2D Correlation - OriginPro
Back to Top
OriginLab Corporation - Data Analysis and Graphing Software - 2D graphs, 3D graphs, Contour Plots, Statistical Charts, Data Exploration, Statistics, Curve Fitting, Signal Processing, and Peak Analysis     

2D Correlation

2D Correlation can be used to detect the correlation between a pair of 2D signals. It is widely used in spectroscopy and pattern recognition.

2D Correlation tool in OriginPro...

The 2D Correlation tool in OriginPro performs 2D correlation on two input matrices. Both linear and circular correlation can be computed. Two computation methods are available: a fast algorithm based on FFT and an accurate method based on shift accumulation. In addition, you can choose to normalize the result. With a normalized result, it is easier to tell the degree to which the two input signals are correlated.

2D Correlation Dialog

In the following example, we use the 2D Correlation tool to detect whether a pattern is contained in a 2D signal. The image views of the 2D signal and the pattern are as follows:

2D Signal Pattern

2D Signal to detect the presence of the pattern


These two matrices are assigned as input matrices in the 2D Correlation tool. We choose to perform normalization to the result by selecting the Normalize Result check box in the 2D Correlation tool. After the computation, we can get a result matrix. The image view of it is as follows:

Result of 2D Correlation Example

We can see three white spots in the result (white spot means a value closed to 1 in this case). This indicates that the pattern is likely to present in the 2D signal. The positions of the white spots also suggest the possible positions of the presented patterns.

OriginPro includes the ability to automatically recalculate the analysis results of the 2D Correlation operation any time you change the parameters or update your source data. In addition, the settings for this analysis routine can be saved to an analysis theme for use later with similar data.

© OriginLab Corporation. All rights reserved.