20.1 Introduction

Intro-Image-Processing

Image window

From version 2021b, Origin introduces a new Image window to load and save images. It employs OpenCV for processing and analysis routines. The new image window is a single-layer window, like the one below. It has an information bar on the top (otherwise you can select View: Show Information to turn it on) and if a video or multi-frame image, a navigation bar at the bottom (or View: Show Navigation to turn it on). It is responsible for almost all image processing performance.

Single-Frame Image	Multi-Frame Image or Video

Image window vs. Matrix window Before Origin 2021b, images were stored in the Matrix window and the processing tools were supported by LeadTools. On the other hand, the new Image window is supported by OpenCV for processing and analysis routines. The matrix window is a multi-layer window that can contain many sheets. However, an image window has one sheet only. The matrix window is responsible for almost all built-in image processing tools. Also, if a process or analysis outputs some data results, such as 2D FFT, the matrix window is used. On the other hand, image window is used in most of image/video processing Apps. You can always convert an image window to matrix and reverse back. See below.

Import Images and Videos

Refer to this document page for details.

Menus for Basic Operation on Window

Edit: Cut: clear the active image window.

Edit: Clear: if a ROI is selected, this menu will delete the ROI; if the image window is selected, this menu will clear the window.

Edit: Copy: if a ROI is selected, this menu will copy the ROI. If the image window is selected, this menu will copy the image. You can then paste it to a graph window as an object or to an image window.

Edit: Paste: paste the copied ROI or image.

View: Information: show/hide the information bar

View: Navigation: show/hide the Navigation bar

Export Images and Videos

Refer to this page for details.

Basic Operations on Image Window

Besides tools under Image main menu, toolbar buttons and context menus are provided for quickly access to basic manipulations.

Mini-Toolbar

Clicking anywhere inside the image layer will pop up a mini toolbar.

Single frame Image:

Video:

Multi-frame stacked Image:

Applying Palettes to Grayscale Images

Color palettes can be applied to grayscale images in the Image window.

1. Click the Mini Toolbar Gray button , if necessary, to convert an image to grayscale.
2. Apply a palette to the grayscale image by clicking the Palette button on the Style toolbar.
3. Click the Mini Toolbar Reverse Palette button to reverse application of the chosen palette.

From the Palette list on the Style toolbar, click More Palettes to open the Color Manager dialog box and add more palettes.

Context Menu

Change Background Color

• Use Fill Color button on toolbar to change image’s background color.

• Use Labtalk script page.basecolor= to change image’s background color. Refer to Labtalk Page object

page.basecolor= 3; // Change image background color to green.

Region of Interest

To add a ROI box

• Click the Add ROI button on minitoolbar (the shape of the ROI box is the <Last used> one)

Or,

• Click the triangle ▼ next to Rectangle Tool button on Tools toolbar, select one of the shapes (Rectangle, Circle, Polygon, Region Tool) from the sub-list. Drag a region on the image.

Or,

• Right click inside Image window and select Add ROI context menu (the shape of the ROI box is the <Last used> one)

To add multiple ROI boxes to a single Image window:

1. After adding a ROI box to the image,
   • click the Add ROI button on minitoolbar, or
   • right click and select Add ROI context menu

to add another ROI box.

Or,

1. After adding a ROI box to the image, double click on it to open the ROI Position dialog box. Rename it to something like "ROI1".
2. Add a new ROI box by one of the 3 ways in above section.

Once added, you can view all the ROIs in Object Manager and perform basic operations.

Double-click on the ROI to:

• Change the ROI name
• Re-position the ROI by specifying XY coordinates of the top-left and bottom-right anchor points.

Right-click on the ROI to:

• Clip: Available when the image is opened from a graph (e.g. by double-clicking the graph image). This clips the graph image to the Image window ROI. The original graph image size is reduced to the size of the clipped region. To restore the full image in the graph window, right-click outside of the ROI and choose Remove Clipping.
• Crop: Crop Image window image and graph image (if exists) to the ROI. Graph image size and position are kept.
• Copy: Copy the ROI as an image. You can then paste it to other windows inside Origin, e.g. a graph or worksheet window, or to other applications. For multi-frame image, only the active frame is copied.
• Apply Scale to Graph: Available when the image is inserted into a graph by Image from Image Window. This updates the graph image to the Image window ROI.
• Copy Positions: Copy the ROI box coordinates of 4 vertices. You can choose to paste it as plain numbers, or, Paste Position to another ROI box. For the latter option, you can choose to paste only ROI Width/Height, or Horizontal/Vertical location.
• Export ROI: Saves details (e.g. ROI name, position of vertices, etc.) to a .ROI file and later.
• Import ROI: Applies details of a previously-saved ROI, to the selected ROI.
• Create New from XY: Create a new ROI box of arbitrary shape from a set of XY coordinates (boundary) and index data. See also Create ROI from XY in NetCDF data.
• Create Linked Image: Outputs the ROI to a new Image window. If you move ROI or adjust the ROI box size or shape, the new window will update accordingly.
• Create Linked Matrix: Outputs the data inside ROI to a new Matrix window. The matrix will update according to the Image window ROI.
• Intensity Profile: Perform intensity profile on ROI and output descriptive statistics value to a worksheet. See below section for details.

The degree of image darkening outside of the Image Window ROI is controlled by system variable @IMGDI. Setting @IMGDI = 1000 will cause the outside area to display as black; @IMGDI = 0 turns off darkening. For information on changing the value of a system variable, see this FAQ.

Intensity Profile on ROI

Right click on the ROI box and select Intensity Profile from context menu.

This will open the cvroiprofile dialog.

• You can choose to take profile on single ROI (Active ROI) or multiple ROIs (All ROIs on Image). If the Image window contains multi-frames, every frame's result will be output.
• Select the descriptive statistics value to output under Statistics branch: Mean, SD, Min, Max, Median, Sum, XY of Min and XY of Max.
• If Weighted ROI check box is selected, pixels on/near to the ROI border (ROI cannot fully corver one pixel) will be weighted by the percentage of area inside the ROI. This checkbox is usefaul when the ROI is tiny, in which case the edge pixels have a bigger effect on the whole ROI.

Refer to the X-Function cvroiprofile for more details.

Create Graph from Image

Select menu Image: Plot Image.

• This will create a new graph window and set the current image as its background.
• The created graph window will automatically set layer isometric (Plot Details > layer level > Size tab > Link Axis Length to Scale with X/Y RATIO = 1).
• The graph page is auto-rescaled to fit to layer size.
• If the Image's XY coordinaes change, e.g. the Image is a linked image created from a ROI box on another Image and the ROI is relocated, the graph will be rescaled accordingly. You can turn off this feature by clicking Auto Rescale Background Image button on the layer level mini-toolbar.

Learn more about inserting an image to a graph on this page.

Adjustments and Conversion

Besides buttons and menus mentioned above, Origin provides a sets of tools useful in image processing and analysis. They are sub-categorized as follow:

• Setting Image Scale and Dimensions
• Converting Image

Converting to Matrix and Reversion

As different ways of dataset display and storage, sometimes you need to convert betwenn Image window and Matrix window. Data type can be specified during the conversion, which should take in account when the post-processing has a requirement on the data type.

Convert an image (single chanel) into matrix,

• Select menu Image: Convert to Matrix to open the Cv2Mat dialog.

Following data types are supported:

Convert a matrix to image window,

• You can do it in Python. Below is a sample.

import originpro as op
import numpy as np
ml = op.find_sheet('m')
a = ml.to_np3d()
iw=op.new_image()
iw.setup(1,True)
iw.from_np(a)
iw.set_int('NAV',1)