| 摘要: |
| 绿视率是用于绿色空间感知的直观评价标准，传统研究的绿视率多基于平面影像进行计算，不能完全反映三维空间中人对绿量的主观感受。基于全景影像，提出全景绿视率的概念，通过全景相机获取球面全景照片，将等距圆柱投影转换为等积圆柱投影，利用基于语义分割的卷积神经网络模型，自动识别植被区域面积以实现全景绿视率自动化识别和计量。通过比较5项卷积神经网络模型对绿视率的识别效果，显示出Dilated ResNet-105神经网络模型具有最高的识别准确度。以武汉市武昌区紫阳公园为例，对各级园路和广场的全景绿视率进行计算和分析。将卷积神经网络的识别结果同人工判别结果进行对比研究，结果显示：使用Dilated ResNet-105卷积神经网络对绿植范围识别的平均交并比（mIoU）为62.53%，与人工识别的平均差异为9.17%。全景绿视率自动识别和计算可以为相关研究提供新的思路，实现客观准确、快速便捷的绿视率测量评估。 |
| 关键词: 风景园林 绿视率 全景影像 等积投影转换 深度学习 卷积神经网络 语义分割 |
| DOI：10.14085/j.fjyl.2019.10.0089.06 |
| 分类号:TU986 |
| 基金项目:国家自然科学基金(编号51808245）；中央高校基本科研业务费专项资金(编号2662017QD037） |
|
| Research on Automatic Identification and Measurement of Panoramic Visible Green Index |
|
ZHANG Wei, ZHOU Yuxing, YANG Mengqi
|
|
Huazhong Agricultural University
|
| Abstract: |
| Visible green index is a visual evaluation standard for green space perception. In previous researches, visible green index was usually calculated on the basis of 2D images, which cannot fully reflect the subjective feeling of people on green volume in 3D space. The concept of panoramic visible green index is proposed based on the panoramic photography. Spherical panoramic photos can be obtained with a panoramic camera, and then the equal-distance cylindrical projected images are transformed into equal-area cylindrical projected images. The images are semantically segmented by convolutional neural network (CNN) models to identify the vegetation areas automatically to calculate the panoramic visible green index. Five CNN models are selected and compared. The result shows that the Dilated ResNet-105 neural network model has the highest recognition accuracy. Finally, the Ziyang Park in Wuchang District of Wuhan is taken as a case study to calculate and analyze the panoramic visible green indices of roads and squares in the park. Compared with the traditional manual identification method, Dilated ResNet-105 achieves an average Intersection over Union (IoU) of 62.53% with an average difference of 9.17% for vegetation area identification. The automatic recognition and calculation of the panoramic visible green index can provide new ideas for related researches and achieve an accurate, high-speed and easy-to-use evaluating method for visible green areas. |
| Key words: landscape architecture visible green index panoramic image equal-area projection deep learning convolutional neural network semantic segmentation |
| 引用本文: | 张炜,周昱杏,杨梦琪.全景绿视率自动识别和计算研究[J].风景园林，2019，26（10）：89-94. |
| ZHANG Wei,ZHOU Yuxing,YANG Mengqi.Research on Automatic Identification and Measurement of Panoramic Visible Green Index[J].Landscape Architecture Journal, 2019, 26(10):89-94. |
|
