A real-time algorithm of optical tactile sensor for surface characterization / Nurul Fathiah Mohamed Rosli

Many advances have been made in terms of sensor technology for sensitivity optimization. Mostly available optical tactile sensors are bulky, inflexible and lack dexterity for applications with limited or space constraint. In view of this, current tactile sensors cannot be used in endoscopic inspection applications. Hence, this issue aims to provide analysis on the behaviour of a newly developed tactile sensor for sensing surface characteristics as the main objective of the research work. In this research, the main objective of this research work is to analyze the behaviour of a newly developed optical tactile sensor for sensing surface characteristics. In this research, the characteristics of the developed tactile sensor are calibrated with respect to changes in diameter, area, perimeter and centroid of the inner surface deformation when subjected to normal and shear forces at the point of interaction. A surface exploration computer algorithm to obtain the sensing information was developed in order to analyse the characteristic of the optical tactile sensor. The overall image analysis technique involves the following algorithms: image acquisition (capturing of images), processing (thresholding, noise filtering and boundary detection) and recognition (force measurement). The measured forces were then compared with the actual forces to determine the accuracy of the tactile sensor's characteristics. The image acquisition technique provides high quality image resolution while the image analysis algorithm generated a measurement that has good accuracy and recognized image consistently. The outcome of the research showed that the functionality of the newly developed image computer algorithm analysis coupled with the silicone tactile sensor is able to facilitate in biomedical applications such as in endoscopic surgery for measurement of tissue softness thus overcoming some of the current limitations of tactile sensor.