Related Experiment Video
Updated: Feb 11, 2026

Image-based Lagrangian Particle Tracking in Bed-load Experiments
Published on: July 20, 2017
A New Distance Measure Based on Generalized Image Normalized Cross-Correlation for Robust Video Tracking and Image
Arie Nakhmani1, Allen Tannenbaum
1A. Nakhmani is with the Department of Electrical and Computer Engineering, Boston University, Boston, MA.
We introduce two new normalized cross-correlation-based distance measures for image matching. These novel measures offer improved performance for specific computer vision tasks like image recognition and object tracking.
Area of Science:
- Computer Vision
- Image Processing
- Pattern Recognition
Background:
- Image matching is crucial for various computer vision tasks.
- Existing distance measures have limitations depending on the application.
- Natural images exhibit high spatial correlation between pixels.
Purpose of the Study:
- To propose two novel, normalized distance measures for image matching.
- To evaluate the performance of these measures against established methods.
- To determine the suitability of the proposed measures for different computer vision applications.
Main Methods:
- Development of two novel distance measures based on normalized cross-correlation.
- Normalization of measures to a range between 0 and 1.
- Comparative analysis with Normalized Cross-Correlation (NCC) and Image Euclidean Distance (IMED).
Main Results:
- The proposed measures leverage pixel spatial correlation for enhanced accuracy.
- One measure demonstrates superior performance in tracking applications.
- The other measure is more appropriate for image recognition tasks.
Conclusions:
- The novel distance measures offer distinct advantages for specific image matching applications.
- The choice of distance measure should be tailored to the requirements of the computer vision task.
- These findings contribute to the development of more robust image matching algorithms.
Related Concept Videos
Distance Problem
Electronic Distance Measuring Instruments
Design Example: Measuring Distance Between Two Points with Obstructions
Distance Measurements by Taping
Correlations
Short-distance Transport of Resources

