A Vertical Line Feature Extraction and Optimization Method Based on Grid Maps

In structured environments, vertical line features are prevalent. However, inconsistent feature extraction can impact map fusion relying on vertical line feature extraction and matching. This paper proposes a method for extracting and optimizing vertical line features based on grid maps. An optimization algorithm is applied to the detected and extracted line features to ensure their consistency across different maps. The proposed method initiates by detecting and extracting edges within the map using the Canny edge detection algorithm. Subsequently, the Hough transform is applied to detect and extract straight-line segments composed of the previously identified edges. And retain the line segments that form the vertical relationship as vertical line features. Then the optimization algorithm proposed in this article evaluates vertical line segments within a specified range. The algorithm chooses the optimal vertical line segment as the optimized result and updates its position information to ensure consistent extraction of line features within the overlapping area of the two sub-maps. Finally, obtain the accurate transformation matrix by performing feature matching based on point feature information corresponding to consistent line features. Experimental results show that this method can effectively extract and optimize vertical line features within the map and lead to a successful fusion of the two sub-maps.