3D vision combined with deep learning software for automated fruit recognition
Basler, together with Berlin-based Data Spree, has developed a highly accurate and robust real-time solution for fruit sorting. Agricultural products can vary greatly in shape and color which poses a great challenge for traditional image processing methods. The vision solution described here therefore uses 3D data for object recognition.
The images captured by the Basler blaze-101 Time-of-Flight camera are evaluated by the Data Spree software "Deep Learning DS", a software solution based on deep neural networks (deep learning).
Learn in our Use Case "3D Vision Combined with Deep Learning Software for Automated Fruit Recognition" which steps were necessary for the implementation and which advantages Basler's solution offers.