Mask R-CNN and OBIA Fusion Improves the Segmentation of Scattered Vegetation in Very High-Resolution Optical Sensors

Guirado, Emilio; Blanco‐Sacristán, Javier; Rodríguez‐Caballero, Emilio; Tabik, Siham; Alcaraz‐Segura, Domingo; Martínez‐Valderrama, Jaime; Cabello, Javier

doi:10.3390/s21010320

Cited by 33 publications

(16 citation statements)

References 83 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Vila‐Viçosa et al 2020); vegetation structure (Betbeder et al 2017); or soil moisture, texture, and salinity (Kim et al 2020). Finally, artificial intelligence, in particular deep learning, can boost these tasks through the exploitation of petabytes of remote sensing data, such as in the detection and counting of seedlings or adult crowns (Buters et al 2019; Albuquerque et al 2020), the identification of tree species (Egli & Höpke 2020), and the detection of scattered trees (Brandt et al 2020; Guirado et al 2021) that could serve for postdisturbance regeneration. In summary, the fusion of remote sensing, ecological niche modeling, and artificial intelligence may help to identify the appropriate locations (e.g.…”

Section: Precision Forest Restorationmentioning

confidence: 99%

See 1 more Smart Citation

Precision restoration: a necessary approach to foster forest recovery in the 21st century

Castro

Morales‐Rueda

Reyes

et al. 2021

Restoration Ecology

Self Cite

View full text Add to dashboard Cite

Forest restoration is currently a primary objective in environmental management policies at a global scale, to the extent that impressive initiatives and commitments have been launched to plant billions of trees. However, resources are limited and the success of any restoration effort should be maximized. Thus, restoration programs should seek to guarantee that what is planted today will become an adult tree in the future, a simple fact that, however, usually receives little attention. Here, we advocate for the need to focus restoration efforts on an individual plant level to increase establishment success while reducing negative side effects by using an approach that we term “precision forest restoration” (PFR). The objective of PFR will be to ensure that planted seedlings or sowed seeds will become adult trees with the appropriate landscape configuration to create functional and self‐regulating forest ecosystems while reducing the negative impacts of traditional massive reforestation actions. PFR can take advantage of ecological knowledge together with technologies and methodologies from the landscape scale to the individual‐plant scale, and from the more traditional, low‐tech approaches to the latest high‐tech ones. PFR may be more expensive at the level of individual plants, but will be more cost‐effective in the long term if it allows for the creation of resilient forests able to provide multiple ecosystem services. PFR was not feasible a few years ago due to the high cost and low precision of the available technologies, but it is currently an alternative that might reformulate a wide spectrum of ecosystem restoration activities.

show abstract

Section: Precision Forest Restorationmentioning

confidence: 99%

“…Artificial intelligence may further boost and enhance this task (e.g. Brandt et al 2020; Guirado et al 2021).…”

Section: Precision Forest Restorationmentioning

confidence: 99%

Precision restoration: a necessary approach to foster forest recovery in the 21st century

Castro

Morales‐Rueda

Reyes

et al. 2021

Restoration Ecology

Self Cite

View full text Add to dashboard Cite

show abstract

“…This is among the most difficult challenges in computer vision. This employed the contemporary mask regional-CNN system (regions using convolutional neural networks) in this work that is an extension of the quicker Regional-CNN classification models [ 33 ]. Mask regional-CNN examines the contribution image and generates a three-outputs for every item class: (i) a category label indicating the object-class description, (ii) a boundary box delimiting every object class, and (iii) a mask delimiting the pixels which comprise every object class.…”

Section: Methodsmentioning

confidence: 99%

An Efficient Deep Learning Mechanism for the Recognition of Olive Trees in Jouf Region

Alshammari

Shahin²

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Olive trees grow all over the world in reasonably moderate and dry climates, making them fortunate and medicinal. Pesticides are required to improve crop quality and productivity. Olive trees have had important cultural and economic significance since the early pre-Roman era. In 2019, Al-Jouf region in a Kingdom of Saudi Arabia’s north achieved global prominence by breaking a Guinness World Record for having more number of olive trees in a world. Unmanned aerial systems (UAS) were increasingly being used in aerial sensing activities. However, sensing data must be processed further before it can be used. This processing necessitates a huge amount of computational power as well as the time until transmission. Accurately measuring the biovolume of trees is an initial step in monitoring their effectiveness in olive output and health. To overcome these issues, we initially formed a large scale of olive database for deep learning technology and applications. The collection comprises 250 RGB photos captured throughout Al-Jouf, KSA. This paper employs among the greatest efficient deep learning occurrence segmentation techniques (Mask Regional-CNN) with photos from unmanned aerial vehicles (UAVs) to calculate the biovolume of single olive trees. Then, using satellite imagery, we present an actual deep learning method (SwinTU-net) for identifying and counting of olive trees. SwinTU-net is a U-net-like network that includes encoding, decoding, and skipping links. SwinTU-net’s essential unit for learning locally and globally semantic features is the Swin Transformer blocks. Then, we tested the method on photos with several wavelength channels (red, greenish, blues, and infrared region) and vegetation indexes (NDVI and GNDVI). The effectiveness of RGB images is evaluated at the two spatial rulings: 3 cm /pixel and 13 cm /pixel, whereas NDVI and GNDV images have only been evaluated at 13 cm/pixel. As a result of integrating all datasets of GNDVI and NDVI, all generated mask regional-CNN-based systems performed well in segmenting tree crowns (F1-measure from 95.0 to 98.0 percent). Based on ground truth readings in a group of trees, a calculated biovolume was 82 percent accurate. These findings support all usage of NDVI and GNDVI spectrum indices in UAV pictures to accurately estimate the biovolume of distributed trees including olive trees.

show abstract

“…Object-based image analysis (OBIA) has been widely used in the past two decades, such as in urban-land-cover mapping [1][2][3], ecological monitoring [4][5][6], disaster evaluation [7][8][9], and crop-type identification [10][11][12]. In recent years, OBIA has also been incorporated into the deep learning model to capture the precise outlines of ground objects and model spatial-context relationships [13][14][15]. The most intriguing feature of OBIA is that homogeneous pixels are merged into one segment, and the segment serves as the smallest unit for image analysis [16,17].…”

Section: Introductionmentioning

confidence: 99%

Improvement of Region-Merging Image Segmentation Accuracy Using Multiple Merging Criteria

et al. 2021

View full text Add to dashboard Cite

Image segmentation plays a significant role in remote sensing image processing. Among numerous segmentation algorithms, the region-merging segmentation algorithm is widely used due to its well-organized structure and outstanding results. Many merging criteria (MC) were designed to improve the accuracy of region-merging segmentation, but each MC has its own shortcomings, which can cause segmentation errors. Segmentation accuracy can be improved by referring to the segmentation results. To achieve this, an approach for detecting and correcting region-merging image segmentation errors is proposed, and then an iterative optimization model is established. The main contributions of this paper are as follows: (1) The conflict types of matching segment pairs are divided into scale-expression conflict (SEC) and region-ownership conflict (ROC), and ROC is more suitable for optimization. (2) An equal-scale local evaluation method was designed to quantify the optimization potential of ROC. (3) A regional anchoring strategy is proposed to preserve the results of the previous iteration optimization. Three QuickBird satellite images of different land-cover types were used for validating the proposed approach. Both unsupervised and supervised evaluation results prove that the proposed approach can effectively improve segmentation accuracy. All explicit and implicit optimization modes are concluded, which further illustrate the stability of the proposed approach.

show abstract

Mask R-CNN and OBIA Fusion Improves the Segmentation of Scattered Vegetation in Very High-Resolution Optical Sensors

Cited by 33 publications

References 83 publications

Precision restoration: a necessary approach to foster forest recovery in the 21st century

Precision restoration: a necessary approach to foster forest recovery in the 21st century

An Efficient Deep Learning Mechanism for the Recognition of Olive Trees in Jouf Region

Improvement of Region-Merging Image Segmentation Accuracy Using Multiple Merging Criteria

Contact Info

Product

Resources

About