Advanced algorithms for identifying targets from a three-dimensional reconstruction of sparse 3D ladar data

Berechet, Ion; Berginc, Gérard

doi:10.1117/12.896708

Cited by 7 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The idea is to apply a tomographic algorithm on reflective projections [13,11,6,2]: a filtered backprojection. It offers several ways of representing the scene [7,3,5,9] in three-dimensional optical imaging. The most intense values of the reconstruction are located near the original surfaces, up to artifacts.…”

Section: Reflective Tomographymentioning

confidence: 99%

Multiresolution greedy algorithm dedicated to reflective tomography

Bellet¹

2018

The SMAI Journal of Computational Mathematics

View full text Add to dashboard Cite

show abstract

Section: Reflective Tomographymentioning

confidence: 99%

Multiresolution greedy algorithm dedicated to reflective tomography

Bellet¹

2018

The SMAI Journal of Computational Mathematics

View full text Add to dashboard Cite

show abstract

“…We obtain the three-dimensional reconstruction by a cone-beam algorithm [5], [6], [7] , which is a convolution backprojection algorithm deduced from the Radon transform. This algorithm uses a set of two-dimensional projections which contain the data collected by the pixels of a focal plane area.…”

Section: D Reflective Tomographic Algorithm -Application For Non-conmentioning

confidence: 99%

“…Figure 7 shows an example of three-dimensional incomplete reconstruction. The implementation of a data-driving algorithmic process [6][7] [9] can help fill data in the incomplete areas and generate the object surface. This algorithmic process is based on partition of initial incomplete point clouds in significant areas by integration of cloud minimum energy, completion of incomplete significant areas and surface generation using a MultiLayer Perceptron (MLP) [14] approach with sensibility calculation for higher capacity of generalization and fusion of partially generated surfaces.…”

Section: D Completion Algorithm -Application For Sparse 3d Reconstrumentioning

confidence: 99%

3D laser imaging for concealed object identification

Berechet¹,

Berginc

Berechet³

2014

SPIE Proceedings

Self Cite

View full text Add to dashboard Cite

This paper deals with new optical non-conventional 3D laser imaging. Optical non-conventional imaging explores the advantages of laser imaging to form a three-dimensional image of the scene. 3D laser imaging can be used for threedimensional medical imaging, topography, surveillance, robotic vision because of ability to detect and recognize objects. In this paper, we present a 3D laser imaging for concealed object identification. The objective of this new 3D laser imaging is to provide the user a complete 3D reconstruction of the concealed object from available 2D data limited in number and with low representativeness. The 2D laser data used in this paper come from simulations that are based on the calculation of the laser interactions with the different interfaces of the scene of interest and from experimental results. We show the global 3D reconstruction procedures capable to separate objects from foliage and reconstruct a threedimensional image of the considered object. In this paper, we present examples of reconstruction and completion of three-dimensional images and we analyse the different parameters of the identification process such as resolution, the scenario of camouflage, noise impact and lacunarity degree.

show abstract

“…We shall focus on refl ective tomography algorithms for reconstructing optical three-dimensional scenes. [110][111][112] We develop a cone- reconstruction of the vehicle under trees. Trees are modeled by polarized refl ectance applied to the different facets generated by the CAD model.…”

Section: Applicationsmentioning

confidence: 99%