5. Results and discussions - radhikahorti/Dynamic-Display-of-Histogram-and-Tone-curve GitHub Wiki

In this chapter, we discuss the results and compare the results to the desired output.

5.1 Result Analysis

We were able to obtain the Bayer Histogram and Tone Curve on preview for a given frame during live capture from the camera sensor after executing our Sample Code.Also we were able to record the video to which the tone curve had been applied using the Gstreamer encoder and store the said video in .mp4 format. To give the user a better idea of the tonal distribution of the given frame, we recorded the window of the frame with the previews present using ffmpeg. We also verified the tone curves for various indoor and out doors conditions.The frames collected after the tone curve was applied to them are shown in the following pictures, along with the frame’s associated Bayer histogram for indoor lighting conditions:

(a) Linear Tone Curve (b) Non-Linear Tone Curve
image image

We have presented the frames to which the linear tone curve and non linear tone curve have been applied. As observed the contrast increases and the brightness decreases.

(c) Reinhard Tone Curve (d) Exponential increase Tone Curve
image image
(e) S-shaped Tone Curve (f) Delayed S-shaped Tone Curve
image image

The exponential tone curve and s-shaped tone curve increases the contrast and decreases the brightness.

We may utilise the Reinhard tone curve, as seen from the results in Figure 5.2b, to better comprehend the colour distribution and intensity of the image.

The photographer can employ all of these tone curves to improve performance by adjusting the exposure, contrast, and brightness of the frame.

The frames collected after the tone curve was applied to them are shown in the following pictures, along with the frame’s associated Bayer histogram for outdoor lighting conditions:

(a) No Tone Curve (b) Linear Tone Curve
image image
(c) Non-Linear Tone Curve (d) Reinhard Tone Curve
image image
(e) Exponential increase Tone Curve (f) S-shaped Tone Curve
image image

Table 5.1: Comparison of tone curves

Tone curve Contrast Brightness CPU utilization
Linear Increases Decreases 3%-12%
Non-linear Increases Decreases 3%-15%
Reinhard No visible change Increases 3%-11%
Exponential increase Increases Decreases 3%-12%
S-shaped Increases Decreases 3%-12%
Delayed S-shape Increases Decreases 3%-13%

From Table 5.1, we can infer how to apply the tone curves in various situations. The Reinhard tone curve has been regarded as the most effective among all since it provides the best results in both indoor and outdoor settings. We can better comprehend the distribution of color and light intensities in an image by using the Reinhard tone curve.