No abstract

No abstract

No abstract

A demonstration of a real-time image correlator based upon acousto-optic and charge-coupled devices will be presented. This architecture performs the two-dimensional correlation utilizing one-dimensional input devices. Each input image line is read into the system one line at a time using an acousto-optic device (AOD). The reference templates are stored in an electronic memory and introduced into the optical system in row parallel using a one-dimensional laser diode array. As the information propagates down the length of the AOD it is correlated against all of the rows of the reference template; the result accumulates on a 2-D charge-coupled device (CCD). The correlation in the vertical direction is accomplished by operating the CCD in a shift-and-add mode. The demonstration involves tracking a moving grayscale object using a video cassette recorder as the input image source. The ability to rapidly correlate with multiple templates will also be demonstrated. The optics are mounted inside a 12 in. × 15 in. × 6 in. enclosure. The memory and control electronics are housed in a single VME card cage.

Binary phase -only filters (BPOFs) are a viable candidate for the replacement of matched filters in real -time image processing and pattern recognition applications. The original BPOFs were binarized versions of the real or imaginary parts of the Fourier transform. Recently, a filter has been proposed that is the binarized Hartley transform. In this paper, the noise performance of the binarized Hartley phase -only filter is analyzed and compared with other BPOFs as well as with the phase -only and matched filters. It is well known that the matched filter optimizes the signal -to -noise ratio. Further, it can be shown that the matched filter phase is the optimum phase -only filter. A bound on the BPOF signal -to -noise ratio relative to that of the phase -only filter is derived. Finally, the analysis is applied to the generalized form of the binarized Hartley transform.