HedeDanmark LogSystems
In collaboration with HedeDanmark, Talmo has developed both prototypes, devised the camera design and implemented the final LogStackPro production software. LogStackPro is a Linux-based, hand-held and certified measuring device, that has been developed and customized for digital photo measurements of raw wood stacks in the forest.
Based on input from a stereo camera the device reconstructs the metric 3D structure of the stack for measuring both diameter distributions, density and final stack volumes without requiring any references in the image.
The LogStackPro software is based upon a wide variety of custom vision techniques, including photogrammetry, simultaneous localization and mapping along with deep learning-based detection and segmentation for precise estimation of log positions and stack density.
Kozoom Carom Billiard
For Kozoom Carom Billiard, Talmo has developed a desktop application that is capable of analyzing and generating 4k video at 60 fps captured through Blackmagic DeckLink equipment.
The developed application detects the Carom table, tracks the three balls and freezes the exact moment when the shot is made. The analyzed results are overlaid on the input video and output through the Decklink card for mixing with other camera sources for visualizing details after the shot has been made. The developed application enables the video operator to control which balls are tracked and visual details regarding the graphics overlay.
The video analysis operates at 60 fps hence requiring highly optimized routines for video input processing and scaling, color space and bit depth conversions along with the actual object tracking.
SBS Spare Parts
Together with brake pad manufacturer SBS, Talmo has worked on prototypes for detecting and identifying spare parts from images. Motor-cycle brake pads come in many shapes and sizes, where each brake pad is compatible with only a certain subset of end-user braking systems.
To alleviate existing SBS CAD-based data for each known brake pad, Talmo has worked on a rendering solution, where multiple synthetic images can be generated with random wear and tear for each model, from various viewpoints, lighting setups and background scenes.
The generated images can be used for training both detection and segmentation systems without the need for manually acquiring and annotating millions of images to support all combinations. The trained inference models can be used both for factory quality control, but also for end-user apps capable of identifying the actual brake pads from phone images, while finally providing suggested replacement model numbers.
Terma Counter Drone
Talmo has helped Terma develop video-based drone detection software for observing and tracking intruding UAVs in high-end surveillance cameras for deployment at critical infrastructure sites.
The solution performs registration of the video stream for continuously learning the camera pose, along with employing different deep learning-based detectors trained on both real and synthetic data. Finally, the system performs object tracking based on motion models and Kalman filtering.
The tasks performed involved evaluating different detector architectures with respect to both their training and inference capabilities for deployment in a large system with many parallel camera streams. Moreover, the tasks involved devising ideas for fusing radar and camera tracks.
Christmas tree yield estimation
Talmo has been part of a consortium along with University of Southern Denmark (SDU), Hermes Traffic Intelligence and Green Team Group exploring how drones can be used to give a status of the Christmas trees in a given plantation including estimating the number of Christmas trees and the heights of the individual trees.
In the agriculture industry, like in many other sectors, there is a great interest in optimizing specific processes in the production. In order to optimize a process, you need information about the conditions of the system that the process is affecting. If for example you need to plan a sale, it is important to know what is in stock.
Talmo has supplied knowledge about photogrammetry, object detection and segmentation along with 3D data analysis to the project.
TCBS Carom Billiard Trainer
Together with carom billiard champion Tonny Carlsen an advanced ball tracking and training system has been developed by Talmo.
Based on an industrial vision camera mounted above the billiard table, the developed application assists players through an intuitive cross-platform user interface. The player receives immediate feedback on the chosen trajectory and different training modes enable use by both beginners and pros.
The video analysis operates at 60 fps for detecting and tracking objects (cue and balls) in real-time. Talmo has optimized the cross-platform software for running on entry-level hardware.
Robust and custom table, cue and ball detection algorithms has been developed to support a variety of lighting setups, tablecloths and cue designs.