuring the deformation of compliant dome. To perform the object manipulation tasks effectively for robot system, at least two types of tactile information is required which is contact sense and slippage. To define contact sense, normal force by means of static tactile sensing is measured. To define slippage, shear force by dynamic tactile sensing is measured. Therefore the tactile sensor system is able to measure force at three axes direction [2]. Nagoya University research team has developed a novel optical three axes tactile sensor system based on an optical waveguide transduction method capable of acquiring normal and shear force [3]. In order to help surgeon to determine the health of the tissue, Fath El Bab et al (2009) uses information from a miro-machined piezoresistive type tactile sensor to detect reciprocal of stiffness of a soft tissue [4]. A capacitive type tactile sensor using copper clad laminated with flexible polyimide substrates is fabricated by Anastasios Petropoulus et al (2009) [5]. A tactile sensor based on tridimensional piezoelectric aluminium nitride (AIN) membranes is proposed by Tobias Polster et al (2009) [6] Most of the sensors using piezoelectric, piezoresistive and capacitive sensing technique only describe the magnitude of the force and hardness of the object and cannot react to force changes and also can’t determine the direction of the forces. Waveguide transduction method technique can react to the forces and direction changes. That is why this method is more convinced than other method. However, its individual protruding column feeler is very small about 6 mm and will be easily damage when handled roughly and harshly. All the mostly using piezoelectric, piezoresistive and capacitve based principle. 2012 IEEE 8th International Colloquium on Signal Processing and its Applications 978-1-4673-0961-5/12/$31.00 ©2012 IEEE 23