nir-vana innovation network

The German company consists of a young and interdisciplinary team working on automation and digital solutions in the agritech industry. Their goal is to meet the needs of the livestock industry, helping farmers and employees fully automate the documentation in animal breeding and farming. The company developed an embedded system, based on artificial intelligence (AI), computer vision and image processing. This technology allows the automatic counting of animals, thus facilitating the work of farmers and strengthening their efficiency and competitiveness. Currently, the technology is used in the farming sector to count large groups of pigs during different processes. The technology consists of an innovative camera system that automatically registers the number of animals during activities such as buying, selling or the restocking and relocation of pigs. This automated pig counting avoids errors by 100%. The camera system automatically generates the counts on video, which can also be used to automate the documentation and ensure a comprehensive tracking. Furthermore, the system is also capable of including an option for automatic reporting to the corresponding national authorities. The technology is based on a tracking algorithm which uses object detection and is connected with the hardware and the user interface. The technology is based on AI and uses innovative alghorithms for object detection. These algorithms are updated regularly. By using different labels, the technology can be applied to several animals and objects. Additionally, the technology is able to detect different classes of objects only in one picture. Depending on the use cases, different neuronal networks and algorithms like Faster R-CNN (object detection) or Mask R-CNN (instance segmentation) can be applied. After detecting the objects, they need to be tracked in the videos. Therefore, a highly stable and accurate tracking algorithm has been developed. Accurate counting is only possible through the combination of detecting and tracking. The front-end of the system is user friendly and can be accessed via App or PC. Remote access to the collected data is also possible. The hardware is developed for mobile use and is accessible offline without any internet connection. Moreover, the hardware is CE-certified. It consists of a camera, a steering unit and a computing unit. The SME is looking for suitable partners for the further technological development and commercialisation of the technology. A technological cooperation could be realised under a technical cooperation agreement for the development and improvement of the device. Furthermore the company is willing to sign commercial agreements to introduce the technology to various international markets and promote the technology on an international level. Distribution partnerships could also be considered in this context.