Cognition Network Technology (CNT), also known as Definiens Cognition Network Technology, is an object-based image analysis method developed by Nobel laureate Gerd Binnig together with a team of researchers at Definiens AG in Munich, Germany. It serves for extracting information from images using a hierarchy of image objects (groups of pixels), as opposed to traditional pixel processing methods. To emulate the human mind's cognitive powers, Definiens used patented image segmentation and classification processes, and developed a method to render knowledge in a semantic network. CNT examines pixels not in isolation, but in context, building up a picture iteratively and recognizing groups of pixels as objects. It uses color, shape, texture, and size of objects as well as their context and relationships to draw conclusions and inferences, similar to human analysis.
History
In 1994, Professor Gerd Binnig, a Nobel laureate in physics, founded Definiens AG in Munich. CNT was first available with the launch of the eCognition software in May 2000. In June 2010, Trimble Navigation Ltd (NASDAQ: TRMB) acquired Definiens' business assets in earth sciences markets, including eCognition software, and also licensed Definiens' patented CNT. In 2014, Definiens was acquired by MedImmune, the global biologics research and development arm of AstraZeneca, for an initial consideration of $150 million. This acquisition shifted the company's focus toward biomedical applications, particularly digital pathology.
Core Principles
CNT operates on the principle of object-based image analysis (OBIA), which contrasts with traditional pixel-based methods. Instead of analyzing individual pixels in isolation, CNT groups pixels into meaningful objects based on spectral, spatial, and contextual properties. These objects are organized in a hierarchical network, allowing the system to analyze images at multiple scales. The method leverages a semantic network to represent knowledge about image content, enabling the system to reason about relationships between objects, similar to human visual cognition. This approach is particularly effective for complex images where context and object relationships are crucial, such as remote sensing and medical imaging.
Software Implementations
eCognition
The eCognition suite is the primary software implementation of CNT for earth sciences and remote sensing. It offers three components: eCognition Developer, a development environment for creating rule sets and applications; eCognition Architect, which allows non-technical users to configure and execute workflows; and eCognition Server, which provides batch processing capabilities. eCognition is used in diverse applications, including rapid mapping, change detection, landcover mapping, urban analysis (e.g., impervious surface area for taxation), forestry (biomass measurement, species identification), agriculture (precision farming, crisis response), and marine and riparian environments (ecosystem evaluation, disaster management). The online eCognition community was launched in July 2009 and had 2,813 members as of July 9, 2010. User conferences are held regularly, and the bi-annual GEOBIA (Geographic Object-Based Image Analysis) conference is heavily attended by eCognition users.
Definiens Tissue Studio
Definiens Tissue Studio is a digital pathology image analysis software application based on CNT, designed for biomarker translational research. It is intended for use with formalin-fixed, paraffin-embedded tissue samples treated with immunohistochemical staining or hematoxylin and eosin (H&E). The software facilitates machine learning from example digital histopathology images to derive an image analysis solution for measuring biomarkers and histological features within defined regions of interest on a cell-by-cell basis, including sub-cellular compartments. The derived solution is automatically applied to subsequent images to objectively measure multiparametric features, aiding in understanding biological processes underlying diseases like cancer. Image processing can be performed on a local workstation or a server grid.
Applications and Impact
CNT has been applied across various fields, from geospatial analysis to biomedical research. In earth sciences, it enables detailed and accurate analysis of satellite and aerial imagery, supporting applications in environmental monitoring, urban planning, and disaster response. In digital pathology, CNT-based tools like Definiens Tissue Studio assist researchers in quantifying biomarkers and histological features, contributing to advancements in personalized medicine. The method's ability to incorporate context and object relationships makes it a powerful alternative to traditional pixel-based approaches, particularly in scenarios where high accuracy and interpretability are required. As of the early 2020s, CNT remains a foundational technology in object-based image analysis, with ongoing developments in Machine learning and Deep learning enhancing its capabilities.
See Also
- active appearance model
- Computer vision
- image-segmentation
- semantic-network