Browsing by Author "Walsh, Stephen J."

Now showing 1 - 2 of 2
  • Thumbnail Image
    Assessing Deep Convolutional Neural Networks and Assisted Machine Perception for Urban Mapping
    Shao, Yang; Cooner, Austin J.; Walsh, Stephen J. (MDPI, 2021-04-15)
    High-spatial-resolution satellite imagery has been widely applied for detailed urban mapping. Recently, deep convolutional neural networks (DCNNs) have shown promise in certain remote sensing applications, but they are still relatively new techniques for general urban mapping. This study examines the use of two DCNNs (U-Net and VGG16) to provide an automatic schema to support high-resolution mapping of buildings, road/open built-up, and vegetation cover. Using WorldView-2 imagery as input, we first applied an established OBIA method to characterize major urban land cover classes. An OBIA-derived urban map was then divided into a training and testing region to evaluate the DCNNs’ performance. For U-Net mapping, we were particularly interested in how sample size or the number of image tiles affect mapping accuracy. U-Net generated cross-validation accuracies ranging from 40.5 to 95.2% for training sample sizes from 32 to 4096 image tiles (each tile was 256 by 256 pixels). A per-pixel accuracy assessment led to 87.8 percent overall accuracy for the testing region, suggesting U-Net’s good generalization capabilities. For the VGG16 mapping, we proposed an object-based framing paradigm that retains spatial information and assists machine perception through Gaussian blurring. Gaussian blurring was used as a pre-processing step to enhance the contrast between objects of interest and background (contextual) information. Combined with the pre-trained VGG16 and transfer learning, this analytical approach generated a 77.3 percent overall accuracy for per-object assessment. The mapping accuracy could be further improved given more robust segmentation algorithms and better quantity/quality of training samples. Our study shows significant promise for DCNN implementation for urban mapping and our approach can transfer to a number of other remote sensing applications.
  • Thumbnail Image
    Stream fish colonization but not persistence varies regionally across a large North American river basin
    Wheeler, Kit; Wenger, Seth J.; Walsh, Stephen J.; Martin, Zachary P.; Jelks, Howard L.; Freeman, Mary C. (2018-07)
    Many species have distributions that span distinctly different physiographic regions, and effective conservation of such taxa will require a full accounting of all factors that potentially influence populations. Ecologists recognize effects of physiographic differences in topography, geology and climate on local habitat configurations, and thus the relevance of landscape heterogeneity to species distributions and abundances. However, research is lacking that examines how physiography affects the processes underlying metapopulation dynamics. We used data describing occupancy dynamics of stream fishes to evaluate evidence that physiography influences rates at which individual taxa persist in or colonize stream reaches under different flow conditions. Using periodic survey data from a stream fish assemblage in a large river basin that encompasses multiple physiographic regions, we fit multi-species dynamic occupancy models. Our modeling results suggested that stream fish colonization but not persistence was strongly governed by physiography, with estimated colonization rates considerably higher in Coastal Plain streams than in Piedmont and Blue Ridge systems. Like colonization, persistence was positively related to an index of stream flow magnitude, but the relationship between flow and persistence did not depend on physiography. Understanding the relative importance of colonization and persistence, and how one or both processes may change across the landscape, is critical information for the conservation of broadly distributed taxa, and conservation strategies explicitly accounting for spatial variation in these processes are likely to be more successful for such taxa.