{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

4.Object_Recognition - S Venkannah Mechanical and...

Info icon This preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
S. Venkannah Mechanical and Production Engineering Department UNIVERSITY OF MAURITIUS FACULTY OF ENGINEERING ROBOTICS TECHNOLOGY – MECH 4041 B. Eng (Hons.) Mechatronics Level 4 Prepared by S. Venkannah Object Recognition Not even the simplest machine vision tasks can be solved without the help of recognition. Pattern recognition is used for region and object classification. There is a need to make some decision about the content of an image or about the classification of an object that it contains: e.g. the user of a notebook computer may be able to give input using hand-printed characters the classification of an object- whether an ‘A’ or an ‘8’ would be based on the features of its optical image or perhaps of a pressure footprint which is also an image like representation. The classification process might actually fail because: The character is badly made The person has invented a new character Usually a reject class is included in a system design and the reject class might be examined again later at some higher level Imagine an ATM using a camera to verify that a current user is indeed authentic. Here the image of the current person’s face is to be matched to a stored image, or images, attached to the current account and stored either on a computer network or in the bank card itself. In another application a food market recognition system would classify fruits and vegetables placed on the checker’s scale. Object recognition The problem in object recognition is to determine which, if any, of a given set of objects appear in a given image or image sequence. Thus object recognition is a problem of matching models from a database with representations of those models extracted from the image luminance data. Early work involved the extraction of three-dimensional models from stereo data, but more recent work has concentrated on recognizing objects from geometric invariants extracted from the two-dimensional luminance data. 1 Of course, the representation of the object model is extremely important. Clearly, it is impossible to keep a database that has examples of every view of an object under every possible lighting Faculty of Engineering Robotics Technology MECH 4041 B. Eng (Hons.) Mechatronics
Image of page 1

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon