Deep learning in remote sensing scene classification: a data augmentation enhanced convolutional neural network framework

Yu, Xingrui; Wu, Xiaomin; Luo, Chunbo; Ren, Peng

doi:10.1080/15481603.2017.1323377

Cited by 236 publications

(103 citation statements)

References 28 publications

Supporting

Mentioning

103

Contrasting

Order By: Relevance

“…Given the unbalanced sampling among the three semantic classes, we incorporated sample weighting, with weights calculated as the inverse frequency of each class, to enhance the robustness of our model. To improve the accuracy of the network, we also augmented the training data by randomly shifting, rotating and reflecting it to create multiple versions of the data [45,46]. With a fully specified model, we trained it using mini-batch stochastic gradient descent with momentum (SGDM) as the optimizer with 75% of the labeled data and 15% of the data for validation.…”

Section: Deep Learning Semantic Segmentation and Model Fittingmentioning

confidence: 99%

A Deep Learning Semantic Segmentation-Based Approach for Field-Level Sorghum Panicle Counting

Malambo

Popescu

et al. 2019

Remote Sensing

View full text Add to dashboard Cite

Small unmanned aerial systems (UAS) have emerged as high-throughput platforms for the collection of high-resolution image data over large crop fields to support precision agriculture and plant breeding research. At the same time, the improved efficiency in image capture is leading to massive datasets, which pose analysis challenges in providing needed phenotypic data. To complement these high-throughput platforms, there is an increasing need in crop improvement to develop robust image analysis methods to analyze large amount of image data. Analysis approaches based on deep learning models are currently the most promising and show unparalleled performance in analyzing large image datasets. This study developed and applied an image analysis approach based on a SegNet deep learning semantic segmentation model to estimate sorghum panicles counts, which are critical phenotypic data in sorghum crop improvement, from UAS images over selected sorghum experimental plots. The SegNet model was trained to semantically segment UAS images into sorghum panicles, foliage and the exposed ground using 462, 250 × 250 labeled images, which was then applied to field orthomosaic to generate a field-level semantic segmentation. Individual panicle locations were obtained after post-processing the segmentation output to remove small objects and split merged panicles. A comparison between model panicle count estimates and manually digitized panicle locations in 60 randomly selected plots showed an overall detection accuracy of 94%. A per-plot panicle count comparison also showed high agreement between estimated and reference panicle counts (Spearman correlation ρ = 0.88, mean bias = 0.65). Misclassifications of panicles during the semantic segmentation step and mosaicking errors in the field orthomosaic contributed mainly to panicle detection errors. Overall, the approach based on deep learning semantic segmentation showed good promise and with a larger labeled dataset and extensive hyper-parameter tuning, should provide even more robust and effective characterization of sorghum panicle counts.

show abstract

Section: Deep Learning Semantic Segmentation and Model Fittingmentioning

confidence: 99%

A Deep Learning Semantic Segmentation-Based Approach for Field-Level Sorghum Panicle Counting

Malambo

Popescu

et al. 2019

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…However, data labeled as center pivot systems were much less than data labeled as non-center pivot systems. To overcome the limited number of samples, image transformations, such as flipping, rotating, and cropping have been proposed [44,45]. In this work, a sampling approach was proposed to augment the number of training samples.…”

Section: Training Datamentioning

confidence: 99%

Automatic Identification of Center Pivot Irrigation Systems from Landsat Images Using Convolutional Neural Networks

Zhang

Yue

et al. 2018

Agriculture

View full text Add to dashboard Cite

Being hailed as the greatest mechanical innovation in agriculture since the replacement of draft animals by the tractor, center pivot irrigation systems irrigate crops with a significant reduction in both labor and water needs compared to traditional irrigation methods, such as flood irrigation. In the last few decades, the deployment of center pivot irrigation systems has increased dramatically throughout the United States. Monitoring the installment and operation of the center pivot systems can help: (i) Water resource management agencies to objectively assess water consumption and properly allocate water resources, (ii) Agro-businesses to locate potential customers, and (iii) Researchers to investigate land use change. However, few studies have been carried out on the automatic identification and location of center pivot irrigation systems from satellite images. Growing rapidly in recent years, machine learning techniques have been widely applied on image recognition, and they provide a possible solution for identification of center pivot systems. In this study, a Convolutional Neural Networks (CNNs) approach was proposed for identification of center pivot irrigation systems. CNNs with different structures were constructed and compared for the task. A sampling approach was presented for training data augmentation. The CNN with the best performance and less training time was used in the testing area. A variance-based approach was proposed to further locate the center of each center pivot system. The experiment was applied to a 30-m resolution Landsat image, covering an area of 20,000 km 2 in North Colorado. A precision of 95.85% and a recall of 93.33% of the identification results indicated that the proposed approach performed well in the center pivot irrigation systems identification task.

show abstract

“…The stacked autoencoder is an unsupervised learning model, which trains an one-hidden layer neural network to reconstruct input data from the latent representation [13]. In recent years, neural network models [14] show a great potentials in learning hidden representations. Given a set of data instances…”

Section: A Stacked Autoencodermentioning

confidence: 99%

Federated Learning Based Proactive Content Caching in Edge Computing

Min

et al. 2018

2018 IEEE Global Communications Conference (GLOBECOM)

123

View full text Add to dashboard Cite

Content caching is a promising approach in edge computing to cope with the explosive growth of mobile data on 5G networks, where contents are typically placed on local caches for fast and repetitive data access. Due to the capacity limit of caches, it is essential to predict the popularity of files and cache those popular ones. However, the fluctuated popularity of files makes the prediction a highly challenging task. To tackle this challenge, many recent works propose learning based approaches which gather the users' data centrally for training, but they bring a significant issue: users may not trust the central server and thus hesitate to upload their private data. In order to address this issue, we propose a Federated learning based Proactive Content Caching (FPCC) scheme, which does not require to gather users' data centrally for training. The FPCC is based on a hierarchical architecture in which the server aggregates the users' updates using federated averaging, and each user performs training on its local data using hybrid filtering on stacked autoencoders. The experimental results demonstrate that, without gathering user's private data, our scheme still outperforms other learning-based caching algorithms such as m--greedy and Thompson sampling in terms of cache efficiency.

show abstract

Deep learning in remote sensing scene classification: a data augmentation enhanced convolutional neural network framework

Cited by 236 publications

References 28 publications

A Deep Learning Semantic Segmentation-Based Approach for Field-Level Sorghum Panicle Counting

A Deep Learning Semantic Segmentation-Based Approach for Field-Level Sorghum Panicle Counting

Automatic Identification of Center Pivot Irrigation Systems from Landsat Images Using Convolutional Neural Networks

Federated Learning Based Proactive Content Caching in Edge Computing

Contact Info

Product

Resources

About