About

Anthony Filippi is an Associate Professor and Director of Graduate Programs at Texas A&M University College of Arts and Sciences, within the Department of Geography. His research specializes in remote sensing and geographic information-processing (GIP), with principal interests in imaging spectroscopy, hyperspectral optical remote sensing of rivers and the coastal ocean, GIS-based modeling, spatial analysis, and data fusion. His work combines remote sensing, aquatic optics, GIScience, and machine learning to study riverine, floodplain, coastal marine, and terrestrial optical systems. Dr. Filippi's research includes developing hyperspectral remote-sensing inversion algorithms to estimate water-column properties, bathymetry, bottom optical properties, and bottom-type information from remote-sensor images over coastal and other waters. He also investigates coastal wetland mapping, floodplain environments, terrestrial land-cover, vegetation, agricultural studies, and hazardous/radiological waste site monitoring using airborne and satellite remote sensing. As a faculty member of the Fluvial-GEOS Lab, he focuses on using remote sensing and GIS to address riverine and floodplain research issues.

Research topics

• Computer Science
• Artificial Intelligence
• Geography
• Ecology
• Cartography
• Remote sensing
• Environmental science
• Agronomy
• Systems engineering
• Risk analysis (engineering)
• Engineering
• Biology
• Agroforestry
• Business
• Computer vision
• Forestry

Selected publications

• Forest restoration for environment and well-being is associated with empowered local governance over long time horizons

  Environmental Research Letters · 2025-07-18 · 1 citations

  articleOpen access

  Abstract Forest restoration is widely recognized as a global priority to sequester carbon, conserve biodiversity, and support the livelihoods of rural and indigenous people. Contemporary interventions often target landscapes with a substantial human presence, and they regularly call for stakeholder participation during project implementation. However, there is a lack of empirical evidence linking local involvement with multiple forest benefits over long time horizons. Using a unique dataset of four decades of government-sponsored tree planting in North India, we find that both substantive local influence over planning projects and sustained control over management into the present—a favorable combination of long-term, empowered local governance—is associated with greater livelihood benefits and improvements in forest canopy cover over time. Our work points toward complex socio-ecological relationships, which may be explained by a positive interaction between empowered local governance, interventions that align with local needs, and long-term local care for planted forests. This implies that current financial commitments may need to be accompanied by institutional reforms that give communities meaningful control over planning and build capacities for self-governance that can endure into the future. In light of this work, we suggest that a paradigm of ‘people-centered restoration’ may offer the best opportunity to support long-term environmental goals in densely settled landscapes in the Global South.

  PublisherOA PDFDOI

• Object- and Image Endmember-based Riparian Forest Classification of Narrow-Band UAS Image Data: A Case Study of the River Gail and River Drau, Austria

  2024-03-11

  preprintOpen access1st author

  Studies that directly compare classification accuracies of object-based image analysis (GEOBIA) and endmember-based algorithms for the exploitation of very-high-spatial-resolution (VHR) images (e.g., unmanned aircraft systems (UAS) images) are quite limited. We employ an endmember-extraction algorithm in conjunction with an endmember-mapping method, and we separately utilize a multiresolution segmentation/object-based classification algorithm. We then classify riparian forest and other land covers and compare the classification accuracies obtained from the application of these respective classifiers to narrow-band, VHR UAS images acquired over two river reaches (of the River Gail and River Drau, respectively) in Austria. We determine the effect of pixel size on classification accuracy and assess performances associated with multiple image-acquisition dates. Our results indicate markedly higher classification accuracies for the GEOBIA approach, relative to those of the endmember-based method, where the former generally entails overall accuracies in excess of 85%. Poor endmember-mapping classification accuracies are most likely a function of: the very small pixel sizes associated with the UAS images; the large number of information classes; and the relatively small number of (albeit narrow) bands available for analysis.

  PublisherDOI

• Enhancing Flood Resilience: Geomorphological Insights into Lowland Riverscapes for Nature-Based Solutions

  2024-03-08

  preprintOpen accessSenior author

  Aimed at achieving environmentally and economically smart growth in lowland riverscapes in the face of exacerbating flood threats, the elements of natural riverscapes, such as floodplain landforms, riparian forests, and wetlands can provide solutions to flood risk reduction. Geomorphological knowledge is crucial to working effectively with river processes and landforms in addressing flood hazards. In addition to unique landforms and habitats that can support flood mitigation, landscape-level geomorphological characteristics, such as geomorphological heterogeneity and connectivity, can also impact the attenuation and retention of downstream fluxes of water, sediment, and other materials, and thus resistance and resilience to floods. In this study, we employ a geomorphological approach to delineate the natural elements of lowland riverscapes as geomorphological habitats to assess their susceptibility to floods and erosion/sedimentation as well as their capacity to alleviate the negative impacts of floods. To delineate geomorphological habitats, we utilize a range of classification approaches and geospatial data including LiDAR-derived digital terrain models, airborne and satellite images, raster/vector data on vegetation, soils, and land-cover land-use. We then quantify the diversity, heterogeneity, and connectivity of delineated habitats using landscape ecological approaches and in the context of flood impacts and mitigation. Our geomorphological approach to riverscape characterization provides new insights on fundamental knowledge of natural elements as geomorphological habitats and their interconnections and interdependencies. This new knowledge has a high potential for developing geomorphologically derived nature-based solutions to flood management and enhancing flood resilience of lowland riverscapes.

  PublisherDOI

• Modelling Red–Crowned Parrot (Psittaciformes: Amazona viridigenalis [Cassin, 1853]) distributions in the Rio Grande Valley of Texas using elevation and vegetation indices and their derivatives

  PLoS ONE · 2023-12-06 · 2 citations

  articleOpen accessCorresponding

  Texas Rio Grande Valley Red-crowned Parrots (Psittaciformes: Amazona viridigenalis [Cassin, 1853]) primarily occupy vegetated urban rather than natural areas. We investigated the utility of raw vegetation indices and their derivatives as well as elevation in modelling the Red-crowned parrot's general use, nest site, and roost site habitat distributions. A feature selection algorithm was employed to create and select an ensemble of fine-scale, top-ranked MaxEnt models from optimally-sized, decorrelated subsets of four to seven of 199 potential variables. Variables were ranked post hoc by frequency of appearance and mean permutation importance in top-ranked models. Our ensemble models accurately predicted the three distributions of interest ([Formula: see text] Area Under the Curve [AUC] = 0.904-0.969). Top-ranked variables for different habitat distribution models included: (a) general use-percent cover of preferred ranges of entropy texture of Normalized Difference Vegetation Index (NDVI) values, entropy and contrast textures of NDVI, and elevation; (b) nest site-entropy textures of NDVI and Green-Blue NDVI, and percent cover of preferred range of entropy texture of NDVI values; (c) roost site-percent cover of preferred ranges of entropy texture of NDVI values, contrast texture of NDVI, and entropy texture of Green-Red Normalized Difference Index. Texas Rio Grande Valley Red-crowned Parrot presence was associated with urban areas with high heterogeneity and randomness in the distribution of vegetation and/or its characteristics (e.g., arrangement, type, structure). Maintaining existing preferred vegetation types and incorporating them into new developments should support the persistence of Red-crowned Parrots in southern Texas.

  PublisherOA PDFDOI

• Comparison of Image Endmember- and Object-Based Classification of Very-High-Spatial-Resolution Unmanned Aircraft System (UAS) Narrow-Band Images for Mapping Riparian Forests and Other Land Covers

  Land · 2022 · 11 citations

  1st authorCorresponding
  • Computer Science
  • Artificial Intelligence
  • Remote sensing

  Riparian forests are critical for carbon storage, biodiversity, and river water quality. There has been an increasing use of very-high-spatial-resolution (VHR) unmanned aircraft systems (UAS)-based remote sensing for riparian forest mapping. However, for improved riparian forest/zone monitoring, restoration, and management, an enhanced understanding of the accuracy of different classification methods for mapping riparian forests and other land covers at high thematic resolution is necessary. Research that compares classification efficacies of endmember- and object-based methods applied to VHR (e.g., UAS) images is limited. Using the Sequential Maximum Angle Convex Cone (SMACC) endmember extraction algorithm (EEA) jointly with the Spectral Angle Mapper (SAM) classifier, and a separate multiresolution segmentation/object-based classification method, we map riparian forests/land covers and compare the classification accuracies accrued via the application of these two approaches to narrow-band, VHR UAS orthoimages collected over two river reaches/riparian areas in Austria. We assess the effect of pixel size on classification accuracy, with 7 and 20 cm pixels, and evaluate performance across multiple dates. Our findings show that the object-based classification accuracies are markedly higher than those of the endmember-based approach, where the former generally have overall accuracies of >85%. Poor endmember-based classification accuracies are likely due to the very small pixel sizes, as well as the large number of classes, and the relatively small number of bands used. Object-based classification in this context provides for effective riparian forest/zone monitoring and management.

  PublisherOA PDFDOI

• Limited effects of tree planting on forest canopy cover and rural livelihoods in Northern India

  Nature Sustainability · 2021 · 159 citations

  • Geography
  • Agroforestry
  • Forestry
  PublisherDOI

• Decades of tree planting in Northern India had little effect on forest density and rural livelihoods

  Research Square · 2021-03-19 · 3 citations

  preprintOpen access

  Abstract Myriad scholars, policymakers, and practitioners advocate tree planting as a climate mitigation strategy and to support local livelihoods. But, is the broad appeal of tree planting supported by evidence? We report estimated impacts from decades of tree planting in Northern India. We find that tree plantings have not, on average, increased the proportion of dense forest cover, and have modestly shifted species composition away from the broadleaf varieties valued by local people. Supplementary analysis from household livelihood surveys show that, in contrast to narratives of forest dependent people being supported by tree planting, there are few direct users of these plantations and their dependence is low. We conclude that decades of expensive tree planting programs have not proved effective.

  PublisherOA PDFDOI

• Computationally efficient sequential feature extraction for single hyperspectral remote sensing image classification

  Abstracts of the ICA · 2021-12-13 · 1 citations

  articleOpen accessSenior author
  PublisherDOI

• Fast Sequential Feature Extraction for Recurrent Neural Network-Based Hyperspectral Image Classification

  IEEE Transactions on Geoscience and Remote Sensing · 2020 · 23 citations

  • Computer Science
  • Artificial Intelligence
  • Computer Science

  Classification is a critical, widely employed type of hyperspectral image (HSI) processing. Recently, deep learning models have been attracting more attention within the hyperspectral remote-sensing community due to their improved classification performance. Among them, recurrent neural networks (RNNs), which were initially used to handle sequential data, have been applied to HSI classification with promising results. The key point for such RNN-based models in a classification context is the extraction of a sequential feature for each individual pixel in a HSI. One popular strategy is to first extract similar pixels compared with a target pixel from the HSI, and then use those similar pixels to encode its sequential feature. However, the computational cost is tremendous, especially if such similarity-calculation search is done on the whole image. In this article, inspired by our previous work regarding similarity measurement-based sequential feature construction, a faster sequential feature extraction framework for long short-term memory (LSTM)-based HSI classification is proposed, where object-based segmentation method is employed for the purpose of imposing spatial constraints and computational acceleration. Within the proposed framework, both the local segment containing the target pixel and nonlocal segments are considered. For a target pixel, similar segments are selected first based on segment-based features, and then similar pixels from selected segments are extracted to construct a sequential feature. During pixel-wise similarity measurement, both spectral and spatial information are considered in such computation. Experimental results on three benchmark HSI data sets illustrate that the proposed methods achieve promising classification performance with markedly lower computation-time cost.

  PublisherDOI

• Hyperspectral Image Classification via Object-Oriented Segmentation-Based Sequential Feature Extraction and Recurrent Neural Network

  2020-09-26 · 7 citations

  articleSenior author

  Recurrent neural networks (RNNs) have been investigated and utilized as classification model in the hyperspectral remote-sensing community due to its great capability of encoding sequential features, especially for multi-temporal images. For non-temporal, individual remote-sensing images, RNNs are still a dominant and powerful classification tool that benefits from sequential feature extraction from a single image. In this article, we propose a computationally-efficient sequential feature extraction method for the long short-term memory (LSTM)-based hyperspectral image classification model. Within the proposed method, object-oriented segmentation was employed first to guide similar-pixel searching in the whole-image scope to a local segment scope. Experimental results on two benchmark hyperspectral datasets indicate that our proposed methods achieve higher classification accuracy with lower computational cost.

  PublisherDOI

Frequent coauthors

• Burak Güneralp

  Texas A&M University

  58 shared

• Andong Ma

  Zhujiang Hospital

  57 shared

• Forrest Fleischman

  52 shared

• Eric A. Coleman

  51 shared

• Bill Schultz

  51 shared

• Rajesh K. Rana

  51 shared

• Vijay Ramprasad

  51 shared

• Harry W. Fischer

  51 shared