View Scatter Plots

来源: https://vis-webcontent.s3.amazonaws.com/quickguides/Explore+Raster+Data/View+Scatter+Plots/index.html#/

Lesson content

Lesson 1 of 1

View Scatter Plots

A scatter plot compares pixel values of two bands in an image. It provides a visual picture of the statistical correlation between two bands. One band provides the X coordinates, and the other band provides the Y coordinates. Scatter plots can be used to find and label groups of pixels based on their two-dimensional (2D) distribution.

Open and Display an Image

1

Select File > Open from the Menu bar. An Open dialog appears. 2. 2

Go to the "data" directory in your ENVI installation path.

Windows: C:\Program Files\NV5\ENVIxx\data (xx is the version number)
Linux: /user/local/NV5/envixx/data
Mac: /Applications/NV5/envixx/data 3. 3

Select the file qb_boulder_msiand click Open. This is a Quickbird multispectral image of Boulder, Colorado. The picture below shows the image at its full extent.

Display a Scatterplot

1

Select Display > 2D Scatter Plot from the Menu bar. The Scatter Plot Tool appears, and a "Scatter Plot" layer is added to the Layer Manager. By default, Band 3 (red) is along the X-axis and Band 2 (green) is along the Y-axis. The black cloud of data points represents the pixels that are displayed in the Image window.

2

Rotate the scroll wheel on your mouse to zoom in and out of the plot. Resize the Scatter Plot Tool if needed. Click and hold the middle mouse button to pan around the plot. 2. 3

Compare the near-infrared band with the red band. To do this, drag the Y-axis slider to Band 4 and keep the X-axis slider at Band 3.

A group of pixels in the image can share the same value in both bands. The black scatter plot does not show where the highest concentration of related pixels occur. You can apply a color gradient (also called a density slice) to show the densities of related pixels.

4

Click the Toggle Density Slice button in the Scatter Plot Tool. The scatter plot is displayed with a blue color gradient. Bright areas represent a higher density of pixels that are similar in value.

Interactively View Pixel Distribution

1

When the Scatter Plot Tool was displayed, the cursor became a small square. Click and drag across some pixels in the image to show where these pixels reside in the Scatter Plot Tool. This dynamic linking is known as dancing pixels.

Example of dancing pixels. Clicking and dragging the cursor over a dry field highlights the corresponding data points in the scatter plot.

2

You can also select different parts of the scatter plot to view their corresponding pixels in the image. Click and hold down the left mouse button to draw a polygon around some data points in the scatter plot. The image pixels associated with those data points are highlighted in the Image window.

Example of drawing a polygon around an area with high correlation between the red and NIR bands. This part of the scatter plot represents bright image features such the large school shown on the right.

3

To clear the highlighted pixels, right-click in the scatter plot and select Clear All. 2. 4

Now that you know how to use the Scatter Plot Tool, take some time to view scatter plots of other band combinations and deduce how they relate to image features. For example, look at a scatter plot of the red and green bands. Are they highly correlated (close to a straight line along the diagonal)? If so, for what specific features?

This quick guide demonstrated some of the core capabilities of the Scatter Plot Tool. It has other features such as creating unique classes for different groups of data points, and interfacing with the Region of Interest (ROI) Tool. You will learn about those features in other quick guides.

Your input is important to us, please take a few moments to fill out ourQuick Guide Feedback(opens in a new tab)form.