Deep Learning for Satellite Imagery Feature Detection

Vladimir Iglovikov

Vladimir & Sergey. 3rd out of 419 teams.

$20,000 prize

  • Sr. Data Scientist at TrueAccord
  • PhD in Physics at UC Davis
  • San Francisco, USA
  • Kaggle Master
  • Deep Learning Engineer
  • BS in Computer Science
  • Angarsk. Siberia. Russia

Problem description

Evaluation metric

Water classes: sometimes simple solutions work better

NDWI = \frac {Green - NIR} {Green + NIR}
NDWI=GreenNIRGreen+NIRNDWI = \frac {Green - NIR} {Green + NIR}

Neural network pipeline

  • Buildings

  • Misc. man-made structures

  • Roads

  • Track

  • Trees

  • Crops

Network ​architecture

Local boundary effects

Local boundary effects

Problem

Prediction quality decreases at the edges

Solution

Added Cropping2D layer

Global boundary effects

Problem

We need integer number of tiles    => 

Problem

Zero Padding creates artifacts    => 

Solution

Zero Padding

Solution

Reflection Padding

Test time augmentation

Results

  1. Water classes => unsupervised
  2. Car classes => did not predict
  3. Other classes => U-net per class
  4. Input 4 indices + M-band + P-band + RGB
  5. Test time augmentation

Summary

Hardware

Vladimir

Core i7

RAM 32 Gb

Titan X (Pascal)

Sergey

2x Xeon E5-2670

RAM 128 GB

GTX 1080

Copy of Using Deep Learning for Satellite Imagery Feature Detection Challenge

By Vladimir Iglovikov

Copy of Using Deep Learning for Satellite Imagery Feature Detection Challenge

  • 1,292