Utilizing remotely gathered imagery from Unmanned Aerial Vehicles (UAVs) and ground platforms of field-grown wheat breeding trials has the potential to improve response to selection by quantifying previously unmeasurable plant phenotypes. Digital phenotypes can precisely quantify phenotypic variation that were previously only classified by breeder qualitative ratings. Within wheat breeding programs, these phenotypes can be utilized directly as a selection criteria or to associate these with regions of the genome using structured mapping populations or unstructured diversity panels. We will present several examples of recent research (past three years) utilizing remotely gathered imagery to estimate yield and biomass, to detect stripe rust incidence and severity, green leaf area duration, spike count and plant spatial arrangement. We have utilized this imagery with a variety of techniques ranging from vegetation index analysis, conventional image analysis, and deep learning with convolutional neural networks. We will discuss the applicability of these techniques and how this information is being utilized in a large commercial wheat breeding program.