Selected article for: "data type and time series"
Author: Kerner, Hannah; Sahajpal, Ritvik; Skakun, Sergii; Becker-Reshef, Inbal; Barker, Brian; Hosseini, Mehdi; Puricelli, Estefania; Gray, Patrick
Title: Resilient In-Season Crop Type Classification in Multispectral Satellite Observations using Growth Stage Normalization
  • Cord-id: qppgfs81
  • Document date: 2020_9_21
    • ID: qppgfs81
    Snippet: Crop type classification using satellite observations is an important tool for providing insights about planted area and enabling estimates of crop condition and yield, especially within the growing season when uncertainties around these quantities are highest. As the climate changes and extreme weather events become more frequent, these methods must be resilient to changes in domain shifts that may occur, for example, due to shifts in planting timelines. In this work, we present an approach for
    Document: Crop type classification using satellite observations is an important tool for providing insights about planted area and enabling estimates of crop condition and yield, especially within the growing season when uncertainties around these quantities are highest. As the climate changes and extreme weather events become more frequent, these methods must be resilient to changes in domain shifts that may occur, for example, due to shifts in planting timelines. In this work, we present an approach for within-season crop type classification using moderate spatial resolution (30 m) satellite data that addresses domain shift related to planting timelines by normalizing inputs by crop growth stage. We use a neural network leveraging both convolutional and recurrent layers to predict if a pixel contains corn, soybeans, or another crop or land cover type. We evaluated this method for the 2019 growing season in the midwestern US, during which planting was delayed by as much as 1-2 months due to extreme weather that caused record flooding. We show that our approach using growth stage-normalized time series outperforms fixed-date time series, and achieves overall classification accuracy of 85.4% prior to harvest (September-November) and 82.8% by mid-season (July-September).

    Search related documents:
    Co phrase search for related documents
    • accuracy drop and long short term memory: 1
    • accuracy drop and low accuracy: 1
    • accuracy drop and lstm model: 1
    • accuracy mid and low accuracy: 1
    • accuracy precision and activation function: 1, 2, 3
    • accuracy precision and adam optimizer: 1, 2, 3, 4
    • accuracy precision and long short term: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12
    • accuracy precision and long short term memory: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12
    • accuracy precision and low accuracy: 1, 2, 3, 4, 5, 6, 7, 8, 9
    • accuracy precision and low negative: 1, 2
    • accuracy precision and lstm model: 1, 2, 3, 4, 5
    • accuracy precision and lstm network: 1, 2, 3, 4, 5, 6
    • accuracy recall and activation function: 1, 2, 3
    • accuracy recall and adam optimizer: 1, 2
    • accuracy recall and long short term: 1, 2, 3, 4, 5, 6, 7, 8
    • accuracy recall and long short term memory: 1, 2, 3, 4, 5, 6, 7, 8
    • accuracy recall and low accuracy: 1, 2, 3
    • accuracy recall and low negative: 1
    • accuracy recall and lstm model: 1, 2, 3, 4, 5