Using Optical Transfer Function and Fuzzy-Neuro Logic for a 3-Dimensional (3D) Camera

The currently available 3-dimensional (3D) cameras are quite expensive because they employ technologies like using three separate lasers to capture XYZ and RGB values, the triangulation method using laser scanner etc. We propose passive methods as a solution to decrease the high cost involved in 3D cameras. For this purpose, we have developed a focus measure and an approximation technique. The focus measure is based on an optical transfer function and it has shown robustness in the presence of noise. For the approximation technique, our algorithm is based on fuzzy-neuro approach. A fuzzy inference system (FIS) is designed and trained using the neural network for the calculation of the depth map. The proposed focus measure and approximation technique can further be employed in any of the passive methods to acquire depth map for a 3D camera