Analyze Tab
Analyze MAPIR camera images
Last updated
Analyze MAPIR camera images
Last updated
To start analyzing images press the "Open File" or "Open Folder" button and select an image or folder of images to analyze.
The image will open in the image view box:
To apply a raster equation, such as the NDVI index to the image, press the EDIT INDEX button to open the 'Raster Index Calculator" pop-up window:
Selecting an index from the "Index:" drop-down will display the index formula photo:
Notice the left-hand side shows the spectrum of light that is typically used in that index formula variable location. For the NDVI index it typically uses near infrared (NIR) light and red (RED) light. The right-hand side shows the variables X, Y, Z that correspond to the X, Y, Z drop-downs below the index formula photo.
MCC automatically labels the image channel/band names if a Survey3 image is open. In our example, this is the RGN filter model Survey3, so the bands are labeled accordingly. For all our filter models, the name of the filter corresponds with the band order, so filter R,G,N = bands 1,2,3.
For NDVI we need to choose the "X: Band 1: Red" and "Y: Band 3: NIR", as seen below:
Press the APPLY button to see the opened image update without closing the Raster window. Press the OK button to keep the changes and close the Raster window.
On the right side you can see the pixel map legend. The name at the top of the legend shows the index name that was used (NDVI), and the numbers represent the pixel value range (minimum to maximum) of the image. In our example the pixels range from -1 to 1. These are the minimum and maximum pixel end-points.
The grayscale gradient image next to the pixel scale shows the color associated with the corresponding pixel value. This is commonly called a LUT, or "Look Up Table". The minimum value is set to black color and the maximum is set to white pixel. A gradient map is then applied to the pixels in between, with a medium gray representing a value 50% (half-way) between the minimum and the maximum.
Looking at the NDVI index image you can see the surrounding grass is mostly composed of white pixels, since the NDVI index correlates a higher NDVI value with healthy vegetation that is reflecting a lot of near infrared (NIR) light.
Let's apply a new color LUT (Look Up Table) to the pixels so it is easier to see the contrast in the image. Contrast is the difference between the black and white pixels.
Press the EDIT LUT button to open the "Color Map (LUT)" pop-up window:
Click the APPLY button to apply the default color LUT settings to the opened image:
Understanding how to properly adjust the LUT minimum and maximum clipping points may take some time. It helps to know a little about the raster index being used and the index's typical pixel ranges for certain materials. We'll go over it now briefly to help you understand the various options.
The NDVI index has a maximum pixel range of -1 to 1, but most often it is clipped to 0 to 1. We will explain clipping in a moment below. The reason the NDVI range is commonly set from 0 to 1 is that the NDVI index is typically used to analyze vegetation "health", with higher NDVI values correlating to more photosynthesis activity going on (reflecting more NIR light). So a higher NDVI value typically represents a healthier plant. Vegetation typically ranges from dead or low health around 0.2-0.3 to most healthy around 0.8-0.9.
Notice in the previous image that the surrounding grass white pixels are now mostly colored medium green (~0.6) to dark green (~0.9), showing that the majority of the grass is "healthy". Since our LUT minimum and maximum clipping points are still set at the defaults, they are at the image's actual pixel minimum and maximum. We have set our contrast for the LUT to cover the entire pixel range (-1 to 1).
It is common to adjust the contrast of the LUT, meaning to adjust the values for the minimum and maximum clipping points. Making the contrast range less condenses the color ramp/gradient, meaning that the LUT contrast increases within the clipped pixel range.
Let's adjust the clipping range of the opened image. Open the "Color Map (LUT)" pop-up window (EDIT LUT button). Change the "Minimum:" value to 0, as shown below. Click Apply/Ok.
The opened image has now been updated:
Now you can better see the vegetation health in the image. The lower health grass has been colored light green (~0.6) and the healthiest grass is colored dark green (~0.9). Exposed soil in the foreground and perimeter is red (0.0) to orange (~0.3).
Click the left "<" and right ">" arrows to toggle through the other images in the input folder. The same index and LUT will be applied to the other images as well.
Opening up the "Color Map (LUT)" pop-up window (EDIT LUT) we can choose a "Classes:" option:
Classes define how many defined color ranges the pixels are scaled between. The more classes the higher the contrast, the less classes the lower the contrast between pixel regions. You've seen the highest "7 Colors" class option, here is the opened photo below in "3 Colors" for comparison:
Opening up the "Color Map (LUT)" pop-up window (EDIT LUT) we can choose a "Clip:" option:
When the minimum value is increased, or the maximum value is decreased, you are clipping/cutting off the pixels outside that inner clipped range. The pixels that are clipped/cut off can then have their value changed to something else. Those are the clip options explained below:
Pixels outside the clipping range will be set to the LUT default minimum and maximum pixel color. For the Red-Yellow-Green LUT, lower valued pixels are all set to red, and higher valued to green.
Pixels outside the clipping range will be set to transparent (see-through). This is useful to overlay the output image in other image applications, such as as a raster layer in a GIS application.
Pixels outside the clipping range will be set to the pixels from the raster index image. Basically setting the LUT back to the default grayscale for only those pixels.
Pixels outside the clipping range will be set to the pixels from the original image. This is a common option for us to use ourselves, as it tends to easily highlight the contrast we're looking to show.
Here is the opened image with the "Clip" option set to "Background Original":
And the original opened (calibrated) image for comparison:
After reflectance calibration the healthy vegetation is blue, because on this RGN filter model camera the near infrared (NIR) light is captured in the sensor's blue channel.
You could also choose the GNDVI index and process that with this RGN filter model MAPIR camera:
You could also open up an OCN filter model image of a similar scene and process for NDVI:
Above right of the image viewer you will find the navigation buttons (from left to right):
Toggle Left - Toggle Right - Zoom Out - Zoom In - Zoom to Fit
Click the left "<" and right ">" arrows to toggle through the other images in the input folder. The same index and LUT will be applied to the other images as well.
Pressing the + and - buttons will zoom in and out of the image.
Pressing the "[ ]" button will scale the image back to the default fit zoom and centered.
Clicking and holding the cursor on the image while dragging around will move the image around.
The SAVE button allows you to output the various formats of the analyzed image. Pressing the SAVE button icon you see the following options:
The BROWSE button selects the output folder to save the images to.
Selecting the "Save Index Image" option saves the processed raster index images.
Selecting the "Save LUT Image" option saves the processed LUT color images.
The drop-down allows you to choose whether you will save the open image or the entire input folder of images.
Once you have made your save option choices press the SAVE button to save the images. A progress bar will show to the right of the Cancel button while it is saving the images.
Our example image shows our attached to a short tripod on top of some grass. The Survey3 camera was handheld during image capture, from about 1 m distance. There are regions in the grass ranging from dirt, to lower health grass, to higher health grass. This 16-bit TIFF image was previously converted from RAW and calibrated for reflectance using the in MCC.
Please feel free to if you have questions about our products and their many uses. Some of the below text may be a bit technical, so please let us know if you have questions.
