JPEG

JPEG. It is a commonly used method of lossy compression for digital images, particularly for those produced by digital photography. The degree of compression can be adjusted, allowing a selectable tradeoff between storage size and image quality. The format has become the de facto standard for storing and transmitting photographic images on the World Wide Web.

Overview

The format was created by the Joint Photographic Experts Group, from which it derives its name, and published as an International Organization for Standardization standard in 1992. It is designed to exploit the limitations of the human visual system, achieving high compression ratios by discarding information less noticeable to the eye. This makes it exceptionally efficient for continuous-tone images like photographs, but less suitable for sharp edges and text, where formats like Portable Network Graphics or Graphics Interchange Format are preferred. Its ubiquity is supported by near-universal compatibility across software, operating systems like Microsoft Windows and macOS, and hardware devices including digital cameras and smartphones.

Technical details

The core of the compression is based on the discrete cosine transform, which converts image data into a sum of cosine functions oscillating at different frequencies. The process typically operates on 8x8 blocks of pixel components in the YCbCr color space, which separates luminance from chrominance information. Key parameters controlling quality include the quantization tables, which determine how much high-frequency information is discarded, and the subsampling ratio, often employing chroma subsampling like 4:2:0 to reduce color resolution. The final stage uses entropy coding techniques, such as Huffman coding or arithmetic coding, to further compress the data stream.

File format and structure

A JPEG file is more precisely referred to as JPEG Interchange Format or JPEG File Interchange Format, which defines a wrapper for the encoded image data. The file structure consists of a series of segments, each beginning with a marker. Key segments include the Start of Image, which contains quantization tables and Huffman codes, and the Start of Frame, which holds image dimensions and sampling factors. Other common segments can contain Exif metadata from camera manufacturers like Canon or Nikon, ICC profiles for color management, or thumbnail images. While the JFIF specification is the most common implementation, other container formats like SPIFF also exist.

Compression process

The encoding pipeline begins by converting the image from an RGB color model to YCbCr, where the chroma channels are often downsampled. Each channel is divided into 8x8 blocks, and the discrete cosine transform is applied to each block, concentrating visual information into the upper-left low-frequency coefficients. These coefficients are then quantized by dividing them by values in a quantization table, a lossy step that reduces precision, especially for higher frequencies. The resulting matrix is reordered using a zigzag scan and compressed losslessly via run-length encoding followed by Huffman coding. Decoding reverses this process, though the quantized data cannot be fully recovered, leading to artifacts like ringing or blocking.

Applications and usage

The format is the dominant standard for storing and sharing photographic content on the internet, used extensively by platforms like Facebook, Instagram, and Flickr. It is the default output for most consumer digital cameras and is widely supported in image editing software such as Adobe Photoshop and GIMP. Its efficiency makes it suitable for applications where file size is critical, including web pages, email attachments, and digital photo albums. However, for professional photography requiring multiple edits, TIFF or raw image formats from companies like Sony or Leica are often used to avoid generational quality loss from repeated compression.

Development and standardization

The development committee, the Joint Photographic Experts Group, was formed in 1986 as a collaboration between ITU-T and ISO/IEC. The first standard, JPEG (1992), was published as ITU-T Recommendation T.81 and ISO/IEC 10918-1. Subsequent extensions added features like progressive JPEG for web display, lossless JPEG compression, and support for arithmetic coding. Later, more advanced codecs were developed by the same group, including JPEG 2000 which uses wavelet transform, JPEG XR based on technology from Microsoft, and the modern JPEG XL format. The format's maintenance and future development continue under the auspices of the ISO/IEC JTC 1/SC 29 subcommittee.