SRPNet

SRPNet is a web application for digital image processing based on a convolutional deep neural network using the technique of pixel-wise semantic segmentation. The research behind the SRPNet was conducted by the Soil Water Processes group located within the Department of Agronomy at Kansas State University. Our goal is to provide a tool to characterize the land cover of agricutlural fields to help conservation planners, farmers, ranchers, forest producers, and researchers making better decisions to improve soil health and conserve soil and water resources. Accurately quantifying the proportion of each component covering the soil surface is an essential step towards implementing improved management strategies for building soil health and improving soil and water conservation.