Showing posts with label image processing. Show all posts
Showing posts with label image processing. Show all posts

Tuesday

An Experiment in Photography - Saturation and Desaturation

The modern digital camera is an incredible tool, but like all cameras it is really just an instrument to collect data.  While many camera manufacturers make claims about the ability of their camera to make the perfect picture, in reality it is the person who makes the image using a camera to collect data about the scene.  Once the data is collected it must be processed to create a final image that reflects the photographers intent, weather it be a realistic reproduction of a scene or imparting an idea or emotion to the viewer.  What used to take place in a dark room now takes place on a computer with a digital camera.  Specialized equipment was replaced with equipment available to everyone.  The results are many mediocre images made with digital artistic applications, but also some truly spectacular images of the world around us created by everyday people.

The creative act of photography begins not with a camera but with the human mind.  For me it is the visualization of a final image while looking at a scene.  Observing the play of light, shadow, and color and realizing the potential for an image.  On many occasions, whether flying or walking I will suddenly see an image while just looking around.  It will flash in my mind as my head is turning and I will stop, back up and find the specific view that grabbed my attention and being composing an image in my mind.  Then it is out with the camera and start collecting data.  I try to cover the whole scene, overlapping the edges with several images to ensure I miss nothing that later might be useful during data analysis and realizing that the final image will be cropped into a proper and pleasing scene.

Once the data is collected it is back to the computer where all the images are combined with an image compositor into a single large image of the scene.  The image is then examined and cropped to produce a final view which began in my mind.  The next step is processing which involves a decision on whether the scene is to be a reproduction of reality or will try to impart an idea or emotion.  I generally strive to communicate an idea or emotion with my images, with the goal of highlighting the beauty of the Chiricahua and Pelocillo mountains.

The digital manipulation of color is frowned upon by many but I would argue that: 1.  Everyone sees the world differently and my interpretation of reality does not necessarily match yours.  2.  The cameras ability to reproduce reality is flawed by it's software (it represents the view of someone else) and its' hardware.  While I generally do not manipulate color, sometimes when trying to evoke an emotion I will alter saturation levels, either of all the colors simultaneously or individual colors to bring out a specific highlights.  The image below is one I have been working on for about a month, exploring different methods of presentation to evoke different ideas.  In this case the original dataset I collected was altered by selectively saturating or desaturating specific colors which resulted in an artificial, but personally pleasing interpretation of the Chiricahua mountains.

Original 5-image panorama stitched and cropped but unprocessed.

Selective saturation and desaturation of individual colors in this panorama of the Chiricahua mountains (click image to enlarge).

Monday

Creating Depth in Photographs

I'm not a professional photographer, just a guy in the desert with an inexpensive automatic digital camera asking the question "how do people see"?  While the physiological process of vision is fairly well understood, the analysis and interpretation of visual information seems less well defined.  There appears to be variability in the cognitive processes used by humans in interpreting a visual image and one way to qualitatively understand this process is through photography.  In other words "what makes a pleasing image"?  In trying to understand "how people see" a number of techniques have been employed from low level aerial photography to change perspective, in combination with tilt-shift to draw attention to specific elements in a photograph, to multi-image aerial panoramas to provide the viewer with a "big picture" view of the landscape from a different perspective.  But one of the more challenging approaches is the creation of depth in a 2D photograph.  There are a number of approaches to creating depth in a photograph but the most satisfying approach I have found is in-camera layering.  Briefly, a scene is divided into a number of layers then observing differences in contrast, texture, color, and focus between the layers then composing the scene to maximize the differences to make an image.  By maximizing these layer differences it is possible to impart to the viewer a sense of depth in a 2 dimensional photograph without resorting to anaglyphs.  Below are several examples of successfully layered images.  In these images one or more components appear to stand out as a result in differences contrast, texture, color, or focus (depth of field).  It is important to point out all these images were all photographed in the auto mode so depth of field manipulations were not used.

peloncillo mountains hiking
Peloncillo Mountains outcrop



The image above (click to enlarge) is a 4 image panorama taken on a sunny day.  Although similar in color, the outcrop in the foreground is differentiated from the valley floor by differences in texture.  While it is differentiated by color from the Chiricahua mountains in the background resulting in the appearance of an outcrop of rock that appears to float on the background.

creating depth in photographs
Hill in Animas Valley in southern New Mexico


This image is a low level aerial photograph with a tilt-shift effect added.  While employing an overall difference in perspective, in this case differences in contrast between the layers (shadow along the ridge top in comparison to the background desert floor) creates an edge resulting in increased depth.  This effect is enhanced by the tilt-shit which draws the viewer towards the area of increased depth which helps the viewer interpret increased depth in the photograph overall.


photograph of new mexico mountains
Low level aerial view of the Big Hatchet mountains
This image, also a low level aerial photograph of the landscape, relies on layered differences in color to enhance depth.  The foreground layer is brown while the background layer, with Big Hatchet Peak, is blue in color.  This layered color difference helps the foreground peak stand out against the background again resulting in increased depth.

Jhus canyon grave
Headstone in Jhus canyon



This final image, of a headstone in the Chiricahua mountains, uses color and texture differences to impart a sense of depth to the image.  The flat smooth brown headstone is contrasted against a background of shrubs and grass which differ in both color and texture again imparting a sense of depth to the image. 

This technique works at any scale from small to large landscapes, and is just a matter of composition.

While a number of different post production processing techniques may be used on a photograph, generally on a slight decrease in input levels and a slight increase in contrast suffice to bring out the layered depth in a composition.