Why Use Lossy Compression? Algorithms, RLE, Huffman, and Compression Bombs

1 Lossy Compression

Lossy compression exploits the human eye and ear's insensitivity to certain frequency components, allowing some information loss to achieve much higher compression ratios; it is widely used for audio, image, and video data.

After lossy compression, some data is permanently lost and cannot be fully restored, but the trade‑off is acceptable because the lost information rarely affects usability, providing a better cost‑performance ratio.

Common examples include everyday emoji images, where the loss of detail yields smaller, more shareable files.

2 Lossless Compression

Lossless compression removes statistical redundancy, allowing the original data to be perfectly reconstructed; typical compression ratios range from 2:1 to 5:1 and are used for text, program binaries, and specialized images such as fingerprints or medical scans.

Modern lossless algorithms can shrink files to 30‑40% of their original size, though higher compression often means slower decompression.

Examples of lossless techniques include Run‑Length Encoding (RLE), dictionary algorithms, and Huffman coding.

RLE compresses repeated characters by storing the count followed by the character, e.g. bbbbbb y y ttttt e ddd aaa n n ccccccc eee becomes 6b3y5t1e3d3a2n7c4e. Its drawback is that data without long runs can expand.

Dictionary algorithms replace frequently occurring words with short codes, similar to assigning nicknames.

Huffman coding, invented by David Huffman in 1952, assigns shorter bit patterns to more frequent symbols. For the sequence 1,50,20,50,50,18,50,25,32,18, the frequencies produce the following codes:

50: 00

18: 01

1: 100

20: 101

25: 110

32: 111

Applying these codes transforms the original data into 100,00,101,00,00,01,00,110,111,01, which is far more compact than the original binary representation.

3 Compression Bombs

A compression bomb is a tiny archive (often only tens of kilobytes) that expands to an enormous size—potentially petabytes—when decompressed, overwhelming system resources.

Examples include a 42 KB file that expands to 4.5 PB after repeated nested archives, and a 28 KB file that recursively extracts copies of itself, effectively creating an infinite loop.

These bombs exploit the fact that compression removes redundancy; however, because the data is deliberately repetitive, the information entropy is low, allowing extreme size reduction.

Historically, compression bombs have been used to evade antivirus scanning by forcing the scanner to decompress massive amounts of data, though modern security software can detect and mitigate such attacks.