The most successful work has been the most brilliant science writer, Cosmos Carl Sagan, the author relates:
\u0026lt;\u0026lt;"the Backbone of Night", as if the sky was a large animal in which we live ... The Kung believe that the Milky Way holds the night, but for the Milky Way, dark pieces fall, breaking at our feet. It's a smart idea>>
The following image is inspired by the paintings of Jon Lomberg, published in that book, which describes a metaphor about the nature of the Milky Way told Kung people of the Republic of Botswana. I recommend you view the image at a minimum distance of 1.5-2 meters above the monitor:
More information on how we obtained the final result can be found at: http://www
.astrosurf.com/astro35mm/articulos/procesamiento/procesamiento2/procesamiento2.htm
Well, even so, there are several reasons why the stars do not appear at the end of processing as bright as in the raw image. First, there are several processing techniques with multi-scale wavelet tend to produce that effect. Fortunately it is something that has easy solution, because usually only a matter of giving more or less emphasis on small-scale components, which normally are present most stars. Second, this effect can be caused also by the use of filters at least, and if not handled carefully, the stars tend to leave "dull", which again they lose some of their role. And finally, this aspect can be generated simply by the desire of the author.
When running a portrait photographer for instance, always seeks to highlight the subject from the background, blurring it and preventing any other element can distract the viewer's attention. When photographing deep sky same thing happens, there is a subject and a background. The subject here is the Milky Way and the rest is the background of the sky and many stars in the field. Therefore, if you want your taxable gain greater prominence on the scene, need to reduce the visual impact generated by the stars.
Speaking of pictures, than the mere physical identification (photo ID card, for example), the portrait is intended to reveal the character, feelings, moods, etc ... However, if our intention is to reveal all these features at once, most likely the subject will not be too encouraged. Of course the Milky Cap is not an animate subject and lack of character, or mood. But we do know one very own, such as its dimensions. The Milky Way is a galaxy that is presented before our eyes like an unusually large structure. Therefore, it is in the larger scales where it can be useful to have a more in our processing, it is a quality we want to highlight of our subject, and because our intention is not just a mere physical identification of the object.
On the other hand, there are many forms and ways to process an image. We can be proud to have flexible tools that allow us to express ourselves freely and be creative. Of course, the most important part of processing this image of the Milky Way with a 50mm lens, are the last two steps, in which one tries to enhance both the dark structures of small and medium scale, such as large-scale bright . Logically explain this requires an extensive article with step by step examples. What I think I have already commented here about this type of image processing, is that when processing the images with the wavelets, and try to give more importance or weight to the upper layers (large scale), you just realizing that problem is that stars, especially the brightest, for reflections in the upper layers of wavelets. This means that if you increase the bias in layers where there are no stars, they also exert some influence, creating strange halos. This effect is caused by the fact that the scaling function in wavelet can not completely isolate the stars in certain layers. Fortunately there is an easy solution: subtract the stars, to isolate. You simply can not disable the lower layers wavelet (small scale) because the dark halos affect the result. We must do the opposite: duplicate the image and her first disable all the upper layers, for example, 32 pixels. Then, it maximizes the layers survivors deringing and applied. This will generate an image with the stars alone, without the negative halo. Then you must subtract this image without scaling and then re-add, this time Rescaling. In between subtraction and addition is introduced all the processes to apply to the picture without stars.
In most cases this is usually sufficient to enhance the bright structures. But what about the dark structures? In that case we need to invest the image with no stars before and after processing structures, so that as in the image do not have stars, or their presence is minimal, increasing the influence deringing now avoid generating bright structures.
must be remembered also, for example, that if you want only to isolate the components of small or large, should be applied to image simultaneously the wavelets on the luminance and chrominance, except when they want to enhance details or structures should then be applied only on luminance and then combine the result with the color data (RGB). Finally, the masks play a role. Actually are you at all times to control which areas or components of the image are processed and which not. In the end, the further away you get, the more sophisticated end up being the masks. Trying this picture of the Milky Way at times I found the need to combine several masks at the same time in order to protect small-scale components present in highly lit areas, while protecting against the sky while enhancing high-level details.
Be happy,
0 comments:
Post a Comment