Date of Award:


Document Type:


Degree Name:

Doctor of Philosophy (PhD)


Electrical and Computer Engineering

Committee Chair(s)

Scott E. Budge


Scott E. Budge


Jacob Gunther


Todd K. Moon


Rose Q. Hu


Marvin W. Halling


The term image is often used to denote a data format that records information about a scene’s color. This dissertation object focuses on a similar format for recording distance information about a scene, “depth images”. Depth images have been used extensively in consumer-level applications, such as Apple’s Face ID, based on depth images for face recognition.

However, depth images suffer from low precision and high errors, and some post-processing techniques need to be utilized to improve their quality. Deep learning, or neural networks, are frameworks that use a series of hierarchically arranged nonlinear networks to process input data. Although each layer of the network is limited in its capabilities, the learning capacity accumulated by the multilayer network becomes very powerful. This dissertation assembles two different deep learning frameworks to solve two different types of raw image preprocessing problems. The first network is the super-resolution network, a nonlinear interpolation of low-resolution deep images through the deep network to obtain high-resolution images. The second network is the inpainting network, which is used to mitigate the problem of losing specific pixel data in the original depth image for various reasons.

This dissertation presents deep images processed by these two frameworks, and the quality of the processed images is significantly improved compared to the original images. The great potential of deep learning techniques in the field of deep image processing is shown.