Tree, Shrub, and Grass Classification Using Only RGB Images

Journal Article · Thu Apr 23 00:00:00 EDT 2020 · Remote Sensing

DOI:https://doi.org/10.3390/rs12081333· OSTI ID:1668666

Ayhan, Bulent ^[1]; Kwan, Chiman ^[2]

---

1. Applied Research LLC, Rockville, MD (United States); Applied Research LLC
2. Applied Research LLC, Rockville, MD (United States)

In this work, a semantic segmentation-based deep learning method, DeepLabV3+, is applied to classify three vegetation land covers, which are tree, shrub, and grass using only three band color (RGB) images. DeepLabV3+’s detection performance has been studied on low and high resolution datasets that both contain tree, shrub, and grass and some other land cover types. The two datasets are heavily imbalanced where shrub pixels are much fewer than tree and grass pixels. A simple weighting strategy known as median frequency weighting was incorporated into DeepLabV3+ to mitigate the data imbalance issue, which originally used uniform weights. The tree, shrub, grass classification performances are compared when all land cover types are included in the classification and also when classification is limited to the three vegetation classes with both uniform and median frequency weights. Among the three vegetation types, shrub is found to be the most challenging one to classify correctly whereas correct classification accuracy was highest for tree. It is observed that even though the median frequency weighting did not improve the overall accuracy, it resulted in better classification accuracy for the underrepresented classes such as shrub in our case and it also significantly increased the average class accuracy. The classification performance and computation time comparison of DeepLabV3+ with two other pixel-based classification methods on sampled pixels of the three vegetation classes showed that DeepLabV3+ achieves significantly higher accuracy than these methods with a trade-off for longer model training time.

• Research Organization: Applied Research LLC, Rockville, MD (United States)
• Sponsoring Organization: USDOE Office of Science (SC)
• Grant/Contract Number: SC0019936
• OSTI ID: 1668666
• Journal Information: Remote Sensing, Journal Name: Remote Sensing Journal Issue: 8 Vol. 12; ISSN 2072-4292
• Publisher: MDPI
• Country of Publication: United States
• Language: English

Related Subjects

58 GEOSCIENCES
DeepLabV3+
deep learning
imbalanced data
median frequency weighting
vegetation classification