Motion-adaptive weighted averaging for temporal filtering of noisy image sequences

1992 ◽  
Author(s):  
Mehmet K. Ozkan ◽  
M. Ibrahim Sezan ◽  
A. Murat Tekalp
Keyword(s):  
Image Sequences ◽  
Weighted Averaging ◽  
Noisy Image ◽  
Temporal Filtering

Evaluation of Computer-assisted Image Enhancement in Minimal Invasive Endoscopic Surgery

2004 ◽  
Vol 43 (04) ◽  
pp. 362-366 ◽  
Author(s):  
F. Vogt ◽  
W. Hohenberger ◽  
D. Paulus ◽  
H. Niemann ◽  
C. H. Schick ◽  
...  
Keyword(s):  
Image Processing ◽  
Minimal Invasive ◽  
Image Sequences ◽  
Computer Assisted ◽  
Processing Methods ◽  
Temporal Filtering ◽  
Endoscopic Images ◽  
Filter Size ◽  
Image Pairs

Summary Objectives: This paper focusses on the evaluation of the usage of computer-aided image processing methods for minimal invasive surgery. During video endoscopy of visceral cavities the images are displayed directly on the monitor without further processing. In the course of the operation the former good quality of the images decreases due to typical disturbances like bleeding, smoke or flying particles. These disturbances can be reduced by using image processing methods like color normalization, temporal filtering or equalization. Methods: In this double-blinded analysis, 14 surgeons with different levels of experience evaluated 120 image pairs and 5 image sequences, directly comparing original and processed images or movies. Results: Color normalization and equalization proved to significantly enhance video endoscopic images. With regard to temporal filtering, an improvement could be seen in the image sequences with filter size 5 being a greater enhancement than filter size 3. Comparing the state of experience and its influence on the results, it occurred that the experienced surgeons preferred the original color while altogether agreeing that the color-normalized images were better. Conclusions: The results obtained in the present evaluation show that the image processing methods which were used can significantly improve the quality of video endoscopic images. As a result of this, necessary lavages of the operated area are reduced and a better overview and orientation for the surgeon can be reached.

scholarly journals Random projections and Hotelling’s T2 statistics for change detection in high-dimensional data streams

2013 ◽  
Vol 23 (2) ◽  
pp. 447-461 ◽  
Author(s):  
Ewa Skubalska-Rafajłowicz
Keyword(s):  
Discrete Time ◽  
Data Streams ◽  
Random Projection ◽  
Image Sequences ◽  
High Dimensional ◽  
Random Projections ◽  
Random Subspace ◽  
Noisy Image ◽  
Method Of Dimensionality Reduction ◽  
Diagnostic Properties

The method of change (or anomaly) detection in high-dimensional discrete-time processes using a multivariate Hotelling chart is presented. We use normal random projections as a method of dimensionality reduction. We indicate diagnostic properties of the Hotelling control chart applied to data projected onto a random subspace of Rn. We examine the random projection method using artificial noisy image sequences as examples.

Motion-compensated filtering of noisy image sequences

1994 ◽  
pp. 75-86
Author(s):  
Richard P. Kleihorst ◽  
Jan Biemond ◽  
Reginald L. Lagendijk
Keyword(s):  
Image Sequences ◽  
Noisy Image

HOS-based method of global motion estimation for noisy image sequences

1999 ◽  
Vol 35 (16) ◽  
pp. 1320 ◽  
Author(s):  
E. Ibn-elhaj ◽  
D. Aboutajdine ◽  
S. Pateux ◽  
L. Morin
Keyword(s):  
Motion Estimation ◽  
Image Sequences ◽  
Noisy Image ◽  
Global Motion ◽  
Global Motion Estimation
Sign in / Sign up

Export Citation Format

Share Document