Image MIME Reference
DjVu
DjVu Document
image/vnd.djvu
VS
PCO B16 Image
image/vnd.pco.b16
PCO-B16
A complete technical comparison of DjVu Document and PCO B16 Image — covering compression, feature support, browser compatibility, magic bytes, and when to choose each format.
Equal browser support
Feature Support
| Feature | DjVu | PCO-B16 |
|---|---|---|
| Transparency (Alpha) | Yes | ✕ No |
| Animation Support | ✕ No | ✕ No |
| Progressive Loading | Yes | ✕ No |
| HDR Support | ✕ No | ✕ No |
| EXIF Metadata | ✕ No | ✕ No |
| ICC Color Profile | ✕ No | ✕ No |
DjVu Document
image/vnd.djvu
Extension
.djvu.djv
Container
Interchange File Format (IFF)
Compression
JB2 (Bi-tonal Text) / IW44 (Wavelet Background) / BZZ (General Data/Text)
Algorithm
JB2IW44BZZ
Color Depth
1-bit (Foreground Text), 24-bit (Background)
Developed by
AT&T Labs (Yann LeCun, Léon Bottou, Patrick Haffner, Paul G. Howard)
Released
1996
Magic Bytes
41 54 26 54 46 4F 52 4D
PCO B16 Image
image/vnd.pco.b16
Extension
.b16
Container
PCO B16 Binary Container
Compression
Uncompressed (Raw Sensor Data)
Algorithm
None
Color Depth
16-bit (Monochrome or Bayer Raw Color)
Developed by
PCO AG (Excelitas PCO GmbH)
Released
Early 2000s
Magic Bytes
50 43 4F 2D
Browser Support Comparison
| Browser | DjVu | PCO-B16 |
|---|---|---|
|
|
✕ | ✕ |
|
|
✕ | ✕ |
|
|
✕ | ✕ |
|
|
✕ | ✕ |
|
|
✕ | ✕ |
DjVu
DjVu Strengths
- Unmatched compression ratios for high-resolution scanned documents (often 5 to 10 times smaller than equivalent JPEG/TIFF-based PDFs)
- Separates text layers cleanly from backgrounds, allowing for highly legible text even at low file sizes
- Embedded OCR (Optical Character Recognition) text layers allow for fast searching
Limitations
- Lacks native support in modern operating systems and web browsers, requiring third-party software
- Largely superseded by PDF/A and modern PDF compression standards in standard corporate workflows
- Not suitable for vector graphics or born-digital documents (like Word exports), where standard PDF is far superior
PCO-B16
PCO-B16 Strengths
- Preserves exact, uncompressed 16-bit raw sensor data, which is critically important for accurate scientific measurements, fluorescence microscopy, and photometry
- The simple, linear struct header is highly predictable and easy to parse in custom Python, MATLAB, or C++ scripts for researchers building bespoke pipelines
Limitations
- Generates massive file sizes due to the complete lack of image compression [1.3.1]
- Completely unsupported by standard operating systems, web browsers, and consumer image editing suites
- Strictly tied to the hardware and software ecosystems of a single camera manufacturer
When to choose which format
DjVu
Use DjVu when…
- Scanned Books and Manuals
- Digital Libraries (e.g., Internet Archive)
- Historical Document Archiving
PCO-B16
Use PCO-B16 when…
- Microscopy and Life Sciences
- High-Speed Camera Recording
- Machine Vision and Industrial Inspection
- Scientific Photometry