LandingEdge Release Notes

December 19, 2024: Expanded PLC Options (v2.10.31)

When setting up communication between LandingEdge and a Programmable Logic Controller (PLC), you can now configure custom signals. For example, you can set up a signal that sends the "lot name" of a batch of images from a PLC to LandingEdge.

Custom Signal in LandingEdge

To allow you to include custom signals in Results Processing scripts in LandingEdge, we've added the ReadInput and WriteOutput methods. Use these methods to convert the signal values to variables that you can use in Python or C# scripts.

For detailed information about setting up custom signals or using them in scripts, go to PLC Workflow.  

August 27, 2024: OCR with LandingEdge & Docker

LandingAI now offers an optical character recognition (OCR) solution through LandingEdge and Docker. Use OCR to extract text from images to convert unstructured content into structured, actionable data. Our OCR solution supports English and Simplified Chinese character sets and can detect multiple languages in one image.

This OCR solution is available as an add-on for users on the LandingLens Enterprise plan. 

When LandingEdge runs OCR on an image, the predicted text displays on the image and is included in the JSON response. For detailed information, go to OCR.

The Predicted Text Displays as an Overlay in LandingEdge

March 25, 2024 - v2.7.204

We released a new version of LandingEdge today! The newest version includes these updates:

• For Windows users only: When you install LandingEdge, LandingEdge will now check if your system has Microsoft Visual C++ Redistributable. If this package is not detected, LandingEdge will install it. For context, LandEdge requires the Microsoft C and C++ (MSVC) runtime libraries. The Microsoft Visual C++ Redistributable package installs these libraries.
• We fixed an issue that caused OPC UA controller connections to PLCs to fail. For more information about using PLCs with LandingEdge, go to PLC Workflow.

February 22, 2024 - v2.7.188

You can use LandingEdge to upload images to a LandingLens project without running inference on those images. The images are uploaded to the Deploy page for the project in LandingLens. Images uploaded to the Deploy page can then be added to the project dataset.

This method is particularly helpful if you want to upload images directly from a camera into LandingLens. For example, if you have an industrial camera on a manufacturing line, you can set up an Inspection Point in LandingEdge to capture images from that camera. That way, you don’t need to first download images from the camera and then upload them into LandingLens.

To upload images using LandingEdge without running inference:

1. Open LandingEdge.
2. Open or create an Inspection Point.
3. Set up the Image Source, Inspection Start, and Cloud Connection sections.
4. Leave the Model section blank.
5. In the Image Saving section, select the Upload Results to LandingLens checkbox.
   Set Up the Inspection Point for Upload Only
6. Click Save Configuration.
7. When you're ready to upload images, click Run Now and upload the images using the method you selected in the Image Source section. For example, if you selected Folder Watcher, move the images you want to upload to the folder you selected.
8. To see the images in LandingLens:
   1. Open the project in LandingLens.
   2. Click Deploy.
   3. In the Self-Hosted Deployment section, click the name of the Inspection Point that you created in LandingEdge.
   4. The images display.
      
      Images Uploaded from LandingEdge Display on the Deploy Page

February 15, 2024 - v2.7.167

In the newest version of LandingEdge, we’ve improved processing time and web API latency. We’ve also added a few improvements and bug fixes. Keep reading to learn more!

Expanded Support for Industrial Cameras

We’ve added an additional method, called "GigE Vision”, to run inference on images from industrial cameras. This is an alternate option to the “GenICam” option that is designed to be more efficient and robust. For more information on using industrial cameras, go to GenICam and GigE Vision Cameras.

JSON Results Include Model ID and Latency

When you run inference on an image via LandingEdge, the JSON results now include the Model ID (model_id) and Latency (latency) information.

Send Images for Reference via URL

If using web APIs to upload images for inference, you can now send a URL of the image. Previously, only an image file could be sent. Sending the image URL is supported by the /images endpoint. For more information about using web APIs, go to Run Inference with Web APIs.

Bug Fixes

• We fixed an issue that occurred in certain use cases with segmentation mask calculation.
• We fixed issues related to the user interface.
• We fixed issues related to running multiple GPUs.

September 26, 2023 - v2.6.87

In the latest version of LandingEdge, you can now send images from different IP addresses (different devices) for inference using web APIs. You can enable this functionality by selecting the Allow External Access checkbox when setting up your Inspection Point. For more information, go to Run Inference with Web APIs.

Select "Allow External Access" to Send Images from Multiple Devices

Bug Fixes

• After running inference on an Anomaly Detection model, the results now display correctly in the user interface. (Anomaly Detection is deprecated and only available to legacy users.)
• Role names are now easier to see in the User Menu.

August 24, 2023 - v2.6.45

The newest version of LandingEdge offers these new features:

• Roles and Permissions
• Assign GPUs to Inspection Points
• Configurations

Roles and Permissions

Often, LandingEdge is deployed to a shared computer, such as a machine on a manufacturing floor. Several users might access both the machine and LandingEdge. To prevent users from accidentally changing Inspection Points or stopping the inspection process, LandingEdge now offers Roles to limit who can access different parts of the system:

• Admin: Can access and edit all parts of LandingEdge. 
• Engineer: Can access and edit all parts of LandingEdge except for editing Role permissions.
• Operator: Has read-only access. 

If LandingEdge has been idle for 15 minutes, the current user is logged out of the Role. You can then enter a password to log in as an Admin or Engineer. Or, you can log in as an Operator, who by default has read-only access.

To manage Roles in LandingEdge, go to Settings > Permissions.

Roles and Permissions

Do I need to do anything now that there are Roles?

If you would like to use Roles, set up the passwords for the Admin and Engineer Roles. For detailed information, go to Roles and Permissions.

If you don't plan on using Roles—for example, if LandingEdge runs on a computer that only one person has access to—you don’t need to set up passwords or any of the Permissions settings. If you don’t set a password for any of the Roles, you won’t be logged out after the Idle Time.

Assign GPUs to Inspection Points

When setting up an Inspection Point, you can now assign one or more specific GPUs from your system to power the Inspection Point. This helps free up your CPU for other processes. For more information, go to GPUs and LandingEdge.

Assign GPUs

Configurations

You can now set up multiple configurations for each Inspection Point. A configuration is the collection of settings for the Inspection Point. Set up multiple configurations if you want to switch between settings without changing the Inspection Point itself.

If you're using a PLC, you can set it up to automatically switch between configurations. For more information about using a PLC with LandingEdge, go to PLC Workflow.

If you're not using a PLC, you can manually change the configuration in the LandingEdge interface.

Configurations are managed in the Configuration field at the top of the Inspection Point settings.

Configurations

If You Don't Need Multiple Configurations

If you don't need multiple configurations for an Inspection Point, ignore the Configuration field.

July 11, 2023 - v2.5.10

We've added more custom scripting capabilities to LandingEdge. You can now read out the raw scores from a Classification model. We've also introduced the AllClasses Dictionary, which provides access to all names and indices of all Classes in the model. For more information, go to Custom Processing.

Here is a sample C# post-processing script for reading out the raw scores from a Classification model:

public void Run()
{
    if (Result.PrimaryResult is not ClassificationResult clf)
    {
        // not a classification result
        // this script is only applicable to classification, so exit the script
        return;
    }

    Log($"All classes in model: {string.Join(", ", AllClasses)}");
    Log($"Class scores for image: [{string.Join(", ", clf.Predictions.RawScores)}]");

    for (int i=0; i<AllClasses.Count; i++)
    {
        Log($"Class '{AllClasses[i].Name}' score = {clf.Predictions.RawScores[i]}");
    }    
}

July 7, 2023

The updated version of LandingEdge includes these updates:

• You can now use LandingEdge to pull stills (individual frames) from videos and run inference on those images. For more information, go to Image Source.
• Support for the new API Key system. For more information, go to API Keys.
• If you use LandingEdge with a PLC, you can now configure an Error Code output. For more information, go to Outputs. 
• A few behind-the-scenes bug fixes.

June 14, 2023

Download a Free Trial of LandingEdge

You can now download a free trial of LandingEdge directly from LandingLens! 

To download the free trial, open a Project to the Deploy page and click LandingEdge. Then follow the on-screen prompts to complete the download. For detailed information, go to LandingEdge Overview.

Download LandingEdge

June 8, 2023

What's New

• The Settings icon is now permanently on the LandingEdge homepage, even if you haven’t created any Inspection Points yet. Previously, you could only access Settings after setting up an Inspection Point. 

  
  This makes it easier to get your License Key, which you can access in Settings > Update License.

  
  Click the Settings Icon
• We've made it easier to see when your license will expire. If your license will expire within 120 days, a message on the Settings page notifies you how many days are left until the expiration date.
  Countdown Until Your License Expires
  Previously, if the license expired within 120 days, that notification would always be “Your license will expire in less than 120 days.”
• If you try to upload an invalid license, you now get a pop-up letting you know that the license isn’t valid.
• To make it easier to understand some of the features, we updated some of the user interface text throughout LandingEdge.

Bug Fixes

• We fixed a bug that caused issues when stopping inspections in some situations.
• We fixed a few broken external references in our C# scripts.

May 12, 2023

The latest release of LandingEdge makes it much easier to activate your Deployment License. Most notably, you can now get a 14-day free trial of LandingEdge! This is the perfect way to vet LandingEdge for your deployment needs. 

What’s New

• You can now get a 14-day free trial of LandingEdge.
  Get Trial License
• LandingEdge now notifies you when your LandingEdge Deployment License will expire within 120 days, so that you can request a new license. LandingEdge also notifies you once your license has expired. Once you have your updated license, click the notification to add your license to LandingEdge.
  License Expiring Soon
• You can now update your LandingEdge license even if your current license is still valid. Previously, you could only add a LandingEdge license if your previous one was expired. With this new workflow, you can now avoid lapses to your LandingEdge functionality.
  Update License
• If you’ve enabled the Upload Results to LandingLenssetting, images uploaded to LandingLens via LandingEdge now have image metadata. The metadata follows this format:

  {
   "imageId": "28587.jpg",
   "inspectionStationId": "11",
   "locationId": "California#11",
   "captureTimestamp": "2021-10-11T12:00:00.00000"
   }

• If you’ve enabled the Upload Results to LandingLens setting, you can run a script to automatically skip saving certain images. For more information, go here.

Bug Fixes

• Fixed a bug that caused changes to Inspection Point names to not save. LandingEdge now correctly saves changes to Inspection Point names.
• Fixed a bug that caused the PLC information to not display in the Communication settings. The information displays correctly now.

April 12, 2023

LandingEdge now supports Linux! You can run LandingEdge on Linux Ubuntu 20.04 and 22.04.

When you download LandingEdge, select the Linux version. For detailed download instructions, go to Download and Install LandingEdge.

Download LandingEdge for Linux

April 5, 2023

What's New

• The Settings module now shows your LandingEdge version. For more information, go to Version.
  LandingEdge Version
• You have more ways to trigger image capture. For more information, go to Inspection Start.
  • Self: Configure your camera or hardware device to trigger when the image will be taken. This is the only option for WebAPI and Folder Watcher. 
  • Manual: In Manual mode, click the Grab a Frame button on the Monitor page to trigger the camera to capture the image.
    Manual Mode

    
    

    "Grab a Frame" Is Available in Manual Mode to Capture an Image
• You can now add image metadata by using the Custom Scripting feature. For more information, go to Add Metadata to Images.
  Post-Processing Script to Update Metadata

   

  Customized Metadata
• When you attempt to close LandingEdge, you are now prompted to confirm the action, because closing the application stops all Inspection Points.
  Prompt When You Close LandingEdge

Bug Fixes

• You can now download images that display in the image Preview. Previously, the download button was deactivated.
  Download an Image from Image Preview
• Resolved issues with incorrect results for Object Detection Projects that occurred in some systems. 
• Resolved issues with high load conditions.
• Fixed miscellaneous bugs. 

March 23, 2023

You can now customize and apply your own automation using C# or Python scripts to your Inspection Point in LandingEdge. This allows you to automate an action to occur before or after inference. For more information on custom processing, go to Manage Inspection Points.

To use this feature, download the latest version of LandingEdge. To learn more about how to download LandingEdge, go here.

Apply Custom Automations to LandingEdge Inspection Points 

March 10, 2023

LandingEdge v2 is now available! This newest version of LandingEdge includes some bug fixes and performance improvements. If you're using LandingEdge v1, we recommend upgrading to the newest version to get the latest developments.

Please Note:

• Any Inspection Points from LandingEdge v1 will not carry over to LandingEdge v2 and will need to be recreated as needed.
• Webcams are not supported in LandingEdge v2.
• The background application LandingEdge Inference Engine has been integrated into LandingEdge v2 and will no longer display as a separate application on your desktop.