JPEG compression is commonly described as lossy. Which statement correctly reflects this property?

• A. It is lossless
• B. It is lossy
• C. It uses lossless for color channels but lossy for grayscale
• D. It does not compress at all

Answer: (B)

JPEG compression is lossy because it deliberately discards some image information during the encoding process. After converting the image to a suitable color space (often YCbCr) and breaking it into small blocks, the data is transformed into frequency components using the Discrete Cosine Transform. The key step is quantization, where these frequency coefficients are reduced to fewer levels. This rounding and reduction mean some original details are permanently lost, so the image cannot be perfectly reconstructed from the compressed data. Chroma subsampling (reducing color resolution) further contributes to lossiness by trimming color detail, which is particularly noticeable in areas with fine color variation.

Because data is irreversibly discarded, JPEG compression never fully preserves the exact original image, which is why the statement is that JPEG is lossy. Even grayscale JPEGs follow the same principle: they still rely on quantization and block-based processing, leading to information loss. This is in contrast to lossless formats (like PNG) where no original data is discarded during compression.