Road Green Belt Extraction Algorithm Based on Morphology and Color Index

Taking the highway images captured by UAV aerial photography as the research object, a highway greening information extraction algorithm based on digital image processing morphology and visible light color index is proposed. Firstly, the experimental image is enhanced by HSV image. An improved vegetation color index - red-green-blue difference index (RGBDI) is established by comprehensively using the three optical bands of visible light red, green and blue. Then, the maximum entropy threshold segmentation method is used to binarize the image. The recognition results are modified by using the digital image processing morphology to reconstruct the open operation and the closed operation, and the final experimental results are obtained. The results are compared with the common 6 color indexes: green leaf index (GLI), normalized green-red difference index (NGRDI), normalized green-blue difference index (NGBDI), relative greenness index (GCC), over-green index (ExG) and over-green minus red index (ExG - ExR) extraction effects. The results show that the green belt extraction algorithm combining morphology and ultra-green leaf index (EGLI) can extract the green space information of the study area well. The kappa coefficient reaches 88.0%, which is the best effect.