Compression (data, mechanical, audio, and medical)

Data and image compression

Data compression encodes information using fewer bits than the original representation. Two broad approaches exist: lossless methods, which allow exact reconstruction, and lossy methods, which discard some detail to achieve higher reduction. Lossless techniques are common for text, executables and some images; lossy is widespread for photographs, audio and video where perfect fidelity is unnecessary.

• Common lossless tools and ideas: Huffman coding, run-length encoding, Lempel–Ziv variants, arithmetic coding, and entropy concepts.
• Common lossy approaches: transform coding (e.g., discrete cosine transform), quantization and psychoacoustic or perceptual models used in JPEG, MP3 and modern video codecs.

Signal and audio compression

In audio engineering, compression can mean reducing dynamic range so quieter and louder sounds sit closer together. Dynamic range compressors modify signal gain using threshold, ratio, attack and release controls, and are essential in mixing, broadcasting and live sound to improve intelligibility and perceived loudness.

Mechanical and material compression

In mechanics, compression refers to forces that press materials together and to the resulting stresses and deformations. Compressive strength measures how much load a material can bear before crushing. In engines, the compression ratio compares cylinder volume before and after piston movement and influences power and efficiency.

Medical and textile compression

Compression garments—stockings, sleeves, bandages—apply controlled pressure to limbs to improve circulation, reduce swelling and prevent venous problems. Compression therapy is a standard treatment for venous insufficiency, lymphedema and some sports recovery uses.

Trade-offs, history and importance

Compression involves trade-offs: smaller size versus computational cost, latency, quality loss and error sensitivity. The mathematical foundations trace to information theory and entropy, and practical algorithms evolved through mid-20th-century advances and later standardization of widely used formats. Compression remains central to storage, communications, multimedia delivery and many engineering applications.