Improving Vegetation Segmentation with Shadow Effects Based on Double Input Networks Using Polarization Images
Fractional vegetation cover (FVC) plays an important role in the study of vegetation growth state, and the key issue is accurately segmenting and extracting green vegetation from the background. However, the shadows generated by natural lights produce extreme illuminance differences in images, which greatly reduces the vegetation extraction accuracy. The polarization information of ground objects is independent of the physical state of the reflectivity of ground objects, and it can eliminate the influence of strong reflections in images to a certain extent, reduce the illuminance differences under extreme sunlight conditions, and help improve the vegetation recognition effect under shadow conditions. To improve the accuracy of vegetation segmentation under shadow conditions, this study introduces polarized reflection information for vegetation and proposes an improved double input semantic segmentation network based on DeepLabv3plus (DIR_DeepLabv3plus). The network extracts low-level features and high-level features at different spatial scales from both light intensity (RGB) images and degree of linear polarization (DoLP) images independently by a deep residual network and ASPP structure, effectively improving the accuracy of vegetation segmentation in shadow situations. The results suggest that the pixel accuracy (PA), mean intersection over union (mIoU) and Kappa coefficients of shadow vegetation segmentation are 97.54%, 88.81% and 84.83%, respectively. Compared with the color index method, the proposed method has a greatly improved extraction accuracy, and compared with the method without polarization information, it has a 1.54% higher mIoU and 3.05% higher Kappa coefficient. This study provides a new idea for vegetation segmentation and improves the accuracy of FVC calculations under shadow conditions
Year of publication: |
[2022]
|
---|---|
Authors: | Yang, Lan ; Chen, Wei ; Bi, Pengshuai ; Tang, Hongzhao ; Zhang, Fengjiao ; Wang, Zhe |
Publisher: |
[S.l.] : SSRN |
Saved in:
freely available
Saved in favorites
Similar items by person
-
Yu, Shi, (2018)
-
Zhang, Fengjiao, (2023)
-
Zhang, Fengjiao, (2023)
- More ...