Release notes for IDS peak 2.4

Introduction

These release notes describe the changes in IDS peak 2.4. This version introduces additional features in IDS peak comfortSDK such as edge enhancement and AI-assisted image analysis with neural networks from IDS lighthouse.

IDS peak Cockpit

New "Inference" dialog

In the new "Inference" dialog, you can load and execute neural networks for classification and object detection. The prerequisite is that the networks have been downloaded from IDS lighthouse for use with IDS peak.

Optionally, after installing IDS peak 2.4, you find three simple neural networks for testing under <IDS peak installation directory>/cnn. These networks were created solely for testing purposes and do not claim to be operational.

For classification: item_classification.pcla, cats_dogs_classification.pcla
For object detection: person_detection.pdet

Rework of the ROI functionalities

The ROI functionality has been reworked and newly arranged. You will now find the ROI settings in the respective corresponding dialogs:

Set the camera image ROI in the "Position and Size" dialog.
Set the ROI for auto brightness in the "Brightness/frame rate" dialog.
Set the ROI for the auto white balance in the "Color" dialog.
Set the ROI for autofocus in the "Focus/image sharpness" dialog.
NEW: You can also set the ROI for sharpness measurement in the "Focus/image sharpness" dialog.

Further modifications

In the "Host Functions" dialog, you can enable the option for edge enhancement in the image.
The sharpness measurement feature has been moved from the top toolbar to the "Focus/image sharpness" dialog and extended with a sharpness measurement ROI.

IDS peak comfortSDK

You can now execute AI-assisted image processing via the IDS peak comfortC interface. Neural networks for classification and object detection are supported. The prerequisite is that the networks have been generated in IDS lighthouse for use with IDS peak.
Functions for edge enhancement have been added.
Functions for sharpness measurement have been added.

IDS peak genericSDK

The .NET bindings have been updated to use the .NET Standard 2.0. Note the new requirements for .NET version: .NET Framework 4.6.1 or higher / .NET Core 2.0 and higher (including .NET 5 and higher).

IDS peak IPL

The .NET bindings have been updated to use the .NET Standard 2.0. Note the new requirements for .NET version: .NET Framework 4.6.1 or higher / .NET Core 2.0 and higher (including .NET 5 and higher).
Methods for edge enhancement have been added.
Added new methods for image processing: Image:Scale, Image:ScaleInto, and Image:Crop.
New in-place methods were added for hot pixel correction.
The functions for image rotation and image mirroring have been optimized.
Optimizations for the renderer for ARMv8 (64-bit) and ARMv7 Cortex-A/ARMv8 (32-bit).

IDS peak AFL

Added new interfaces for C++, .NET Standard 2.0, and Python.

GEV Transport Layer / U3V Transport Layer / uEye Transport Layer

In the StreamBufferHandlingMode node, the "NewestOnly" mode has been added.
For statistical data of buffer processing there is a new BufferHandlingMonitor category.

General improvements

Version 4.6.1 or higher is required for .NET Framework.
IDS peak now supports Python 3.11. Support for Python 3.5 has been removed.

Component versions used in IDS peak 2.4

Component	Version Windows	Version Linux
IDS peak comfortSDK API	1.4.0.0	1.4.0.0
IDS peak genericSDK API	1.6.0.0	1.6.0.0
IDS peak IPL	1.7.0.0	1.7.0.0
IDS peak AFL	1.1.0.0	1.1.0.0
IDS peak Cockpit	1.5.0.0	1.5.0.0
IDS Vision Cockpit	1.4.4.11	1.4.4.11
Tool Device Command	1.4.2.0	1.4.2.0
Tool Device Update	1.4.3.1	1.4.3.1
Tool IP Config	1.4.1.0	1.4.1.0
Tool Device Password	1.0.0.0	1.0.0.0
GenTL GigE Vision user space	1.11.0.0	1.11.0.0
GenTL GigE Vision kernel	1.11.0.0	-
GigE Vision kernel driver (WHQL)	1.3.1.0	-
GenTL USB3 Vision user space	-	1.11.0.0
GenTL USB3 Vision kernel	1.11.0.0	-
USB3 Vision kernel driver (WHQL)	1.4.6.0	-
GenTL uEye (uEye Transport Layer)	1.11.0.0	1.11.0.0
GenICam	3.2	3.2

Known issues

If network adapters are added, removed, activated or deactivated after opening the IDS peak Cockpit, the changes are only visible after restarting the IDS peak Cockpit. Alternatively, you can manually update the list in the camera manager by clicking on the update button in the upper left corner.
LabVIEW does not support template functions as they are used for FindNode() for example (IDS peak genericSDK). As a workaround, all possible templates of FindNode, FindInvalidatedNode, FindInvalidatingNode, FindSelectedNode and FindSelectingNode are provided as separate functions in addition to the normal generic function.
If network adapters are added, removed, activated or deactivated after opening the IDS Vision Cockpit, the changes are only visible after restarting the IDS Vision Cockpit.
If multiple cameras are open in tiled view in IDS Vision Cockpit, no scroll bar is displayed at the bottom.

Discontinuations

The IDS Vision Cockpit has been discontinued and is therefore no longer recommended for new applications. In this version, it will be distributed for the last time.
With the introduction of the IDS peak comfortC API, users of the IDS Software Suite can quickly and easily switch to IDS peak without having to learn a new programming methodology (e.g. GenICam). For this reason, the IDS sherpa migration guide is not being updated anymore. We recommend the use of the IDS peak comfortC API, which is part of IDS peak from version 2.0 on.

Status: 2023-04-03