Dense labeling of large remote sensing imagery with convolutional neural networks: a simple and faster alternative to stitching output label maps

05/30/2018
by   Bohao Huang, et al.
1

In this work we consider the application of convolutional neural networks (CNNs) for pixel-wise labeling (a.k.a., semantic segmentation) of remote sensing imagery (e.g., aerial color or hyperspectral imagery). Remote sensing imagery is usually stored in the form of very large images, referred to as "tiles", which are too large to be segmented directly using most CNNs and their associated hardware. As a result, during label inference, smaller sub-images, called "patches", are processed individually and then "stitched" (concatenated) back together to create a tile-sized label map. This approach suffers from computational ineffiency and can result in discontinuities at output boundaries. We propose a simple alternative approach in which the input size of the CNN is dramatically increased only during label inference. This does not avoid stitching altogether, but substantially mitigates its limitations. We evaluate the performance of the proposed approach against a vonventional stitching approach using two popular segmentation CNN models and two large-scale remote sensing imagery datasets. The results suggest that the proposed approach substantially reduces label inference time, while also yielding modest overall label accuracy increases. This approach contributed to our wining entry (overall performance) in the INRIA building labeling competition.

READ FULL TEXT

page 1

page 2

research
04/29/2018

TreeSegNet: Automatically Constructed Tree CNNs for Subdecimeter Aerial Image Segmentation

For the task of subdecimeter aerial imagery segmentation, the fine-grain...
research
07/24/2018

Feature Fusion through Multitask CNN for Large-scale Remote Sensing Image Segmentation

In recent years, Fully Convolutional Networks (FCN) has been widely used...
research
08/08/2019

Apache Spark Accelerated Deep Learning Inference for Large Scale Satellite Image Analytics

The shear volumes of data generated from earth observation and remote se...
research
08/02/2016

Dense semantic labeling of sub-decimeter resolution images with convolutional neural networks

Semantic labeling (or pixel-level land-cover classification) in ultra-hi...
research
03/22/2018

Aligning Across Large Gaps in Time

We present a method of temporally-invariant image registration for outdo...
research
04/28/2019

Non-myopic Planetary Exploration Combining In Situ and Remote Measurements

Remote sensing can provide crucial information for planetary rovers. How...
research
03/08/2021

The Weakly-Labeled Rand Index

Synthetic Aperture Sonar (SAS) surveys produce imagery with large region...

Please sign up or login with your details

Forgot password? Click here to reset