‧͙⁺˚*･༓☾🐺☽༓･*˚⁺‧͙

Image Compression

3 min read

Parents : Digital Image Processing

Date and time note was created$= dv.current().file.ctime
Date and time note was modified$= dv.current().file.mtime

Image Compression is the process of removal of the redundant data from the image.

There are several types of image compression methods used to reduce the size of image files. Here are some common types of image compression methods:

  1. Lossless Compression: Lossless compression methods aim to reduce the file size of an image without any loss of information. The original image can be perfectly reconstructed from the compressed version. Examples of lossless compression methods include:

    • Run-Length Encoding (RLE)
    • Huffman Coding
    • Arithmetic Coding
    • Lempel-Ziv-Welch (LZW) Compression
    • Burrows-Wheeler Transform (BWT) Compression

  2. Lossy Compression: Lossy compression methods achieve higher compression ratios by sacrificing some amount of image quality. While the reconstructed image may not be identical to the original, it aims to maintain a visually acceptable level of quality. Lossy compression methods are commonly used in applications where storage or bandwidth constraints are critical. Examples of lossy compression methods include:

    • Discrete Cosine Transform (DCT) used in JPEG compression
    • Fractal Compression
    • Transform coding (e.g., Wavelet Transform)
    • Vector Quantization (VQ)
    • Neural Network-based compression techniques (e.g., Convolutional Neural Networks)

  3. Hybrid Compression: Hybrid compression methods combine elements of both lossless and lossy compression to achieve a balance between compression efficiency and image quality. These methods often involve using lossy compression for certain components of the image data and lossless compression for others. One example of a hybrid compression method is the JPEG2000 standard, which incorporates both wavelet-based lossy compression and entropy coding techniques.

  4. Spatial Compression: Spatial compression methods exploit redundancies present within individual pixels or local neighborhoods in the image. These methods include techniques such as predictive coding, delta encoding, and spatial domain transformations.

  5. Transform Compression: Transform compression methods apply mathematical transforms to convert the image data from the spatial domain to the frequency domain. This allows for more efficient representation of image data by concentrating most of the energy in fewer coefficients. DCT-based compression (used in JPEG) and wavelet-based compression are common examples of transform compression methods.

These are just a few examples of image compression methods, and there are variations and combinations of these techniques used in various image compression algorithms and standards. The choice of compression method depends on factors such as the desired compression ratio, acceptable image quality, and specific application requirements.

Related

References

Graph View

Backlinks