Specifically, the spin polarization of atoms trapped near the surface of a microfabricated semiconductor chip will be controlled, eventually to transport the atom spin along the surface of the chip. Ultimately, this could lead to an atom-based spin transistor. Spin-polarized atoms will open up more possibilities for advanced technology in computation, signal processing, and sensing than are available with spin polarized electrons. Coherent control and entanglement of neutral atom qubits confined on an atom chip by means of an optical frequency comb can potentially replace the well-known trapped ions. This technology relies on a laser to cool atoms to very low temperatures. The temperature achieved in a laboratory demonstration of this technology was a remarkable 40 × 10–6 K.

Ultraviolet Lasers

SEDD, in collaboration with the University of Central Florida, has demonstrated high-quality AlGaN/Ga:MgZnO grown on sapphire, which is extremely important for implementing high-efficiency near-ultraviolet lasers.

Signal and Image Processing Division

A large part of the signal and image processing effort within SEDD is collaborative work with CISD as part of the Network Science Program. This research is discussed in Chapter 2. This section highlights additional projects within the Signal and Image Processing (S&IP) Division.

Image Processing for Facial Recognition

SEDD researchers have demonstrated a capability to enhance the performance of low-resolution security cameras so that the images can be used to improve the verification rate of automatic facial recognition algorithms. This important result will enable more effective use of security cameras to protect guarded facilities and will allow law enforcement to make far better use of security cameras to address serious threats.

A parallel effort has focused on devising machine recognition methods for detecting humans in stationary scenes. This collaborative effort draws on SEDD’s machine recognition skills and super-resolution technology and CISD’s social-cognitive work. Because human experts have almost always outperformed every other approach, the long-term promise of this technique is enormous. This work demonstrated the ability of recognition algorithms to identify humans in challenging scenes, such as those with partially obscured targets.

OPPORTUNITIES AND CHALLENGES

SEDD faces the challenge of building an innovation ecology that generates novel technology to enable the warfighter. In doing so, it needs to create an organic and bottom-up environment for its researchers to innovate, yet this environment should be constrained to insure that the needed technological innovation is there to support the warfighter. Balancing the need to constrain an open technology innovation environment is not an easy challenge. SEDD leadership is aware of the delicate balance that they need to have in place to catalyze innovation and the need to actively, but not intrusively, intervene to keep the ecology in synchrony with the SEDD mission.

The rotating SEDD directorship is not healthy for the organization. Talented division chiefs have served as director for 4-month intervals, which is not sufficient time to allow for (1) sustained and con-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement