FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3185118158> ?p ?o ?g. }

• W3185118158 endingPage "112599" @default.
• W3185118158 startingPage "112599" @default.
• W3185118158 abstract "Multi-temporal deep learning approaches have exhibited excellent classification performance in large-scale crop mapping. These approaches efficiently and automatically transform remote sensing time series into high-dimensional feature representations to identify crop types. The lack of interpretation, however, is regarded as a major drawback of these high-performance approaches. Interpreting deep learning approaches in multi-temporal crop mapping is critical for verifying their reliability. This study aims to quantify the impact of multi-temporal information in input time series on classification performance and develop a multi-perspective interpretation pipeline for deep learning models. The pipeline involves three interpretation approaches: evaluating input feature importance, analyzing hidden features, and monitoring temporal changes in model's soft output. An experiment is conducted to classify corn and soybean in the U.S corn belt in 2018. The study area consists of three sites each encompassing millions of pixel-level samples at 30 m resolution. The Landsat Analysis Ready Data are used as the input remote sensing time series and Cropland Data Layer is used as the ground reference. Attention-based Long Short-Term Memory (AtLSTM) and Transformer models are built as multi-temporal deep learning models, and compared to Random Forest (RF). Complete time series input in the correct order achieves a higher overall accuracy of 97.8% than using single-window or out-of-order inputs, indicating multi-temporal information facilitates crop classification. An assessment of the input feature importance demonstrates that the AtLSTM, Transformer, and RF models all consider the period from weeks 11 to 20 (early-July to late-August) as a key growth period and the shortwave infrared band as the critical band for corn and soybean discrimination. Hidden feature analysis suggests that the AtLSTM model accumulates the useful information over the growth period, while the Transformer model extracts the temporal dependencies that contribute important information to high-level feature learning. The learned features contain more effective and refined information than the raw input features and thus are better suited for crop classification. The soft output analysis in the in-season classification scenario demonstrates that increased length of input time series improves the model's confidence in the classification results. The further comparison of input feature importance in different sites and years demonstrates the applicability of the interpretation approach at larger spatiotemporal extents with heterogeneous landscapes and interannual variability. This study provides a multi-perspective evaluation to identify key features in multi-spectral and multi-temporal remote sensing data, and yields a practical approach to integrate agronomy knowledge in deep learning-based crop mapping." @default.
• W3185118158 created "2021-08-02" @default.
• W3185118158 creator A5000165736 @default.
• W3185118158 creator A5032201368 @default.
• W3185118158 creator A5045675946 @default.
• W3185118158 creator A5050008056 @default.
• W3185118158 creator A5050209085 @default.
• W3185118158 creator A5056050667 @default.
• W3185118158 creator A5072139431 @default.
• W3185118158 creator A5083868441 @default.
• W3185118158 date "2021-10-01" @default.
• W3185118158 modified "2023-10-14" @default.
• W3185118158 title "Towards interpreting multi-temporal deep learning models in crop mapping" @default.
• W3185118158 cites W1154758367 @default.
• W3185118158 cites W1982487747 @default.
• W3185118158 cites W1985514099 @default.
• W3185118158 cites W1991361881 @default.
• W3185118158 cites W1998281138 @default.
• W3185118158 cites W2008085934 @default.
• W3185118158 cites W2011500029 @default.
• W3185118158 cites W2027866306 @default.
• W3185118158 cites W2064675550 @default.
• W3185118158 cites W2081054451 @default.
• W3185118158 cites W2118037698 @default.
• W3185118158 cites W2150355110 @default.
• W3185118158 cites W2155632266 @default.
• W3185118158 cites W2261059368 @default.
• W3185118158 cites W2290326488 @default.
• W3185118158 cites W239666967 @default.
• W3185118158 cites W2438450043 @default.
• W3185118158 cites W2531168480 @default.
• W3185118158 cites W2548875948 @default.
• W3185118158 cites W2578830027 @default.
• W3185118158 cites W2604086375 @default.
• W3185118158 cites W2606707877 @default.
• W3185118158 cites W2621021710 @default.
• W3185118158 cites W2730238284 @default.
• W3185118158 cites W2791592925 @default.
• W3185118158 cites W2793461576 @default.
• W3185118158 cites W2793935644 @default.
• W3185118158 cites W2803946774 @default.
• W3185118158 cites W2885406917 @default.
• W3185118158 cites W2886493749 @default.
• W3185118158 cites W2888391613 @default.
• W3185118158 cites W2903282641 @default.
• W3185118158 cites W2919115771 @default.
• W3185118158 cites W2954994501 @default.
• W3185118158 cites W2960863152 @default.
• W3185118158 cites W2973353633 @default.
• W3185118158 cites W2979544394 @default.
• W3185118158 cites W2981830988 @default.
• W3185118158 cites W2990480734 @default.
• W3185118158 cites W2995568574 @default.
• W3185118158 cites W2997552745 @default.
• W3185118158 cites W2999585470 @default.
• W3185118158 cites W3000309135 @default.
• W3185118158 cites W3009382267 @default.
• W3185118158 cites W3024943905 @default.
• W3185118158 cites W3037002701 @default.
• W3185118158 cites W3047601730 @default.
• W3185118158 cites W3091952054 @default.
• W3185118158 cites W3104839310 @default.
• W3185118158 doi "https://doi.org/10.1016/j.rse.2021.112599" @default.
• W3185118158 hasPublicationYear "2021" @default.
• W3185118158 type Work @default.
• W3185118158 sameAs 3185118158 @default.
• W3185118158 citedByCount "63" @default.
• W3185118158 countsByYear W31851181582021 @default.
• W3185118158 countsByYear W31851181582022 @default.
• W3185118158 countsByYear W31851181582023 @default.
• W3185118158 crossrefType "journal-article" @default.
• W3185118158 hasAuthorship W3185118158A5000165736 @default.
• W3185118158 hasAuthorship W3185118158A5032201368 @default.
• W3185118158 hasAuthorship W3185118158A5045675946 @default.
• W3185118158 hasAuthorship W3185118158A5050008056 @default.
• W3185118158 hasAuthorship W3185118158A5050209085 @default.
• W3185118158 hasAuthorship W3185118158A5056050667 @default.
• W3185118158 hasAuthorship W3185118158A5072139431 @default.
• W3185118158 hasAuthorship W3185118158A5083868441 @default.
• W3185118158 hasConcept C108583219 @default.
• W3185118158 hasConcept C119666444 @default.
• W3185118158 hasConcept C119857082 @default.
• W3185118158 hasConcept C121332964 @default.
• W3185118158 hasConcept C124101348 @default.
• W3185118158 hasConcept C127313418 @default.
• W3185118158 hasConcept C138885662 @default.
• W3185118158 hasConcept C146849305 @default.
• W3185118158 hasConcept C150899416 @default.
• W3185118158 hasConcept C151406439 @default.
• W3185118158 hasConcept C153180895 @default.
• W3185118158 hasConcept C154945302 @default.
• W3185118158 hasConcept C169258074 @default.
• W3185118158 hasConcept C2776401178 @default.
• W3185118158 hasConcept C41008148 @default.
• W3185118158 hasConcept C41895202 @default.
• W3185118158 hasConcept C62520636 @default.
• W3185118158 hasConcept C62649853 @default.
• W3185118158 hasConcept C77277458 @default.

• next