Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Research on Pavement Marking Recognition and Extraction Method
Pavement marking is an important part of urban traffic. Effective automatic identification and extraction of markings can help urban traffic to implement efficient operation. In the pavement marking image, to improve the contrast of the image, the redundant part of the acquired image is preprocessed, the image is denoised by comparing median filter, average filtering and Gaussian filter, then enhanced image processing by using a linear grayscale transform. It is effectively improving the contrast between the object of interest and the background in the marking images, For the extraction of markings, it is interactive extracted based on edge detection operator, threshold segmentation, k mean method and regional growth method. According to optimizing the extraction algorithm after programming, it lays the foundation for the automatic extraction of pavement markings.
Research on Pavement Marking Recognition and Extraction Method
Pavement marking is an important part of urban traffic. Effective automatic identification and extraction of markings can help urban traffic to implement efficient operation. In the pavement marking image, to improve the contrast of the image, the redundant part of the acquired image is preprocessed, the image is denoised by comparing median filter, average filtering and Gaussian filter, then enhanced image processing by using a linear grayscale transform. It is effectively improving the contrast between the object of interest and the background in the marking images, For the extraction of markings, it is interactive extracted based on edge detection operator, threshold segmentation, k mean method and regional growth method. According to optimizing the extraction algorithm after programming, it lays the foundation for the automatic extraction of pavement markings.
Research on Pavement Marking Recognition and Extraction Method
Zhang, Zhihua (Autor:in) / Zhang, Xinxiu (Autor:in) / Yang, Shuwen (Autor:in) / Yang, Jun (Autor:in)
23.07.2021
769801 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch