Cat on the wall -1

Blogs and comments are the best way to get people talking. I like to listen to people discussing on a few murky aspects in remote sensing.

The first debate is on "Will contrast enhancement improve supervised classification accuracy?"

Preamble:

Analysts perform contrast enhancement to better visualize their objects of interest. So would this enhancement improve the between-class separability?

Scenario 1:

The analyst contrast stretches the image and chooses the training pixels. He burns the LUT (look up table) to create a new image. He uses the same training pixels for classification in the original image and on the stretched image.

Scenario 2:

The analyst uses a different set of traning pixels on both the images.

I would like to know your opinion on the results of the two scenarios. Which one would you consider better and the reason.

Above the abode of clouds

My two-month long training programme came to an jolting halt. After the good byes and contact exchanges, I was left stranded in the Ahmedabad airport. I was the lone person from the group leaving to Bangalore. I patiently waited for two long hours for my much coveted window seat. It certainly seemed childish for the co-passengers to watch a grown up fellow peeping only through the window all through the journey. Yet, I was generous enough to let them look through the window during the takeoff and landing; perhaps they had an idea of waving to their loved ones as they flew over their rooftops.

  Air travel is like physics practical to me. Every time I fly, I learn a lot new things, encounter several natural phenomena and get several questions answered. This time is no exception. We took off in the early evening at 4:30 pm, the sun shone brightly at midway to the horizon from the zenith. We were travelling due south and the sun was to my right. I certainly had to thank Suganya for booking the seat on the sunny side; although, my co-passengers had a tough time getting sleep, as mine was the only window open in the entire right side of the plane.

Much as expected, the skies of the peninsular India were 

 blanketed with clouds. Most of them perhaps should be of convective origin. While most of them were seen stitched to one another, there were a 

few gaps and individual cloud groups formed a spectacle in the sky. The sun being the only source of illumination, but from an infinite distance sends its rays parallel to Earth. These clouds with high albedo and density remain largely opaque to the light. The result, beautiful parallel rays of light, just as you see in devotional pictures.

At 35,000 feet (11,666 meters) above the ground, we were flying at the higher end of the troposphere. The temperature as the captain echoed was sub zero outside and the pressure was considerably less. Only at these low pressures, or less denser atmospheres, we get the chance to see, of what I would call an invaluable picture. The sun above the horizon shone as a clear white ball, the clouds appeared white, the sky above was blue, but the reflection of the sun from the ocean below was red! How could this happen? To me, at 11.6 km above the ground, sun shines directly and appears least distorted in white, but to the waves on the beach, the rays have to inevitably pass through the denser atmosphere. Further, the evening condition takes the rays through a longer path through the atmosphere where wavelengths up to green get scattered as per Rayleigh law and only reddish orange light reaches. This light further has to pass through the same atmosphere to reach my eye where after additional scattering appears reddish.

As we moved a little further, the cloud cover started growing denser. Soon, it was beyond a point where I could no longer see any ground. These low altitude dense clouds should have been cumulous. They had a fluffy appearance and the evening sun shining obliquely caused small shadows adding to their coarse texture. All of a sudden, I wondered if I were flying above the arctic pole! These very much resembled the polar ice caps.

About an hour passed since we started, and I guessed we must have been at the mid section of the peninsula. The cloud cover had died with the calm interior of the Arabian sea beneath. Although the sea surface in general had a moderate brightness due to scattering because of the wave induced surface roughness, there were a few apparently dark spots seen. As we kept flying, my look angle changed, and when I came directly in line with them, they shone off too brightly. It was just like a mirror. Only then I understood these must have been oil spills. 

These notorious spills form a layer on the waters which are comparatively static and hence behave as specular reflectors. Further, they inhibit the mixing of gases with the sea water at the boundary layer reducing the dissolved oxygen content and causing a potential peril to the aquatic lives.

I wondered whether at this height, I could see a boat or a ship; a little later I found the answer to be yes and no. Although it is not possible to see a small boat, one can see the trails it leaves behind. Generally, the trails are much larger than the boat itself and because of the churning, they appear white and this can be seen from the sky. I recollected how satellite images are used for surveillance; while they cannot resolve tiny boats, they effectively pick up their trails and help calculate the velocity and direction of the boats.

Another hour passed, at 6:30 pm the earth has rotated considerably and sun had declined below the ground horizon; its rays now had to travel through the notorious atmosphere reaching me in red. I feel it is this atmosphere which makes our planet special, for life to exist and climates to sustain. This seemingly transparent and innocuous layer has telltale effects on our lives. Atmosphere means differently to different people, for me, it is a domain less understood, less modeled, an adventure and my area of research.