
eCognition – Trimble’s Software for Image intelligence.
We offer a suitable package tailored to the many possibilities of the software.
eCognition 10.5 – Release March 2025 – Highlights

© Trimble
eCognition 10.5 offers even more flexibility in using external CNN models
With the integration of the ONNX format (Open Neural Network Exchange), a wide range of pre-trained models can now be incorporated into the eCognition workflow.
ONNX Model Support
ONNX (Open Neural Network Exchange) now supported in eCognition as a third CNN-format (eCognition model, tensorflow model, ONNX model) allows users to import and run larger number of pretrained models.
SAM-based Segmentation
Version 10.5 introducing a new algorithm for segmentation of image data based on the Segment Anything Model (SAM) algorithm*. The algorithm offers three input modes for segmentation flexibility: Everything | Bounding Boxes | Points
Point Cloud Editing Toolbar
The new toolbar designed specifically for point cloud editing. It allows to select points in various ways (rectangle, polygon, cuboid, all) and assign a classification to selected points. The toolbar automatically opens when point cloud data is imported.
Tama Goup gives an insight into the highlights of eCognition 10.5 in short videos
The first video focuses on the support for ONNX / Open Neural Network Exchange introduced with eCognition 10.5. Michael Voltersen | Senior Ruleware Developer Tama Goup explains with practical steps the process and what needs to be considered when using ONNX in eCognition. (DE)
Office software
Trimble eCognition
Transform data into information.
Here you will find essential information about eCognition from Trimble.
We have compiled examples of possible applications for various areas to show the range of possibilities with eCognition.
eCognition Distribution & Support
We are pleased to offer a customised licensing and support package that is tailored to the many possibilities of the software. This includes a maintenance package that integrates software maintenance, level 1 & 2 support and training options. To provide you with optimal support, our offers are based on the Tama Core Process for eCognition.


Tama Core Process
A five-step guide to achieving image intelligence with eCognition. The Tama Basic Process outlines the fundamental steps of the automated analysis process using eCognition. We have also taken into account the possibilities of CNNs and the integration of Python scripts. Our range of training is based on these fundamental steps.
After creating a project by (1) importing all available data, additional (2) layers have to be generated to increase the information content for the following analysis. The subsequent (3) segmentation and (4) classification of the newly generated objects can be repeated in mutual influence any number of times. Finally, the (5) export of extracted information is done, may be spatially located in a map or statistical edited in chart form.