Kinetic inductance bolometer (KIB) technology is a candidate for scalable submillimeter wave imaging systems, particularly suitable for person security screening applications. We have previously shown that the basic figures of merit are compatible with room-temperature radiometric imaging applications, and demonstrated the functionality of kilo-pixel detector arrays. In this article, we report on our imaging system based on 8208 KIBs organized on a 2D focal plane. We provide an overview on the basic components, including the detectors, optics, and cryogenics, and describe aspects relevant in system integration. Moreover, we demonstrate the capacity in actual concealed object detection by presenting datasets revealing metallic and dielectric objects hidden under the clothes of a test person.

Passive imaging cameras at sub-millimeter wavelengths with large format focal plane arrays are being developed as the next generation of security screening systems. In this contribution, a dual-band focal plane array (FPA) for security imagers at submillimeter wave frequencies is presented. The detectors are based on bolometric superconducting kinetic inductance resonators, which allows the development of large FPAs at medium cooled temperatures. Two frequency selective absorber (FSA) sets coupled to superconductive resonator lines are designed to implement a dual color security imager. The performance of the dual band imager is evaluated using spectral analysis approach that combines Fourier optics with a Floquet mode field representation. The geometry of the unit cells is based on a Jerusalem cross configuration and the designed FSAs show a stable angular response and a rejection 1 to 3 of the undesired bandwidth. The detectors in the dual band FPA are distributed over a hexagonal grid to maximize their physical size and then improve their sensitivity. The effective point spread function of the imager coupled to a black body point source over a wide frequency band (1:6) was demonstrated experimentally with excellent agreement to the one estimated by using the proposed spectral technique.