 |
Pakur AI-based counting, classification and inspection of palletized packages during goods receipt and inventory using optical methods on mobile devices Incoming goods inspection is still done manually in many SMEs. Automation of these processes optimizes incoming goods, reduces errors along the entire supply chain and creates competitiveness in the market for transport and warehouse logistics. The implementation of this automation, be it through own development or use of existing solutions on the market, is very costly and not feasible for many SMEs. This is where the research project "Pakur" comes in, in order to enable SMEs in the logistics segment to implement (partially) automated data acquisition in incoming goods inspection or inventory. Recent breakthroughs in the field of image processing using neural networks are to be used to develop an easy-to-use, automatic, digital standard solution for identifying and counting packages based on images of the palletized goods. In doing so, the employee is to be supported by an app in order to accelerate the process of receiving and inventorying goods while minimizing potential errors. Here, algorithms are to be developed and neural networks trained that are capable of recognizing the individual elements, such as packages or bags, on a pallet without error, even in heterogeneous environments, analyzing their packing pattern and then deriving the number of elements correspondingly per unit load. This information can then be passed on directly to a possible inventory management system. Errors are thus detected at an early stage and incorrect information in the system is avoided. The focus of the development is on the creation of the algorithms, based on current, innovative research. The transfer into practice is realized by a ready-to-use, open source software library that can be easily used by third parties and an open source demo application for the smartphone. This ensures that third parties can also actively use the result and apply it to other areas. Contact persons: N. Jathe   (Project manager)B. Staar Funded by: BMWi / AiF Duration: 01.02.2023 - 31.05.2025 See project's publications List all projects |
