Over the years infrared imaging technology has improved significantly and many new applications, such as multicolor infrared imaging, thermal medical diagnostics, and remote sensing, have emerged. This growth has been made possible by dramatic technical advances in the infrared sensor manufacturing as well as remarkable improvements in algorithms for image recognition, analysis and processing. However, the readout integrated circuit (ROIC), which is the interface between infrared sensors and the post-processing unit, has shown little improvement over the years. This can be attributed, for the most part, to the fact that ROICs have been designed traditionally as an application-specific integrated circuit (ASIC), which cannot support reconfiguration nor adaptivity. The main objective of this project is to design and prototype an intelligent readout integrated circuit (iROIC) with built-in programmable analog blocks, capable of performing basic spatio-temporal image-processing and image-recognition operations utilizing temporal and spatial (at the pixel-level) tunable bias unit cell. This project consists of three main components: 1) design and fabrication of iROIC, 2) development of spatial-temporal-spectral processing and detector tuning algorithms to be used for iROIC, and 3) fabrication and growth of the quantum dot in well detectors and hybridization onto the iROIC.
Intellectual Merit: The proposed iROIC has two distinctive features that no existing ROIC can offer: 1) pixel-level tunable bias, and 2) built-in analog signal processing units. These unique features have the potential to fundamentally change the way data-exploiting and post-processing imaging are developed. For instance, currently, to create a desired spectral response, multiple images must be taken of the scene. Utilizing the pixel-level tunable bias, one can take multiple spectral responses within one frame. Moreover, the proposed technology allows a single camera system to be used for panchromatic imagery, multi-color imagery at arbitrary wavelengths, multispectral imagery with arbitrary numbers, locations, and widths of spectral bands, and coarse resolution hyperspectral imagery. The additional features in the iROIC technology provide enormous flexibility to the electronic imaging systems, which, in turn, allow us to implement novel image processing algorithms in-situ.
Broader Impact: The proposed research and education plan revolves around broadening participation from underrepresented groups in Mathematics, Science, Engineering and Technology by involving students, such as minorities and women in the research and educational activities. At the University of New Mexico, approximately 40% of the undergraduate students belong to minority groups such as Hispanics, African-American, Native-American and members of the various Indian tribes. Also, roughly 40% of our undergraduate students are women. These facts provide an excellent opportunity for faculty to involve traditionally underrepresented groups and enhance diversity in their research and education program. Furthermore, the effort proposed here will have a significant impact on graduate and undergraduate education and training in the ECE department at UNM. The research team represents an interdisciplinary cross section of several important areas within electrical engineering, bringing together work in optoelectronics, analog/digital VLSI design, and signal/image processing.
