Three-dimensional location measurement system constructed by dual cameras

This paper presents a three-dimensional (3-D) location measurement system constructed by dual cameras. An artificial neural network (ANN) model is used to analyze the 3-D location measurement system implemented with twin pinhole cameras. Evaluation by simulated data set from a mathematic model and real data set demonstrates that the ANN model can be used for interpolation of the special coordinates. The mathematical model of the dual cameras can calculate the depth information of an object's position and can be used for simulation data to test the ANN model. The images captured by the dual cameras are used to estimate the depth information as the real data test. The ANN is trained by the simulated and real data. Back-propagation neural network (NN) is adopted as a function interpolation purpose. High relation exits between the input coordinates and the estimated coordinates for real and simulated data sets.