Object Recognition By, 3-Dimensional Curve Matching (Classic Reprint) - Softcover

Inhaltsangabe

Unlock how 3-D curves from range and intensity data help identify objects. This book explains a practical approach to recognizing 3-D shapes by matching curves extracted from registered range and intensity images. It shows how simple line-like features can determine size, shape, and position in space.

Readers will see how a curve-based matching method compares observed 3-D curves to known models, using a least-squares measure to pick the best fit. The discussion covers how piecewise linear curves are formed from real data and how a range sensor provides the essential registered images that make these curves possible. Examples come from experiments with everyday shapes and objects, illustrating both the ideas and the limits of the technology.

• How 3-D curves are extracted from aligned range and intensity data


• How a matching algorithm compares curves to identify objects


• How noise, calibration, and sensor design affect accuracy and results


• What experiments reveal about identifying objects from partial curve data

Ideal for readers of robotics, computer vision, and human‑made object recognition, this edition offers a clear path from theory to practical sensing and matching.

Die Inhaltsangabe kann sich auf eine andere Ausgabe dieses Titels beziehen.