✦ScienexySCIENCE KNOWLEDGE LIBRARY
Science reference · SCI-CS-003-132

Digital Image Representation

Digital Image Representation describes how visual information is stored as numerical data using pixels, color channels, and spatial dimensions.

N = W × H
S = W × H × C × B / 8

Explanation

Digital Image Representation explains how computers convert visual scenes into numerical structures. A digital image is commonly represented as a two-dimensional grid of picture elements called pixels.

Each pixel contains one or more numerical values describing the visual information at a particular position. In a grayscale image, a pixel commonly stores an intensity value. In a color image, multiple channels are used to represent color information.

Important concepts include:

1. Pixel: The smallest addressable element of a digital raster image.
2. Resolution: The number of pixels along the width and height of an image.
3. Image dimensions: The width and height of the pixel grid.
4. Bit depth: The number of bits used to represent each pixel or color channel.
5. Color channels: Separate numerical components used to represent color.
6. Color space: A mathematical model used to represent colors, such as RGB, HSV, or grayscale.

A grayscale image may use 8 bits per pixel, allowing 256 possible intensity levels from 0 to 255. A common RGB image uses three 8-bit channels: red, green, and blue. Each channel can represent 256 intensity levels, producing approximately 16.7 million possible RGB combinations.

An image can be represented mathematically as a function of spatial coordinates. For a grayscale image, the function returns an intensity value at each pixel location. For an RGB image, it returns three channel values.

Digital images can also be stored in different file formats. PNG commonly supports lossless compression, JPEG commonly uses lossy compression, and TIFF is frequently used in professional imaging workflows. The selected format affects file size, compression, quality, and supported features.

Understanding image representation is essential because image dimensions, color spaces, bit depth, and storage formats influence how Computer Vision algorithms receive and process visual data.

Variables

SymbolMeaningUnit

N

Total number of pixels

pixels

W

Image width

pixels

H

Image height

pixels

S

Uncompressed image data size

bytes

C

Number of color channels

B

Bits per channel

Example or application

An RGB image with dimensions 100 × 100 pixels and 8 bits per channel contains 10,000 pixels and requires approximately 30,000 bytes of uncompressed pixel data, excluding metadata and storage overhead.

Text

A digital image is represented as a grid of pixels. Each pixel stores numerical information about brightness or color, allowing computers to process and display visual information.

Key Points

  • Digital images are commonly represented as pixel grids.

  • Pixels store numerical intensity or color information.

  • Resolution describes the spatial dimensions of an image.

  • Bit depth determines the number of representable intensity levels.

  • RGB uses red, green, and blue color channels.

  • Color spaces provide different ways to represent visual information.

  • Image formats use different compression and storage methods.

Applications

  • Digital photography

  • Medical imaging

  • Computer Vision datasets

  • Satellite imaging

  • Image editing software

  • Video processing

  • Scientific visualization

Learning Outcomes

  • Explain how digital images are represented.

  • Identify pixels, resolution, and bit depth.

  • Describe the RGB color model.

  • Calculate the approximate uncompressed size of an image.

  • Differentiate common image formats.

Practice Questions

  • What is a pixel?

  • How does image resolution differ from bit depth?

  • How many intensity levels can an 8-bit grayscale image represent?

  • How much uncompressed data does a 200 × 100 RGB image require at 8 bits per channel?

Related Concepts

  • Image Processing

  • Color Spaces

  • Image Resolution

  • Pixel Intensity

  • Image Compression

  • Computer Vision

Common Mistakes

  • Assuming higher resolution always means better visual quality.

  • Confusing bits per channel with total bits per pixel.

  • Ignoring color channels when estimating image storage size.

  • Assuming compressed file size equals uncompressed pixel data size.

Sources and verification

Sources are shown as provided in the project data. Not every source has yet been independently verified. Please report errors or missing citations.

  • Gonzalez, R. C., and Woods, R. E., Digital Image Processing, 4th Edition, Pearson, 2018.
  • Szeliski, R., Computer Vision: Algorithms and Applications, 2nd Edition, Springer, 2022.

Suggest a correction or source · More in Artificial Intelligence & Machine Learning