Import and Export Image Data

You are currently viewing the documentation for the latest version (3.0.0). To access a different version, click the "Switch version" button located in the upper-right corner of the page.

■ If you are not sure which version of the product you are currently using, please feel free to contact Mech-Mind Technical Support.

This topic describes how to import and export image data in an algorithm module.

import export data

Import Image Data

Before importing images, you should follow the data acquisition standard to acquire the required images for the business. In Mech-DLK, there is no need to configure a camera. You only need to import images to train the model.

You can use the following method to import images:

Method 1

Drag and drop images or files into the image list area to import them. Importing datasets by dragging is not supported.

Method 2

  1. In the Modules section at the top right corner of the interface, select the module that you need to import images.

  2. On the top of the image list, click Import/Export.

  3. Select the import method based on the data type:

    • Import from Previous Module: Import images from the previous module.

    • Import Image: Import one or more images.

    • Import Folder: Import all images in the folder (images in the subfolders are not included).

    • Import Dataset: Import datasets in DLKDB format (.dlkdb) or COCO format (for Instance Segmentation and Object Detection) into Mech-DLK.

  4. If the dataset contains duplicate images, you can choose to skip or import them in the Import Images dialog. You can also choose to apply new tags or keep the existing ones for the imported images.

    • Dialog box for the Import Images or Import Folder option:

      example projects duplicate1

    • Dialog box for the Import Dataset option:

      example projects duplicate2
      example projects duplicate3

  • If you import images from Mech-MSR with images in MRAW format, you can set the images as depth maps or intensity images and adjust parameter settings of those images in the Image Preprocessing Tool dialog box.

  • Only the Input module supports importing images in MRAW format, while other modules do not.

  • Synchronized import of depth maps and color images is supported. Please make sure that the two image folders are in the same directory, and the image files correspond and are of the same size. When importing any type of image, the system automatically imports the corresponding image of the other type.

Import from Previous Module

Import from Previous Module is supported only when the previous module (parent module) of the module has been validated. Otherwise, this button is unavailable. The Input module does not support the Import from Previous Module function.

Definitions of parent module and submodule:

  • Parent module: the previous module directly connected to the current module.

  • Submodule: the next module directly connected to the current module.

As shown below, Fast Positioning is a submodule of the Input module and a parent module of Object Detection.

parent module submodule definition

Procedure:

  1. At the top of the image list, choose Import/Export  Import from Previous Module to open the Import from Previous Module window.

  2. In the Import from Previous Module window, complete the Process images and Select images steps, and then click Import.

import from previous module

Parameter description:

Step Parameter Description

Process images

Mask settings

Set a mask for images. Three mask modes are supported: No Mask, Auto Mask, and Inverse Mask.

  • No mask: Do not display any mask. The image is shown in its original state.

  • Auto mask: Use masks to mark the validation result area of the previous module.

  • Inverse mask: Use masks to mark the non-validation result area of the previous module.

Apply validation results

Whether to apply the validation result area of the previous module.

  • When this function is enabled, the validation result area from the previous module will be used as the initial ROI area, marked with a crosshair. You can further adjust the ROI area through ROI and ROI segmentation.

  • When this function is enabled, images marked as OK will not be shown or included in filtering. When the function is disabled, they will be shown normally. (Available only when the previous module is a Defect Segmentation or Unsupervised Segmentation module.)

Enable this function if the current module needs to use the previous module’s validation result area as the initial ROI for image processing. This function is enabled by default. If disabled, the original images that have not been processed by the previous module will be imported.

ROI:

  • Supports adjusting ROI groups By ratio or By pixel.

    • By ratio: Scale and offset each ROI group proportionally.

    • By pixel: Scale and offset each ROI group using the same pixel values.

  • Supports drawing ROIs by clicking Add.

In the visualization area, you can drag an ROI with the mouse to adjust it, or adjust it by setting Width, Height, X-axis offset, Y-axis offset, and Offset angle.

ROI segmentation:

Divide the selected ROI area into specified rows and columns, with X-axis spacing and Y-axis spacing.

When there are many detection targets arranged in a regular pattern, you can use ROI segmentation to batch-generate multiple ROI areas.

  • Whether adjusted by ratio or by pixel, newly added ROIs are always scaled and offset based on the initial ROI imported from the previous module.

  • The reset button icon reset is only available when Apply validation results is enabled. After clicking this button, the ROI parameters will revert to their original default values imported from the previous module.

Select images

Options for duplicates

If there are duplicate images, you can Import or Skip them.

Image tag

Supports Do not add an image tag, Use an existing image tag, or Add a new image tag.

  • For the Pick Anything V2 and Object-Bin Segmentation modules, when Import from Previous Module is selected, only Options for duplicates and Image tag can be configured.

  • The import configuration may vary under different module combination. Please set the parameters according to what is displayed.

Export Image Data

You can use the following method to export image data:

Method 1: Export all data

  1. In the Modules section at the top right corner of the interface, select the module that you need to export image data.

  2. On the top of the image list, click Import/Export.

  3. Select Export Dataset.

  4. In the pop-up Export as DLKDB window, enter a file name and specify a directory to save the file.

Method 2: Export part of the data

  1. In the image list, select the images that you need to export.

  2. Right click the images, and then click Export dataset.

  3. In the pop-up Export as DLKDB window, enter a file name and specify a directory to save the file.

Method 3: Export all images

  1. In the Modules section at the top right corner of the interface, select the module that you need to export images.

  2. On the top of the image list, click Import/Export.

  3. Select Export Image(s).

  4. In the pop-up Select Folder window, select the folder where the data is stored.

Method 4: Export some images

  1. In the image list, select the images that you need to export.

  2. Right-click and select Export Image(s).

  3. In the pop-up Select Folder window, select the folder where the data is stored.

Only Import Images, Import Folder, and Export Image(s) are supported in the Input module.

Is this page helpful?

You can give a feedback in any of the following ways:

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.

  • Preference
  • I accept
  • I don't accept