This paper proposes an object recognition and detection method bya combination of support vector machine classifier (SVM) and rotationinvariant phase-only correlation (RIPOC). SVM is a learning techniquethat is well founded in statistical learning theory. RIPOC is a positionand rotation invariant pattern matching technique. We combined these twotechniques to develop an augmented reality system. This system canrecognize and detect objects from image sequences without special imagemarks or sensors and show information about the objects through ahead-mounted display. Performance is real time
展开▼