Review on Convolutional Neural Networks (CNN) in vegetation remote sensing

Kattenborn, Teja; Leitloff, Jens; Schiefer, Felix; Hinz, Stefan

doi:10.1016/j.isprsjprs.2020.12.010

Cited by 875 publications

(569 citation statements)

References 177 publications

Supporting

Mentioning

369

Contrasting

Unclassified

Order By: Relevance

“…For example, if other types of imagery are included such as multispectral images with near-infrared bands even at a lower spatial resolution [77], other important information may be given to help differentiate species, then the 3 cm spatial resolution may not be a necessity. Data (image) spatial resolution, modalities, quality, collection timing or season, their processing and analysis methods and algorithms, and the complexity of the interested objects together determine the final classification performance [22,78].…”

Section: Trade-off Between Spatial Resolution and Coverage For Encroachment Species Detection And Mappingmentioning

confidence: 99%

“…Remote sensing has been used in forest classification, species distribution mapping and invasive species detection for several years [5,17,[21][22][23][24][25][26][27][28][29][30]. With the development of remote sensing technologies, satellites can now provide images in a spatial resolution as high as a few decimeters, e.g., WorldView-4 provides images in 0.31 m spatial resolution.…”

Section: Introductionmentioning

confidence: 99%

“…Among them, object-based classification is a classical and well-recognized approach to classify different vegetation covers with the advantages of quick and accurate classification based on an object's spectral, textural, and structural features and removing the salt-and-pepper effect compared with other classification approaches that are based purely on single pixels [32]. With the advent of ultra-high resolution aerial imagery and high performance computing systems, convolutional neural network (CNN) based deep learning algorithms have started to be investigated more and more in remote sensing based vegetation mapping [22]. Studies can be found in exploring applications of various CNNbased pixelwise classification or object detection models, including forest mapping [33][34][35], dynamic change monitoring [36], and invasive species encroachment assessment [37].…”

Section: Introductionmentioning

confidence: 99%

“…Studies can be found in exploring applications of various CNNbased pixelwise classification or object detection models, including forest mapping [33][34][35], dynamic change monitoring [36], and invasive species encroachment assessment [37]. CNN models for region-based classifications can be trained with information included in rectangular bounding boxes of the whole individual objects [22]. However, delineating the targeted plant from the background and monitoring its progress and increase in size over time is one of the key topicsin remote sensing for woody plant encroachment.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Early Detection of Encroaching Woody Juniperus virginiana and Its Classification in Multi-Species Forest Using UAS Imagery and Semantic Segmentation Algorithms

et al. 2021

View full text Add to dashboard Cite

Woody plant encroachment into grasslands ecosystems causes significantly ecological destruction and economic losses. Effective and efficient management largely benefits from accurate and timely detection of encroaching species at an early development stage. Recent advances in unmanned aircraft systems (UAS) enabled easier access to ultra-high spatial resolution images at a centimeter level, together with the latest machine learning based image segmentation algorithms, making it possible to detect small-sized individuals of target species at early development stage and identify them when mixed with other species. However, few studies have investigated the optimal practical spatial resolution of early encroaching species detection. Hence, we investigated the performance of four popular semantic segmentation algorithms (decision tree, DT; random forest, RF; AlexNet; and ResNet) on a multi-species forest classification case with UAS-collected RGB images in original and down-sampled coarser spatial resolutions. The objective of this study was to explore the optimal segmentation algorithm and spatial resolution for eastern redcedar (Juniperus virginiana, ERC) early detection and its classification within a multi-species forest context. To be specific, firstly, we implemented and compared the performance of the four semantic segmentation algorithms with images in the original spatial resolution (0.694 cm). The highest overall accuracy was 0.918 achieved by ResNet with a mean interaction over union at 85.0%. Secondly, we evaluated the performance of ResNet algorithm with images in down-sampled spatial resolutions (1 cm to 5 cm with 0.5 cm interval). When applied on the down-sampled images, ERC segmentation performance decreased with decreasing spatial resolution, especially for those images coarser than 3 cm spatial resolution. The UAS together with the state-of-the-art semantic segmentation algorithms provides a promising tool for early-stage detection and localization of ERC and the development of effective management strategies for mixed-species forest management.

show abstract

Section: Trade-off Between Spatial Resolution and Coverage For Encroachment Species Detection And Mappingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Early Detection of Encroaching Woody Juniperus virginiana and Its Classification in Multi-Species Forest Using UAS Imagery and Semantic Segmentation Algorithms

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In recent years, CNNs have shown satisfactory results when applied to tree species classification [25]. CNNs have been applied to classify three-dimensional point clouds of trees [26,27], airborne hyperspectral data [28,29], and high-resolution data combined with LiDAR data [30].…”

Section: Introductionmentioning

confidence: 99%

A New Individual Tree Species Recognition Method Based on a Convolutional Neural Network and High-Spatial Resolution Remote Sensing Imagery

2021

View full text Add to dashboard Cite

Tree species surveys are crucial to forest resource management and can provide references for forest protection policy making. The traditional tree species survey in the field is labor-intensive and time-consuming, supporting the practical significance of remote sensing. The availability of high-resolution satellite remote sensing data enable individual tree species (ITS) recognition at low cost. In this study, the potential of the combination of such images and a convolutional neural network (CNN) to recognize ITS was explored. Firstly, individual tree crowns were delineated from a high-spatial resolution WorldView-3 (WV3) image and manually labeled as different tree species. Next, a dataset of the image subsets of the labeled individual tree crowns was built, and several CNN models were trained based on the dataset for ITS recognition. The models were then applied to the WV3 image. The results show that the distribution maps of six ITS offered an overall accuracy of 82.7% and a kappa coefficient of 0.79 based on the modified GoogLeNet, which used the multi-scale convolution kernel to extract features of the tree crown samples and was modified for small-scale samples. The ITS recognition method proposed in this study, with multi-scale individual tree crown delineation, avoids artificial tree crown delineation. Compared with the random forest (RF) and support vector machine (SVM) approaches, this method can automatically extract features and outperform RF and SVM in the classification of six tree species.

show abstract