A Deep Learning Approach to Mapping Irrigation Using Landsat: IrrMapper U-Net

Colligan, Thomas; Ketchum, David; Brinkerhoff, Douglas; Maneta, Marco P.

doi:10.1109/tgrs.2022.3175635

Cited by 11 publications

(9 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Even with high-resolution images, precise interpretation of the true nature of ground cover may not always be possible, potentially impacting our study. There have been several studies using high-resolution images and deep learning methods to extract irrigation [52,53,59]. These studies have achieved an overall accuracy exceeding 85%.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Mapping Irrigated Croplands from Sentinel-2 Images Using Deep Convolutional Neural Networks

Li,

Sun,

Zhou

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

Understanding the spatial distribution of irrigated croplands is crucial for food security and water use. To map land cover classes with high-spatial-resolution images, it is necessary to analyze the semantic information of target objects in addition to the spectral or spatial–spectral information of local pixels. Deep convolutional neural networks (DCNNs) can characterize the semantic features of objects adaptively. This study uses DCNNs to extract irrigated croplands from Sentinel-2 images in the states of Washington and California in the United States. We integrated the DCNNs of 101 layers, discarded pooling layers, and employed dilation convolution to preserve location information; these are models which were used based on fully convolutional network (FCN) architectures. The findings indicated that irrigated croplands may be effectively detected at various phases of crop growth in the fields. A quantitative analysis of the trained models revealed that the three models in the two states had the lowest values of Intersection over Union (IoU) and Kappa, i.e., 0.88 and 0.91, respectively. The deep models’ temporal portability across different years was acceptable. The lowest values of recall and OA (overall accuracy) from 2018 to 2021 were 0.91 and 0.87, respectively. In Washington, the lowest OA value from 10 to 300 m resolution was 0.76. This study demonstrates the potential of FCNs + DCNNs approaches for mapping irrigated croplands across large regions, providing a solution for irrigation mapping. The spatial resolution portability of deep models could be improved further by designing model architectures.

show abstract

Section: Discussionmentioning

confidence: 99%

“…We treated each band as a separate feature without explicit temporal information. The detailed basis for the data treatment can be found in [59].…”

Section: Collection and Pre-processing Of Remote Sensing Datamentioning

confidence: 99%

Mapping Irrigated Croplands from Sentinel-2 Images Using Deep Convolutional Neural Networks

Li,

Sun,

Zhou

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…However, these models have a poor irrigation accuracy, which makes it difficult to use them to real-time use scenarios [38]. Deep learning with UNets, optimized LSTMs, deep reinforcement learning (DRL), and convolutional neural networks (CNN), all of which aim at augmenting different parameter sets for highly efficient irrigation operations, are proposed as a solution to these problems in the research published in [39]- [42], respectively. Because these models have accuracy levels of more than 95%, it is possible to utilize them for real-time deployments.…”

Section: Review Of Existing Smart Irrigation Techniquesmentioning

confidence: 99%

Design of an efficient IoT based irrigation platform via fuzzy bioinspired multisensory analysis of on-field parameter sets

Bhoyar

Kanojia

Chourasia

et al. 2023

tjes

View full text Add to dashboard Cite

Design of smart irrigation systems is a transdisciplinary work involving selecting optimal field-specific sensing devices, data capturing from these devices, their pre-processing, feature analysis, classification, IoT (Internet of Things) based actuation, and feedback operations. A wide variety of smart irrigation models, including IBM ThingSpeak, Tracxn, Farm Connect, Hydrawise, etc. have been proposed by researchers, and each of them have its own operational & deployment-specific characteristics. These models are either highly complex or have slower response time & low efficiency when applied to real-time irrigation scenarios. Moreover, these models have a higher cost, which limits their scalability levels. To overcome these issues, this text discusses design of a novel efficient IoT based irrigation platform via fuzzy bioinspired multisensory analysis of on-field parameter sets. The proposed platform uses low-cost components for sensing moisture levels, temperature levels, rain probability, NPK (Nitrogen, Phosphorous, and Potassium) levels, and soil types. These sensed values are processed via a Grey Wolf Optimizer (GWO) to identify optimum sensor types, and then a fuzzy decision layer is used for irrigation. This layer assists in the identification of efficient water flow for different crop types. The resulting growth of plants is fed-back into the model, and a Genetic Algorithm (GA) is applied to tune the fuzzy rules for better water-flow under multiple types of crops. Due to the integration of GWO with Fuzzy Logic controller, the proposed model improved yield efficiency by 8.5%, reduced computational delay by 4.9%, and reduced deployment cost compared to standard smart irrigation models. Due to these advantages, the proposed model is capable of deployment for a wide variety of real-time smart irrigation use cases.

show abstract

“…The improved U-Net model was named ISDU-Net, and ISDU-Net was compared with the FCN, SegNet, and U-Net models to analyze the qualitative and quantitative results [51][52][53][54][55][56][57][58][59][60][61][62].…”

Section: Comparative Analysis Of Improved U-net and Classical Network...mentioning

confidence: 99%

Economic Fruit Forest Classification Based on Improved U-Net Model in UAV Multispectral Imagery

Jia

Yang

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

Economic fruit forest is an important part of Chinese agriculture with high economic value and ecological benefits. Using UAV multi-spectral images to research the classification of economic fruit forests based on deep learning is of great significance for accurately understanding the distribution and scale of fruit forests and the status quo of national economic fruit forest resources. Based on the multi-spectral remote sensing images of UAV, this paper constructed semantic segmentation data of economic fruit forests, conducted a comparative study on the classification and identification of economic fruit forests of FCN, SegNet, and U-Net classic semantic segmentation models, and proposed an improved ISDU-Net model. The recognition accuracies of the ISDU-Net were 87.73%, 70.68%, 78.69%, and 0.84, respectively, in terms of pixel accuracy, average intersection ratio, frequency weight intersection ratio, and Kappa coefficient, which were 3.19%, 8.90%, and 4.51% higher than the original U-Net model. The results showed that the improved ISDU-Net could effectively improve the learning ability of the model, perform better in the prediction of short sample categories, obtain a higher classification accuracy of fruit forest crops, and provide a new idea for the research on accurate fruit forest identification.

show abstract

A Deep Learning Approach to Mapping Irrigation Using Landsat: IrrMapper U-Net

Cited by 11 publications

References 28 publications

Mapping Irrigated Croplands from Sentinel-2 Images Using Deep Convolutional Neural Networks

Mapping Irrigated Croplands from Sentinel-2 Images Using Deep Convolutional Neural Networks

Design of an efficient IoT based irrigation platform via fuzzy bioinspired multisensory analysis of on-field parameter sets

Economic Fruit Forest Classification Based on Improved U-Net Model in UAV Multispectral Imagery

Contact Info

Product

Resources

About