Satellite Imagery Semantic Segmentation Razieh Kaviani Baghbaderani, - - PowerPoint PPT Presentation

▶

Oct 11, 2023 38 likes •131 views

Incorporating Spectral Unmixing in Satellite Imagery Semantic Segmentation Razieh Kaviani Baghbaderani, Hairong Qi University of Tennessee, Department of EECS Motivation: Deep learning-based networks need a large dataset for training due to a

SLIDE 1

Incorporating Spectral Unmixing in Satellite Imagery Semantic Segmentation

Razieh Kaviani Baghbaderani, Hairong Qi University of Tennessee, Department of EECS

SLIDE 2

Motivation:

▪ Deep learning-based networks need a large dataset for training due to a huge number of hyper-parameters. ▪ Challenges in remote sensing area:

Difficulty in obtaining a large number of labeled training samples due to the high-

resolution and large coverage of satellite images.

The existence of mixed pixels that would result in mixed spectral readings due to

environmental interference as well as the large footprint of the sensing device.

Encoder Decoder

SLIDE 3

Main Idea:

▪ Employing the spectral unmixing method as a feature extraction step to entangle the intertwined reflectances of the given images and obtain the advanced features as extra domain information. ▪ Decompose a mixture into constituent materials (endmembers, A) and their proportions (abundance, S). ▪ Physical constraints (nonnegativity & sum- to-one) ▪ Unsupervised (both A and S are unknown)

vs. Supervised (A is known)

SLIDE 4

Approach:

MVC-NMF FCLS

SLIDE 5

Dataset:

▪ Dataset was obtained from Kaggle Competition: “Dstl Satellite Imagery Feature Detection”

 Imbalanced data Distributions of target classes

0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% Building Structure Track Trees Crops Waterways Standing water

SLIDE 6

Experimental Results:

Test image:

▪ The probability of pixels belonging to each class

SLIDE 7

Experimental Results:

▪ The segmentation accuracy (without any post processing) for different classes in terms of Intersection Over Union (IOU%)

10 20 30 40 50 60 70 Building Structure Track Trees Crops Avg. Baseline [Pan, RGB, MSI] Unmixing-based [Pan, RGB, MSI] Baseline [Pan, RGB, MSI, 4Ratios] Unmixing-based [Pan, RGB, MSI, 4Ratios]

SLIDE 8

Conclusion:

▪ The abundance maps as extra information along with spectral bands enhance the segmentation accuracy and increase the confidence of network in predicting the label for each pixel. ▪ Adding abundance maps to improve the segmentation accuracy is more effective when the training samples are limited

SLIDE 9