Uji Efektifitas Kompresi Golomb-Rice dan Huffman untuk Metadata EXIF dalam File JPEG

Compression algorithms are now called modern compression algorithms. This improvement is characterized by the combination of various classical techniques and is even based on machine learning and AI. However, the important part of compression is not only the algorithm, but also knowledge of the internal structure and metadata of the file is required. Like JPEG has a file structure that can be changed, cannot be changed, every marker (header), and EXIF metadata. Lack of knowledge of the file structure can cause data damage and file corruption. This study evaluates the compression of EXIF metadata of JPEG files using the Golomb-Rice and Huffman algorithms. Golomb-Rice can produce compression that affects the k parameter, while Huffman is optimal based on symbol frequency, but requires a code table. This study measures the effectiveness of both algorithms based on the compression ratio (CR). The test results of Golomb-Rice are more effective than those of Huffman. So, it can be concluded that the Golomb-Rice algorithm is superior in the context of compressing EXIF JPEG metadata, while Huffman shows lower efficiency in the tested scenarios.