Segmentation 3D Light Field Imaging for Object Detection

Object detection in machine vision and image processing has gained increasing interest due to its social and security potential. Earlier measurement methods suffer from drawbacks due to the complex scene properties (geometric, illuminations, etc.). To improve these limits, different techniques and approach have been proposed to yield efficient solutions. New ones are optical methods. These methods are effective and largely invaded in a computer vision task, in the purpose of describing objects and targets by the computing of, both, changes viewing position an illumination. The newest one was the plenoptic imaging.Plenoptic imaging is a promising optical technique, that furnishes substantial information of the object image or a full 3D scene. In fact, this technique captures information on a light field imaging of the appropriate scene and records them in a single shot. Computed location and the propagation direction information of the object light are used as efficient descriptors to detect and track the object position and displacement.