Recognizing Techniques for 3D Objects on a Scene. Cet article n’est pas disponible.Recognizing Techniques for 3D Objects on a Scene

Aim of this book: To propose a method to recognise 3D object from a single silhouette image. To recognise an object many problems must be solved like size change, translation, rotation around the three axes, partial occlusion, low intensity of light as well as the deformation of the shape. We propose to use the invariant Fourier descriptors coefficients from the contour of the projected area of the object and back propagation neural network. We made use of Fourier descriptors coefficients and back propagation neural network with different numbers of hidden layers to build the optimal classifier of 3D pattern from a single silhouette image. The recognised objects are exposed to different intensities of light, are partially occluded, with size change, translation, rotation about all the axes and we used also deformed shapes. For an enhancement of the results and acceleration of learning and recognition, a principal component analyser (PCA) have been used. The Fourier descriptors are compressed by PCA, in order to be fed to the neural network.

Aim of this book: To propose a method to recognise 3D object from a single silhouette image. To recognise an object many problems must be solved like size change, translation, rotation around the three axes, partial occlusion, low intensity of light as well as the deformation of the shape. We propose to use the invariant Fourier descriptors coefficients from the contour of the projected area of the object and back propagation neural network. We made use of Fourier descriptors coefficients and back propagation neural network with different numbers of hidden layers to build the optimal classifier of 3D pattern from a single silhouette image. The recognised objects are exposed to different intensities of light, are partially occluded, with size change, translation, rotation about all the axes and we used also deformed shapes. For an enhancement of the results and acceleration of learning and recognition, a principal component analyser (PCA) have been used. The Fourier descriptors are compressed by PCA, in order to be fed to the neural network.