Exercise: Sensor linear capture

The aim of the exercise was to simulate a linear image by applying the opposite kind of curve. I opened a jpeg image I'd used in a previous exercise, it was an image that included some texture and shadow which I hoped woule demonsrate the excercise well. I used PSE elements and making sure the image was in 'full edit' mode selected Image > Enhance > Adjust colour > Enhance colour curves which brought up a new box with the before and after image with a linear curve graph and sliders to the side, in order to alter the curve I moved the sliders to adjust highlights, midtone brightness, midtone contrasts and adjust shadows. The sliders were moved to produce the desired effect. See image 1.


Image 1

The image below (Image 2) shows the histogram of the darker image as it is captured by the camera's sensor. For this linear image, most of the tones are squashed strongly to the left. This means that most of the levels available to represent tones are devoted to the brightest part of the image, while the darkest parts - the shadows - at the far left are actually represented by very few levels. This is close to what the image looked like as it was captured and before the camera's processor got to work on it. This has a very important impliction for noise which will be discovered in the next part of the course.

Image 2
 

The image below (Image 3), show what the image looks like after the camera's sensors have got to work on it, the the tones, light and dark, are more evenly spread across the histogram. 

Image 3

The next part of the exercise required a change to the linear curve of the darker image to bring it back close as possible to the original 'normal' looking version. The image below shows the re-edited curve of the darker image which demonsrated that the biggest effect has been on the darkest parts of the image. They have been lightened by what looks like several stops.

Image 4
 

If there is noise in an image, it is concentrated in the shadows. As captures, before processing in the camera, it is in a sense buried from view within the overall darkness. However, the strong curve that has to be applied to lighten the image to a normal appearance has the unfortunate side-effect of exaggerating this noise, because it lightens the shadow areas so strongly.

Although I found this exercise to be initially a bit frustrating due to the incorrect information in the notes, it was, after some trial and error, very satisfying. I persevered with the software and after finding the linear curve and working through the notes I was able to produce what was required. It has been a useful exercise to understand how the camera sees an image and how the sensor captures the image and linear curve. Changing the curve was useful in that I was able to see what effect the highlights, brightness, midtond contrast and shadows had on the curve and ultimately on the image.