This thesis presents a new method to extract depth information from stereo-vision acquisitions using a feature and a correlation based approaches. The main implementation of the proposed method is in the area of Autonomous Pick & Place, using a robotic manipulator. Current vision-guided robotics are still based on a priori training and teaching steps, and still suffer from long response time.
The study uses a stereo triangulation setup where two Charged Coupled Devices CCDs are arranged to acquire the scene from two different perspectives. The study discusses the details of two methods to calculate the depth; firstly a correlation matching routine is programmed using a Square Sum Difference SSD algorithm to search for the corresponding points from the left and the right images. The SSD is further modified using an adjustable Region Of Interest ROI along with a center of gravity based calculations. Furthermore, the two perspective images are rectified to reduce the required processing time. Secondly, a feature based approach is proposed to match the objects from the two perspectives. The proposed method implements a search kernel based on the 8-connected neighbor principle.
The reported error in depth using the feature method is found to be around 1.2 mm