TIFFからJPEGへのコンバーター

Tagged Image File Format

Also known as TIF. Requires tiff-v3.6.1.tar.gz or later. Use -define to specify the rows per strip (e.g. -define tiff:rows-per-strip=8). To define the tile geometry, use for example, -define tiff:tile-geometry=128x128. To specify a signed format, use -define quantum:format=signed. To specify a single-precision floating-point format, use -define quantum:format=floating-point. Set the depth to 64 for a double-precision floating-point format. Use -define quantum:polarity=min-is-black or -define quantum:polarity=min-is-white toggle the photometric interpretation for a bilevel image. Specify the extra samples as associated or unassociated alpha with, for example, -define tiff:alpha=unassociated. Set the fill order with -define tiff:fill-order=msb|lsb. Set the TIFF endianness with -define tiff:endian=msb|lsb. Use -define tiff:exif-properties=false to skip reading the EXIF properties. You can set a number of TIFF software attributes including document name, host computer, artist, timestamp, make, model, software, and copyright. For example, -set tiff:software "My Company". If you want to ignore certain TIFF tags, use this option: -define tiff:ignore-tags=comma-separated-list-of-tag-IDs. Since version 6.9.1-4 there is support for reading photoshop layers in TIFF files, this can be disabled with -define tiff:ignore-layers=true. To preserve compression of the source image, use: -define tiff:preserve-compression=true.

Joint Photographic Experts Group JFIF format

Note, JPEG is a lossy compression. In addition, you cannot create black and white images with JPEG nor can you save transparency. Requires jpegsrc.v8c.tar.gz. You can set quality scaling for luminance and chrominance separately (e.g. -quality 90,70). You can optionally define the DCT method, for example to specify the float method, use -define jpeg:dct-method=float. By default we compute optimal Huffman coding tables. Specify -define jpeg:optimize-coding=false to use the default Huffman tables. Two other options include -define jpeg:block-smoothing and -define jpeg:fancy-upsampling. Set the sampling factor with -define jpeg:sampling-factor. You can size the image with jpeg:size, for example -define jpeg:size=128x128. To restrict the maximum file size, use jpeg:extent, for example -define jpeg:extent=400KB. To define one or more custom quantization tables, use -define jpeg:q-table=filename. These values are multiplied by -quality argument divided by 100.0. To avoid reading a particular associated image profile, use -define profile:skip=name (e.g. profile:skip=ICC).