Related Experiment Video
Updated: Jan 13, 2026
07:05
Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures
Published on: February 15, 2022
2.9K
An Improved Two-Step Strategy for Accurate Feature Extraction in Weak-Texture Environments
Qingjia Lv1, Yang Liu1, Peng Wang1
1School of Mechanical Engineering & Automation, Dalian Polytechnic University, Dalian 116034, China.
Sensors (Basel, Switzerland)
|October 29, 2025
Summary
This study introduces a novel solution for feature extraction and reconstruction in environments with weak textures, crucial for mobile robot perception. The method enhances environmental features and uses a two-step approach for accurate 3D point cloud generation.
Area of Science:
- Robotics
- Computer Vision
- Environmental Perception
Background:
- Feature extraction and reconstruction are challenging in weak-texture environments.
- Mobile robots require robust environmental perception for tasks like navigation and manipulation.
- Existing methods struggle with accuracy and efficiency in feature-poor settings.
Purpose of the Study:
- To propose a solution for feature extraction and reconstruction in weak-texture environments.
- To provide essential data support for mobile robot environmental perception.
- To enhance the capabilities of robots operating in dynamic, low-texture settings.
Main Methods:
- Laser-assisted marking to enhance environmental features.
- A two-step feature extraction strategy using binocular vision.
- An improved SURF algorithm for fast feature point localization (FLM).
- A robust correction method (RCM) using light strip grayscale consistency for precise calibration.
Main Results:
- Generation of a sparse 3D point cloud via feature matching and reconstruction.
- Achieved spatial modeling accuracy of ±0.5 mm at a 1m working distance.
- Demonstrated a relative error of 2‱ and an effective extraction rate exceeding 97%.
- Exhibited strong robustness against interference, ensuring both efficiency and accuracy.
Conclusions:
- The proposed solution effectively supports mobile robots in weak-texture environments.
- Enables precise positioning, object grasping, and posture adjustment for robots.
- Advances environmental perception capabilities for robots in challenging, dynamic settings.
