Johannes Hinckeldeyn

Recep Savas, Johannes Hinckeldeyn

Purpose: Object detection is rapidly evolving through machine learning technology in automation systems. Well prepared data is necessary to train the algorithms. Accordingly, the objective of this paper is to describe a re-evaluation of the so-called Logistics Objects in Context (LOCO) dataset, which is the first dataset for object detection in the field of intralogistics. Methodology: We use an experimental research approach with three steps to evaluate the LOCO dataset. Firstly, the images on GitHub were analyzed to understand the dataset better. Secondly, Google Drive Cloud was used for training purposes to revisit the algorithmic implementation and training. Lastly, the LOCO dataset was examined, if it is possible to achieve the same training results in comparison to the original publications. Findings: The mean average precision, a common benchmark in object detection, achieved in our study was 64.54%, and shows a significant increase from the initial study of the LOCO authors, achieving 41%. However, improvement potential is seen specifically within object types of forklifts and pallet truck. Originality: This paper presents the first critical replication study of the LOCO dataset for object detection in intralogistics. It shows that the training with better hyperparameters based on LOCO can even achieve a higher accuracy than presented in the original publication. However, there is also further room for improving the LOCO dataset.

Markus Knitt, Jakob Schyga, Asan Adamanov et al.

Estimating the pose of a pallet and other logistics objects is crucial for various use cases, such as automatized material handling or tracking. Innovations in computer vision, computing power, and machine learning open up new opportunities for device-free localization based on cameras and neural networks. Large image datasets with annotated poses are required for training the network. Manual annotation, especially of 6D poses, is an extremely labor-intensive process. Hence, newer approaches often leverage synthetic training data to automatize the process of generating annotated image datasets. In this work, the generation of synthetic training data for 6D pose estimation of pallets is presented. The data is then used to train the Deep Object Pose Estimation (DOPE) algorithm. The experimental validation of the algorithm proves that the 6D pose estimation of a standardized Euro pallet with a Red-Green-Blue (RGB) camera is feasible. The comparison of the results from three varying datasets under different lighting conditions shows the relevance of an appropriate dataset design to achieve an accurate and robust localization. The quantitative evaluation shows an average position error of less than 20 cm for the preferred dataset. The validated training dataset and a photorealistic model of a Euro pallet are publicly provided.