Highly sensitive infrared imager with direct optical readout

Abstract

We describe an uncooled IR camera that is based on a Fabry-Perot Interferometer (FPI) IR-to-visible transducer. The FPI-based IR camera converts a thermal-IR image to a video electronic image. IR radiation, emitted by an object in the scene, is imaged onto an IR-absorbing material that is located within an FPI. Spatial variations in temperature of the scene translate into corresponding temperature variations in the IR-absorbing material, forming a temperature image in the FPI. Within the FPI, the temperature variations produce variations in optical thickness for any beam of collimated visible light that is reflected from the FPI. The intensity of visible light reflected by the FPI is a function of optical thickness and thus forms an image, with thickness variations translating into intensity variations. The reflected light traverses visible optics that image the IR-absorbing material onto a visible-detector array, where the reflected light is converted into an electronic image. We will describe in detail the various sources of noise that determine the noise-equivalent temperature difference (NETD) of an FPI-based infrared camera. The dominant sources of noise are (1) shot noise in the visible-detector array and (2) temperature fluctuations (thermal noise) in the transducer. For a typical CCD array, we project a total NETD of approximately 40-50 mK for an FPI-based IR camera that is configured so that shot and thermal noise contribute approximately equally to the noise.