Dale came over and grabbed Carol. “You are not going to believe this,” he said, “what a bonanza. What a fucking bonanza!” He settled down a little. “But first things first. I promised myself that I would not spoil it for you. “He showed her to a seat at the conference table in front of the large screen and then sat down beside her.
He pushed the remote-control switch. Up on the large screen came a still frame of the three whales in the reef area under the boat. The fissure could clearly be seen to the right and beneath the whales. Dale looked at Carol. “I see,” she shrugged, “but what’s the deal? I took pictures with my underwater camera that are just as good.”
Dale turned back to the screen and pushed thc remote several more times. The successive scenes zoomed in on the hole in the coral reef, eventually isolating and centering on a small glint in the lower left side of the fissure. Again Dale looked at Carol. “I have a similar blowup,” she said pensively. “But it’s impossible to tell if something is really there or if it’s an artifact of the photographic process. “She stopped herself. “Although the fact that two distinctly different techniques found the light in essentially the same place suggests that it might not be a processing distortion.” She leaned forward, interested. “So what’s next?”
There was no way he could contain himself Dale jumped up and started pacing around the room. “What’s next,” he began, “could be your ticket to the Pulitzer dinner in New York. Now I am going to show you exactly the same sequence of images, only these were taken in the infrared a fraction of a second later. Watch closely, especially in the center of the fissure.”
The first processed infrared image covered the same area underneath the boat that the first visual image had shown. In the infrared picture, however, what was shown were thermal variations in the scene. In the processing, each pixel (an individual picture element in the image) was given a specific temperature based on the infrared radiation observed from that portion of the frame. Similar temperatures were then grouped together by the computer processing and assigned the same color. This process created isothermal regions, or regions of roughly the same temperature, that were visually connected by color. The result was that in the first picture the whales stood out in red, most of the reef plants were blue, and the normalized water temperature formed a dusky gray background. It took Carol a moment to adjust to the display. Dale was smiling triumphantly. Before Carol had a chance to focus on two small regions, one red and another brown down in the center of the hole in the reef, the zoom process had begun. In a few seconds an infrared close-up of the fissure clearly demonstrated why Dale was so excited.
“I told you there was something under the boat,” he said, walking to the screen and pointing at a brown, elongated object. The object was cylindrical at one end and tapered to a point at the other. The fissure had been blown up by the zoom process so that it almost completely filled the screen. Even with all the magnification, the quality of the infrared image was superb. Inside the opening three or four different colors could be seen; however, only two, the brown and the red, were continuous over a significant number of pixels
“Holy shit,” said Carol, involuntarily rising from her seat and walking over to join Dale, “that brown thing must be the lost missile. It was underneath us all the time. “ She picked up the pointer and waved it at the screen. “But what’s this red area? It looks like the Cheshire cat from Alice in Wonderland.”
“I’m not absolutely certain,” Dale replied, “and it’s probably not anything of major significance. But I do have a crazy idea. Actually it’s based on what you told me about the strange behavior of the whales down there. It may be the head of another whale, back away from the light, looking out of the cave. Or whatever the opening is. Here, look at this. By zooming out a little we obtain one single picture that shows both of the red isothermal regions. Look how the red region in the middle of the fissure and the red from your sentinel whales look the same. Even with additional stretching, the two regions remain comparable in temperature. Not a proof of any kind, but it certainly supports my proposition.”
Carol’s mind was racing ahead. She was already planning her next move. It was essential that she retrieve that missile before anybody knew it was there. She needed to return to Key West as soon as possible. She picked up her purse and her briefcase. “Can someone drive me to the airport, please, Dale? Right now. I want to call that Lieutenant Todd again and scare him a bit. You know, make him a little more cautious and buy some time for us.”
She paused, thinking of a million things at once. “But I can’t call him from here without making him suspicious . . . And I must make some arrangements for a boat for tomorrow . . . Oh, incidentally, I assume you have hard copy of those pictures available for me.”
Dale nodded his head. “I do,” he said. “But first sit down and relax for a second. I want to show you something else. I don’t yet know if it’s a real phenomenon, but if it is . . .” Carol started to protest but there was something in his manner that told her to acquiesce. She sat down. He launched into a discussion of enhancement algorithms, explaining how the information in pictures could be stretched to highlight special features and allow easier interpretation.
“Okay, Okay,” she said at length. “The bottom line is what I need. I know already how clever you and your engineers are.”
Dale put the first infrared image back on the screen, the one that showed the full view of the three whales underneath the boat. “This picture does not have much thermal granularity. Every pixel in the region colored red, for example, does not correspond to exactly the same temperature. In reality, the spread in temperatures for the same color is roughly five degrees. Now if we stretch the image, and make the isothermal regions only cover a total spread of two degrees each, we obtain this picture.”
In the new image there were ten different colors. It was much harder to see individual features, and spurious data points made the picture extremely difficult to interpret. A portion of the front of one of the whales was now a different color from the rest of the animal.
“The limit of accuracy of the equipment, by the time the raw spectral data is converted to temperatures, is about one degree. If we show another stretch of the same picture, with the connected isothermal regions now only covering a total range of one degree each, then the picture almost becomes gibberish. Now there are twenty different colors for the isothermal regions and, because the noise or error in each data point is of the same magnitude as the spread in the isothermal region, it is virtually impossible to see the figures of known objects like the three whales. I tell you all this up front to make certain you realize that what I am about to show you may be completely wrong. It is, nevertheless, absolutely fascinating.”
The next image projected on the screen was a close-up down on the floor of the ocean, just above the trench that Carol had followed when backtracking to find the origin of the tracks. The familiar parallel lines just barely showed up in the infrared image The fissure was almost off the left side of the image. On either side of the trench, blue color broken with some occasional green marked the two reefs. Carol looked at Dale with a puzzled expression on her face.
“This close-up has the same five-degree granularity as the big reference image. There is nothing of note here.” He flashed another picture. “Nor here, where we have increased the number of colors to ten again. But look at this.” One more image went up on the screen. The picture was very difficult to follow, much less interpret. As many as twenty different colors connected odd regions in what appeared to be random patterns. About the only thing that was regular in the picture were the background rocks on which the coral and other sea life were living. And it was those background rocks that had Dale so excited.
“This is what I wanted you to see,” he said, waving his hand at the rocks on the two sides of the trench. “The two reef structures do not have the same color. For some unknown and absolutely inexplicable reason, every background rock area on this reef is coded chartreuse. On the opposite reef, just across the trench a few feet away, all the background rock is yellow. A one-degree difference. Now if some of the yellow pieces were interspersed with the chartreuse, and vice versa, then I would say that the data clearly has no significance and that what we are seeing are noise signatures. But this pattern is compelling.”