Image enhancement

Image enhancement is the process of processing an image so that the result is more suitable than the original for a specific application, usually viewing by humans or further analysis. It does not add new information; it makes existing detail easier to see.

Main aims

• Improve contrast and brightness of dark, washed-out or low-contrast images.
• Reduce noise (random specks from the sensor or transmission).
• Sharpen edges and fine detail that look blurred.
• Highlight features of interest, such as tumours in an X-ray or roads in a satellite image.

Two approaches

1. Spatial domain methods work directly on pixels. Examples: contrast stretching, histogram equalisation, negative of an image, smoothing with an averaging or median filter, sharpening with a Laplacian filter.
2. Frequency domain methods transform the image (for example with the Fourier transform), modify the frequency components, and transform back. A low-pass filter smooths and removes noise; a high-pass filter sharpens edges.

Example: a photo taken in dim light has all its pixel values between 30 and 90. Contrast stretching maps this range to 0-255, so the picture becomes clearly visible.

Enhancement differs from image restoration. Restoration tries to undo a known degradation (such as a known blur), while enhancement is subjective and aims only at a better-looking or more useful picture.