2D Model Editor

This tool is used to generate and edit the models for 2D matching. These models can be used in the Alignment Step to align the input images with the selected model.

Preparation

  1. Make sure that the Alignment Step has normal input.

  2. Under Parameters in the parameter configuration panel of the Alignment Step, click the Edit Model button to open the 2D Model Editor. This tool will automatically obtain the input image of the Step.

Click Get Step input in the parameter panel of 2D Model Editor to update the image.

Usage Workflow

  1. Rename the default model or click + New to create a new model.

    After selecting a model from in the model list, right-click and select the Rename option, enter a custom name, and click OK to modify the model name.
  2. Select feature points. The selected feature points should be representative, and the quantity should be appropriate.

    Instruction: Click the icon on the toolbar to select the appropriate ROI tools and label the image.

    • In the ROI list section at the bottom left of the interface, right-click an ROI and select Enable, Disable, etc., to set the drawn ROI. For overlapped regions, the Disable option is more effective than Enable. In other words, when two overlapped regions are set to Enable and Disable separately, the actual status of the overlapped region is Disable, and the feature points in the overlapped region do not take effect.

    • You can use the ROI Eraser roi eraser to erase unwanted feature points in the created ROI.

2d template rename roi
  1. Set the center point. The center point determines the position of the model in the image to be matched, thus determining the search range for model matching.

  2. Set Feature parameters.

  3. Click Generate model in the lower-right corner to generate and preview the model.

  4. Click Save.

    If you only clicked Save in the lower-right corner and did not generate a model, the edited content would be saved but would not work.

After configuration, close the tool. Click under the Edit Model button in the Parameters section of the Alignment Step and then select the corresponding model in the drop-down menu.

Feature Parameters

Gradient magnitude threshold

Points with gradient magnitude greater than the threshold will be considered feature points. Set this parameter to filter out feature points with small gradient magnitude.

Instruction: When a high threshold is set, only points with large gradient magnitude, i.e., points that vary significantly in the image, will be kept. When a low threshold is set, some noise or insignificant feature points may be kept.

Number of feature points

The expected number of feature points extracted from the model image. A greater value means a larger feature point count.

Instruction: Adding more feature points can improve the matching accuracy, but also increase the calculation complexity and reduce the matching speed.

If the number of feature points in the generated model is smaller than the set one, the number of feature points in the model is subject to the actual one.
Kernel size of NMS

The kernel size of non-maximum suppression (NMS). This parameter is used to regulate the distribution density of the feature points. The larger the value, the sparser the feature points.

Instruction: Set the parameter according to actual situation.

Set angle

Set the angle-related parameters to define the allowable rotation angle range of the image to be aligned.

  • Range (°): Set this parameter to determine the possible rotation angle range of the target object in the image relative to the model during model matching.

  • Step size (°): The angle interval during matching. Setting a small step size will improve the search accuracy but slow down the search process.

  • Attempt count: The number of matching attempts within the given angle range.

Set scale

Set the scale-related parameters to define the allowable scale range of the image to be aligned.

  • Range: Set this parameter to determine the scale range that the target object will appear in the image during model matching.

  • Step size: The scale interval during matching.

  • Attempt count: The number of matching attempts within the given scale range.

We Value Your Privacy

We use cookies to provide you with the best possible experience on our website. By continuing to use the site, you acknowledge that you agree to the use of cookies. If you decline, a single cookie will be used to ensure you're not tracked or remembered when you visit this website.