Whether The Street Is Beautiful? A Novel Evaluation Method Based On Mobile LiDAR Data And Street View

Street is the skeleton of a city, it is an extremely important part. It not only ensures the function of urban traffic and carrying the human business activties, but also play an important role on the whole landscape of the city. At present, the aesthetic evaluation of the road is focused on the theoretical approach. Restricted by the defect of technical means and lack of data, evaluation on road beauty need an complete and quantitative evaluation method. Mobile Light Detection And Ranging System (MLS) running in the streets of the city, it has the unique advantage of providing data of directly measured 3D points of urban objects. It has been widely applied in urban surveying and planning. Street view data, which has wide coverage, huge volume and open access for the public, leads to the inevitable trend of mining the utility value of street view data in urban street studies, and has provided new opportunities for urban street landscape calculation and assessment that primarily focus on human senses. Before using the two data into the field of street beauty evaluation, it is need to address how to get thefeatures and information from the two data acquired from the complex urban environment.To assess urban street beauty degree, the proposed method focus on the space perception index calculation method based on MLS data, the visual perception calculation method based ont Street View and the street beauty degree evaluation method. The specific research contents are as follows:(1) The space perception index calculation method based on MLS data. The street closure degree, three dimensional green volume, landscape diversity index and sunlight index were used as spatial perception index. The quantitative calculation of these indexes needs to quickly and accurately extract the required spatial information from the mass point cloud. First, a new structure, called "voxel group", is applied to further organize voxels based on the voxelization of VLS points. The method will process each voxel group instead of discrete point clouds to accelerate the calculation. Then a simple way is applied to quickly recognize the shape of point clouds in each voxel group. Second, a category-oriented merging is used for merging voxel group by utilizing the shape information to obtain high precision point clouds clusters. At last, a novel characteristic nature of vehicle-borne LiDAR building points, called "horizontal hollow ratio" for efficient extraction of various forms of buildings. The experiment show that, the space perception index calculation method based on MLS data can effectively get the spatial information of the streets, the calculation result of the indexs is consistent with the actual situation, so it can be as a basis of street scenic beauty estimation.(2) The visual perception index calculation method based on Street View. Use of color elements, significant regional characteristics, green visual index, visual entropy, the sky open index to measure the visual perception of people when walking along street. This paper presents a image processing technique flow, include image segmentation and image saliency detection, to extract the useful features from the massive Street View, such as the extraction of the green plant area and sky area of the Street Views. The five visual perception index are calculated, through the experiment found that the visual perception calculation method based on Street View can process the massive Street Views and quantitative calculates the five visual perception index. The color elements significant regional characteristics, green visual index, visual entropy and the sky open index match the actual feelings of people so it can be as a basis for street scenic beauty estimation.The analysis and validation of the final scenic beauty estimation results show that the street scenic beauty evaluation nethod based on MLS Data and Street View is fine.The quantitative calculation result of these elements was fine. The result of the street scenic beauty evaluation is consistent with the actual value, and it has high application value.