|
Efficient estimation of ideal-observer performance in classification tasks involving high-dimensional complex backgroundsPublished in JOSA A, Vol. 26 Issue 11, pp.B59-B71 (2009) |
Spotlight summary: The use of the Bayesian ideal observer approach to image quality assessment can be highly advantageous but has been hampered by computational complexity for real-life statistics. The current procedure develops a fast Markov-chain model for real-time estimation of channelized ideal observer computations.
| OCIS Codes: | (110.3000) Imaging systems : Image quality assessment |
| (330.1880) Vision, color, and visual optics : Detection | |
| (330.6100) Vision, color, and visual optics : Spatial discrimination |
Posted on October 08, 2009
Add Comment
You must log in to add comments.
Previous Spotlights
Related Journal Articles 
- Visual signal detection in structured backgrounds. III. Calculation of figures of merit for model observers in statistically nonstationary backgrounds (JOSAA)
- The Rose model, revisited (JOSAA)
- Role of phase information and eye pursuit in the detection of moving objects in noise (JOSAA)
- Detection improvement in spatially filtered x-ray fluoroscopy image sequences (JOSAA)
- Channelized-ideal observer using Laguerre-Gauss channels in detection tasks involving non-Gaussian distributed lumpy backgrounds and a Gaussian signal (JOSAA)
Related Conference Papers
- Joint Design of Lens Systems and Digital Image Processing
- Impact of the 2D-Structured Noise in the Post-Processing of Hybrid Optical-Digital Imaging Systems
- Information capacity of seeing limited imaging systems
- Information Capacity of Seeing Limited Imaging Systems
- Information-theoretic treatment of axial imaging
