DSP accelerator for the wavelet compression/decompression of high- resolution images
A Texas Instruments (TI) TMS320C30-based S-Bus digital signal processing (DSP) module was used to accelerate a wavelet-based compression and decompression algorithm applied to high-resolution fingerprint images. The law enforcement community, together with the National Institute of Standards and Technology (NISI), is adopting a standard based on the wavelet transform for the compression, transmission, and decompression of scanned fingerprint images. A two-dimensional wavelet transform of the input image is computed. Then spatial/frequency regions are automatically analyzed for information content and quantized for subsequent Huffman encoding. Compression ratios range from 10:1 to 30:1 while maintaining the level of image quality necessary for identification. Several prototype systems were developed using SUN SPARCstation 2 with a 1280 {times} 1024 8-bit display, 64-Mbyte random access memory (RAM), Tiber distributed data interface (FDDI), and Spirit-30 S-Bus DSP-accelerators from Sonitech. The final implementation of the DSP-accelerated algorithm performed the compression or decompression operation in 3.5 s per print. Further increases in system throughput were obtained by adding several DSP accelerators operating in parallel.
- Research Organization:
- Oak Ridge National Lab., TN (United States)
- Sponsoring Organization:
- Department of Justice, Washington, DC (United States); USDOE, Washington, DC (United States)
- DOE Contract Number:
- AC05-84OR21400
- OSTI ID:
- 10191975
- Report Number(s):
- CONF-9309245-1; ON: DE94000337
- Resource Relation:
- Conference: ICSPAT `93: International conference on signal processing applications and technology,Santa Clara, CA (United States),28 Sep - 1 Oct 1993; Other Information: PBD: 23 Jul 1993
- Country of Publication:
- United States
- Language:
- English
Similar Records
The FBI wavelet/scalar quantization standard for gray-scale fingerprint image compression
Compression of fingerprint data using the wavelet vector quantization image compression algorithm. 1992 progress report