UC Irvine

In this thesis, we discuss the theoretical background and the practical aspects of the design of a simulation toolchain for prototyping and testing new consumer imaging technologies during the design cycle. To enable a fully computer-aided design and test process, mechanisms are provided to model and investigate all the entities involved in the imaging system. In particular, modeling of the scenes, optics, and sensors are included.

In addition, the computational and storage bottlenecks of the simulator are examined and more efficient methods are provided when appropriate. In particular, a method based on Singular Value Decomposition (SVD) is proposed to lower the storage-footprint of the scene model. Furthermore, an efficient algorithm based on Principal Component Analysis (PCA) and Fast Fourier Transform (FFT) is proposed to implement the convolution of the scene with the spatially-varying optical kernels. The quality and complexity of the proposed method are compared with previous work and it is shown that higher quality is achievable with lower complexity.