Low tide/high tide [multispectral]

[Aaron E.]

I'm finally getting to post-processing images from my trip to Singapore last month. This set is of the same scene of a mangrove swamp forest taken about 7 hours apart in time (left about 8am vs. right about 3pm) when the tide was low (left) and high (right). Rows (top to bottom) are: visible (KolariVision UV/IR-cut hot mirror); full spectrum (no filter); IR (Kolari 850-nm) and UV (Kolari UV bandpass). Lens is a Soligor 35-mm (KA66274) on a LifePixel-converted Nikon Z-8.

The UV levels in the equatorial sun were, according to my weather app, "extreme," and the white-balancing of the reflected UV images using PhotoNinja gave really weird results. Hence the monochrome views (which also allow better comparison with the IR-850nm image). I'd note that the white balancing was really only a problem of images taken in the forest understory. I'll post some photos later of various tropical flowers, for which I had no trouble doing white balancing.

 

This plate is very reduced in size and resolution. The original is 43 (w) x 104" (h). PDAF banding is not apparent in the images, either (something I was checking on this trip with my new Z8). The glowing "streak" across the middle of the UV image at low tide is a spider web - the actual reason for this scene. I've been taking photos of spider webs, which often glow in UV but are completely invisible in IR and visible spectrum.

 

 

[Aaron E.]

Also, these images are each a subset of larger panorama (6 frames across, each frame of which is a 5-image focus-stack set). The focus stack is responsible for the "ghost" branches in the high-tide UV image.

[colinbm]

Nice collection Aaron

[Andrea B.]

Hi Aaron - This is a wonderful set of images. It is especially cool to see UV/IR images of something many of us have never seen in person - a mangrove swamp! (There are some along the southern US coast.) There is good information in these photos about the reflectivities of such a habitat.

 

question -- in the Visible low tide photo (row1, col1), the swamp area seems to be purple? 

 

comment - in the full spectrum row, it is interesting to observe that the high tide image is much brighter than the low tide image. Then it "clicked" that the water causes more IR-reflectivity by reflecting the IR-reflective foliage. That was a bit convoluted, but I'm sure you get what I'm trying to say. "-) There is also some of this phenomenon in the third row of IR-850 images.

 

*******

Aaron wrote:  the white-balancing of the reflected UV images using PhotoNinja gave really weird results. 

 

Yes, it can occasionally happen that white balancing doesn't work quite like we think it should - even in a typically WB-useful app like Photo Ninja. If you would like me to run one (or more) of the raw reflected UV files through Raw Digger, I would be happy to do so. That sometimes can help to figure out what is going on. You can upload a raw NEF to this topic and I'll pick it up and run it.

 

 

[Aaron E.]

Hi Andrea,

 

Thanks for the comments/feedback!

 

Purple in vis low-tide. That's probably close to the actual color; a combination of low tide, exposed mud, solar reflectance, and various red algae. I haven't done anything with these images other than initial white balancing, focus-stacking, and panorama stitching. 

 

You're right that there's much more IR reflectivity at high tide. Conversely, there's much more UV absorbance (into the water) at high tide, which may explain some of my white-balancing problems.

 

I would be delighted if you wanted to run some of the raw reflected images through Raw Digger. The posted image is actually a crop from a 6-scene panorama, where each scene is 5 images focus-stacked. Rather than upload the 30 48Mb NEFs here, I'll put all of them in a GoogleDrive folder that you can access. As noted in the first post, the lens is a Soligor 35mm (which I'm likely to abandon for relatively wide shots as it's not nearly as sharp as my Steinheil 50mm), and the filter is a Kolari UV bandpass.

 

I'll get the google drive link up in a few hours...

 

Best,

Aaron

[Aaron E.]

HI Andrea,

 

Here's a link to a google drive folder (UV-NEFs) with subfolders \LowTide and \HighTide with 20 NEFs each. The main directory also has photos of PFTE card that I used to attempt white balancing of these images. The link is unrestricted, so you should also be able to write files back into it if you'd like.

 

At low tide, the sun was pointing into the camera, whereas at high tide, it was behind me.

 

Thanks for "digging"!

 

Aaron

[Andrea B.]

Hi Aaron -

Thank you for sharing your images! I was eager to look deeper into some Z8 full-spec files. And some files of an interesting habitat. This helps keep my knowledge, and by extension, UVP's knowledge up-to-date. I will work through some of the images over the next couple of days and get back to you here.

 

*****

Funny about your sharp Steinheil/50 and not-so-sharp Soligor/35. I have the opposite - a kinda messed up Steinheil/50 and a decent enough Soligor/35 !! But it's always difficult to know whether an older lens lacks sharpness (or micro-contrast) because aging wear-and-tear has thrown the lens off its original factory calibration or because the lens was never very sharp or contrasty even when new.

 

Sometimes if you send an older lens to be cleaned, lubed and re-calibrated, you can see small improvements in the images. However I cannot say that I have been consistent with the maintenance of my own set of old lenses!  I have sent some to Keh.com over the years for CLR and been quite pleased to see them returned all sparkly clean with smoother/tighter  aperture/focusing controls. 

 

 

 

[Andrea B.]

Sorry for the delay. I had to take a break to finish up my Income Tax work (due 15 April).

*******

 

Here are three sets of the High/Low Tide photos which I worked up as follows. For each set there are 4 results:

1) Raw composite from Raw Digger. This is a view showing the demosaiced file without the application of any white balance or camera color profiles.

2) Rgb render from Raw Digger. This is a view showing a white balance as applied by Raw Digger based on the characteristics of the sensor. Rgb render typically requires an increase in saturation. (Raw Digger is not a converter, so it applies only the most basic renderings.)

3) Conversion in Photo Ninja, B&W version. Because the dynamics of these files are so extreme, I made the basic exposure, black point, white point, highlight and shadow adjustments on a B&W version of the file. Only then did I restore the color and attempt to white balance it on a pinkish/magenta area of the file.

4) Conversion in Photo Ninja, color version. Yes, the outcome in PN is different than what Raw Digger produced.

 

These are tough files to convert, as Aaron has already mentioned. There is not a lot of light in the foreground. There are reflections bouncing around. And some flare is produced from the UV-bright sky. It is hard to say which rendering is the "proper" one. In reflected UV landscape conversions, we do see dark greens. But typically they are headed toward dark yellow-greens (i.e., like the Raw Digger version) and not toward dark blue-greens (i.e. like the Photo Ninja version.)

 

SET 1 High Tide

f/3.5 for 6" @ISO-400

 

Raw composite from Raw Digger

 

Rgb render from Raw Digger

 

B&W conversion from Photo Ninja

 

Color conversion from Photo Ninja

 

 

SET 2 High Tide

f/3.5 for 6" @ISO-400

 

Raw composite from Raw Digger

 

Rgb render from Raw Digger

 

B&W conversion from Photo Ninja

 

Color conversion from Photo Ninja

 

 

SET 3 Low Tide

I had no success with this set. Just too dark.

f/3.5 for 1/3" @ISO-400

 

Raw composite from Raw Digger

 

Rgb render from Raw Digger

 

B&W conversion from Photo Ninja

 

Color conversion from Photo Ninja

 

 

[Aaron E.]

Thanks for the work on these! I hadn't thought to do the PhotoNinja work on monochrome versions first and then converting back to color. In my own practice, I generally (>90% of the time) use monochrome versions of the UV images as the final/printed/production versions, but I do all the white balancing etc. on the color version before "desaturating" them. But the results are better with PN on b&w image. I will go back and reprocess the originals with that in mind.

 

I've never worked with Raw Digger before, but it looks very useful. I think they have a free trial I can test out before buying it.

 

In the end, though, monochrome vs. color, and what color, comes down to aesthetic decisions...

 

Thanks again!!!