It's Idle time here. I gave myself a "vacation" from blogging last week for the Eid holiday. Not much has been going on except for some shots done in the family gathering - and they were not as vibrant as the years before. I, in fact, spent many days in the holiday sleeping for long hours (8 to 10 hours a day!). I guess I'm compensating for all the sleeping problems I've faced during Ramadhan. Meanwhile, I decided to do a little study on some aspect of tone-splitting technique. This study is not scientific in any way, but it is mainly observational to take some notes. I wonder though if there is a way to "scientify" this process after all.
Since I've been working in Tone-splitting techniques for some time, I wanted to test some matters related to it. Yet, I'm not sure how to make a scientific and bold approach to such matter, since they are artistic in the first place. Thus, the sole reasoning in this experiment is done by heart and feelings initiated by eyes vision. The task here is to investigate the relation between luminance colors and their usage in tone-splitting process; how would they appear and feel when used as one of the tones in the process. But before going through rummaging, I will talk a bit about the process.
Concept
In the beginning, the general idea here is to use 2 colors: a specific chosen color from part of the image, and its complimentary color. Now, these two colors can be bright and dull sometimes, according to how we see it, but the main question was, how to specify this? What makes a color lustrous more than the other? If we have two complimentary colors, they are just opposites on color circle, and the difference between the two is barely in the Hue value, while Saturation and Brightness (or Lightness) have the same value. After searching around and with the help of the colors websites I have at hand, I've found out that the most adequate tool to compare these two is by checking their CIE-LAB values (LAB for short) - and more specifically, the "L" value (which stands for Luminance).
After doing some experiments and comparing various colors, seems my guessing was right: The "L" value is indeed responsible for how "bright" the color is as we see it. So, what now?
Work
The work now was to choose a specific color. In the beginning, I was kind of in dilemma for what color should I choose in each image, as I didn't want to make the whole set is based on two complimentary colors alone - but it should be something of general concept. In the beginning, I tried to use the Average Blur command which makes an average of all the pixels in the image into one color. Sounded a good idea, but unfortunately, the colors resulting from this process were mostly dull and with low L-value, and their complimentary colors have the same characteristics. Thus, I had to do it in the usual manner as I usually do the tone splitting process for some of my images; by picking a lighted area in the image itself and analyze its color, and then pick the complimentary for that color. The two are used then to do the usual job of tone-splitting for the image.
I won't go on with the manual procedure of tone-splitting with the images, but to make things standard as much as possible, the random-picked images were turned into Black and White and using Maximum White preset (which can be found in the menu of the command in Photoshop). This preset sets all values for all tones to 100. Usually in my regular work in tone splitting I would play around with these values to pronounce specific details more than the others, but since I'm going to test several random images here, it is better to make something as a basic procedure.
Now according to these procedures and after picking a color and its complimentary color from the available colors in the image itself, we are in front of two cases, which I called "AB" and "BA" for easiness and affluence, where:
Thus, for each image picked here, we have two cases; one for AB and one for BA.
Vergleichung
Since I've been working in Tone-splitting techniques for some time, I wanted to test some matters related to it. Yet, I'm not sure how to make a scientific and bold approach to such matter, since they are artistic in the first place. Thus, the sole reasoning in this experiment is done by heart and feelings initiated by eyes vision. The task here is to investigate the relation between luminance colors and their usage in tone-splitting process; how would they appear and feel when used as one of the tones in the process. But before going through rummaging, I will talk a bit about the process.
Concept
In the beginning, the general idea here is to use 2 colors: a specific chosen color from part of the image, and its complimentary color. Now, these two colors can be bright and dull sometimes, according to how we see it, but the main question was, how to specify this? What makes a color lustrous more than the other? If we have two complimentary colors, they are just opposites on color circle, and the difference between the two is barely in the Hue value, while Saturation and Brightness (or Lightness) have the same value. After searching around and with the help of the colors websites I have at hand, I've found out that the most adequate tool to compare these two is by checking their CIE-LAB values (LAB for short) - and more specifically, the "L" value (which stands for Luminance).
 |
| Portion from ColorHexa website showing the LAB values (in Red). The Luminance value which I've used for comparison between colors marked in blue. |
After doing some experiments and comparing various colors, seems my guessing was right: The "L" value is indeed responsible for how "bright" the color is as we see it. So, what now?
Work
The work now was to choose a specific color. In the beginning, I was kind of in dilemma for what color should I choose in each image, as I didn't want to make the whole set is based on two complimentary colors alone - but it should be something of general concept. In the beginning, I tried to use the Average Blur command which makes an average of all the pixels in the image into one color. Sounded a good idea, but unfortunately, the colors resulting from this process were mostly dull and with low L-value, and their complimentary colors have the same characteristics. Thus, I had to do it in the usual manner as I usually do the tone splitting process for some of my images; by picking a lighted area in the image itself and analyze its color, and then pick the complimentary for that color. The two are used then to do the usual job of tone-splitting for the image.
I won't go on with the manual procedure of tone-splitting with the images, but to make things standard as much as possible, the random-picked images were turned into Black and White and using Maximum White preset (which can be found in the menu of the command in Photoshop). This preset sets all values for all tones to 100. Usually in my regular work in tone splitting I would play around with these values to pronounce specific details more than the others, but since I'm going to test several random images here, it is better to make something as a basic procedure.
Now according to these procedures and after picking a color and its complimentary color from the available colors in the image itself, we are in front of two cases, which I called "AB" and "BA" for easiness and affluence, where:
- AB: The color of the higher luminance is assigned to the highlights, and the colors of the low luminance is assigned to the shadows.
- BA: The opposite of the above; the low luminance color is assigned to the highlights of the image, while the color of high luminance is assigned to the shadows.
The Images
The images for the test are picked randomly, from various folders and years. Nothing specific was set to be looking for; simply random. The images are below and they are minimized. You can click any for a bit larger view. The total is 12 images.
After checking for myself with these twelve images (which is not enough in fact), I've made a simple statistical count:
AB preferred: 58.3%
BA preferred: 16.6%
Undecided: 25%
I can't surely say that having a luminance color to be taking the place of highlights is a sure thing, but I would say that with the abundance of highlights in some image, and placing a luminous color when splitting the tones in place of the highlights, that would make pretty much a bright image to the extent that, sometimes, it eclipses the complimentary color given for the shadows - thus, probably in such instances of abundant highlight, the AB type is not advisable. On the other hand, the colors chosen would play a great role, but in this instance it is all the role of taste. Those 25% of the undecided, might be so for compositional factors I presume, or ought to a balance in the overall highlights and shadows in the image, making the eye sight torn in between elements of the image without really residing on a specific element. This is all I could think of in the time being, but sure it doesn't end here. I'm wondering what would the next step be? More experimenting? Notice that all of this experiment, from A to Z, is based on color and complimentary color relation - there are other schemes to be investigated as well. Anyway, for the time being maybe I can put a guide line for myself in the matter of tone-splitting which might go like:
And more testing should be coming on the way.In tone-splitting process, the color of the highest luminance, or "L" value in LAB system, is to be assigned to the highlights, and its complimentary would be assigned to the shadows, providing that highlights do not overwhelm shadows in the image. And that is a starting point for creating and contrasty and easy-to-notice image.
Finale
I was supposed to have some rest in this month, regarding my workplace which made it easy with the fingerprint attendance process. However, on Wednesday, I've received some letter asking that I should be doing the fingerprint attendance again after being told that I do have a exclusion permit till the end of the month. Such news really rose my blood pressure levels and added 10 more points to the scale of how much I do hate administrators and this workplace specifically. This, specifically, is an addition for really disrespecting this place even more. I just wish it is September already and I'm ready to go away from this trash just to smell some fresh air in Ireland.
On the other hand, I've checked out for another appointment for my eyes examination instead of the one which was during Ramadhan and I didn't attend. Even though it could have been in the next week but I told the nurse to put it some time in November. The appointment is supposed to take place on November 16th. A special date if you ask me!
 |
| Source: Amazon |
 |
| Source: Amazon |
For now, I've stopped my extreme macro experiments. I still do have problems in lighting opaque subjects from the front, and I couldn't come up with a proper solution specially with my slow progress and my lessened attention to this matter. For this reason I've decided to stop it for now and think of some real photography to do in the few coming days, when possible.
My life looks like been punched with one big vacuum in the middle. Missing out some elements which are hard to be found any where, any time soon. All what I'm hoping for for the time being is to have one nice vacation. If someone doesn't feel bright about work, and doesn't feel at home in home, where would one go? I wonder...
No comments:
Post a Comment