Talk · Proceedings of International Conference on Artificial Life and Robotics · February 22, 2024
A Comparative Analysis of Object Detection Methods for Robotic Grasping
Nikita Kolin, Elvira Chebotareva
Abstract
The objects grasping is one of the fundamental robotic problems.Accurate and efficient real-time object detection is crucial for successful grasping in robots equipped with monocular vision.Deep machine learning has made significant progress in solving problems of object detection and image segmentation.At the same time, classical computer vision methods do not lose their relevance and can also be used for these tasks.In this research, we conduct a comparative analysis of the effectiveness the YOLOv8-seg neural network model versions for solving the image segmentation problem with classical segmentation methods.The obtained results allowed us to formulate some recommendations on the choice of a particular method for object detection depending on the surrounding environment conditions.
Related publications
- Advanced Preprocessing Technique for Tomato Imagery in Gravimetric Analysis Applied to Robotic HarvestingApplied Sciences · 2024 · Citations: 7 · Shared topic: Advanced Neural Network Applications
- Comparative Analysis of Methods for Improving the Classification Accuracy of Morphologically Similar Pollen Grain Classes on the CPD-1 DatasetSensors · 2026 · Shared topic: Advanced Neural Network Applications
- Comparative evaluation of ResNet18, ResNet34, and ResNet50 models in classifying plant leaf diseasesYassawi Journal of Engineering Science · 2026 · Shared topic: Advanced Neural Network Applications
- SEMANTIC SEGMENTATION OF AGRICULTURAL FIELD ANOMALY PATTERNS IN AERIAL IMAGES USING U-NETHerald of Kazakh-British technical university · 2026 · Shared topic: Advanced Neural Network Applications
