CONCLUSIONS 



REFERENCES 



1. In a zero sea state, towing the two sonar fish at an 

 average depth of 17 feet and average speed of 2 knots 

 caused a random error of about 2.6 feet in the 43- 

 foot horizontal baseline. Random error in the relative 

 vertical separation of the two fish was around 7.5 

 feet. Even though the magnitude of these errors was 

 probably a function of the particular sonar fish used 

 in the investigation plus the physical parameters of 

 the tow cable, the outrigger, and the towing vessel, it 

 is concluded that, in same-side stereo, the depth of 

 each fish and the lateral separation of the two fish 

 must be continuously measured with a precision of 

 about 0.5 foot in 40 feet. 



2. In same-side stereo scanning, 100-kHz sonar imag- 

 ery will be suitable for fusing into a three-dimensional 

 illusion if the two fish are towed out of each 

 other's beam and the heave for an individual fish 

 is under 4 feet, peak-to-peak, at off-bottom heights 

 around 70 feet. 



1. J. R. Mittleman and R. J. Malloy. "Stereo 

 side-scan sonar imagery," in Proceedings of Seventh 

 Annual Conference of Marine Technology Society, 

 Washington, D.C., Aug. 16-18 1971, pp. 395-422. 



2. M. M. Thompson, Ed. Manual of photogrammetry, 

 3rd ed., vol. II. Falls Church, Va., American Society 

 of Photogrammetry, 1966, pp. 1030-1035. 



3. Naval Civil Engineering Laboratory. Technical 

 Report 787: Rotating acoustic stereo scanner for 

 positioning loads onto the seafloor: Preliminary 

 observations on an experimental model, by R. D. 

 Hitchcock. Port Hueneme, Calif., April 1973. 



RECOMMENDATIONS 



An analog signal-processing system should be 

 developed for producing an optical-stereo readout in 

 real time from side-looking sonar returns. At-sea 

 testing of this system should be conducted using a 

 commercially available, side-scan system interfaced 

 with conventional acoustic devices for continuously 

 measuring fish depths and fish separation. 



13 



