Artificial intelligence, machine learning and data are at the heart of the digital revolution we are currently experiencing. To support business decision-making, one must go beyond predictions and employ causal data science, which enables you to identify the factors that most directly help influence customer decision-making. We explain it in this video!
In line with the global tendency of automation of industrial processes, Alherpell has identified the need to automate the visual inspection of the raw lambskins that today is carried out by the hands of experienced operators and through which it is identify the defects present in the skins to be able to classify them according to the quality. The technical objective of the project is to design and develop the machine that automatically classifies the skins, making operators intervene only in the placement and removal of skins.
To achieve these objectives, Alherpell, with the collaboration of the knowledge transfer group DataScience@UB, has decided to design and prototype a machine that includes an intelligent vision system based on semantic segmentation using convolutional neural networks; being necessary a system of transport of the skins, the mechanical vision system (camera and controller) and the development of intelligent software for detecting defects.
This automation will allow, on the one hand, to increase the productive capacity of the leather warehouses and, on the other, to define criteria of categorization that can be extrapolated to other agents in the value chain, thereby reducing discrepancies and, therefore, the number of transactions not fruitful. This unification of criteria will translate into a reduction of the resources used in the sector and a reduction in the transport of unaccompanied lots and of the respective polluting emissions.
DataScience@UB designs and prototype a machine that includes an intelligent vision system based on semantic segmentation through convolutional neural networks to automate the classification of skins and leather.
We have won a new competition with the Cartographic and Geological Institute of Catalonia, to apply deep learning techniques in the detection of changes in the network of infrastructures and constructions in the territory with the aim of automating the process of elaboration of the Topographic map of Catalonia.