Digital Camera-based Pallet Detection System Using YOLOX

Shivangi  Keshri; Shubham  Kumar; Vikas  Singh

Abstract

The automation of pallet recognition and localization is a widely investigated and sought topic, notably for forklift robots and pallet-picking instruments. Pallets are essential in warehouses and retail establishments. However, pallet recognition and pallet localization can be time-consuming and prone to errors, which can result in reduced efficiency and safety hazards. To address this, we develop the Pallet Detection System, a computer vision mechanism that detects and locates the coordinates of pallets in the environment, reducing errors and increasing warehouse efficiency. To achieve this, we are using computer vision algorithm YOLOX which is trained by large number of dataset images of pallets and non-pallet objects to optimize its accuracy in recognizing pallets in real time object. We have used a Python framework to support the pallet detection process and connected external webcams to obtain the location of pallets. This system reduces risk in warehouse, which increases the safety, reduces timecomplexity and optimizes the warehouse operations.

Abstract

Keywords