On the toolbar, go to Math-->Split Color Channels. Select "RGB" in the dialog box then hit OK. The next few steps will determine the best component(s) to use for selecting the leaves.
IPLab identifies areas based on color separation. Several options are available for separating the area of interest from the background. The method that best identifies the area of interest is chosen, parameters for that method are determined, and a script is developed so that all images in a given set are analyzed in the same method. The analysis of a batch of images may be automated if files are named in sequential numeric order (0001, 0002, 0003....). Script development time varies with complexity of script required and type of image being analyzed. Sample scripts are available. The following procedure outlines the steps to calculate a given area within an image, in this case the area of leaf coverage.
| Step | Image |
| Open .tiff image in IPLab. Zoom in or out on the image if necessary. | |
On the toolbar, go to Math-->Split Color Channels. Select "RGB" in the dialog box then hit OK. The next few steps will determine the best component(s) to use for selecting the leaves. |
|
| This splits the original image into red, green, and blue components. You may need to zoom in on images and rearrange them (Window-->Tile will do this). We are looking for a component that clearly identifies only the area we are interested in. | |
| Minimize component windows from previous step. Go to the Split Color Channels dialog box and choose "HSV". | |
| Minimize component windows from previous step. Go to the Split Color Channels dialog box and choose "RsGsBs". Of all the components, the Hue seems to identify leaf area the best and will be used in the next step. | |
| Minimize all windows except the original image. Create a copy of the original image with Window-->Duplicate Window. The image with a blue bar at the top is the active window. The next step will be performed on this window. | |
| Go to Analysis--> Segmentation. The Segmentation dialog box will appear on the screen shown above. Under component, select the one chosen previously. Hit "Try it" to see how leaves are identified with the parameters shown in the boxes. Parameters may be adjusted with the tabs or by typing in numbers. Image will look something like the one shown below. Hit "Done" when finished. | |
| Multiple components may be necessary for accurate identification, although it is best to use as few parameters as possible. The combine option boxes provide and/or choices for how the multiple parameters interact. The circle with a dot in it indicates which component the number parameters are for. The reverse option selects everything that has not been defined by the component parameters. | |
| The component and parameters used immediately above have identified the leaf area in the image. The next step is to write a script to analyze the images using the components and parameters selected. | |