Retail system automation with machine learning

  • Author(s) / Creator(s)
  • Lately, technological advancement has made it possible with the accessibility of enormous annotated datasets and artificial intelligence breakthroughs to have sparked a spectacular rise of precise object recognition and analysis. This thesis specifies the development and implementation considerations of an AI-enabled deep learning-based object detection system for the grocery retail industry. This thesis aims to automate the retail experience of fruits and vegetables. In this work, a data collection of images with appropriate pixel segmentation and bounding boxes has been compiled, the relevant theory is described, and we introduce the dataset of fruits images with the experiment results for training a neural network model. The thesis demonstrates the challenges for such a solution, a recommendation for the object detection model to be used, and future work reference.

  • Date created
    2021-09-01
  • Subjects / Keywords
  • Type of Item
    Research Material
  • DOI
    https://doi.org/10.7939/r3-8jpb-3h53
  • License
    Attribution-NonCommercial 4.0 International