Original papersA framework for the management of agricultural resources with automated aerial imagery detection
Graphical abstract
Introduction
Located in the Pacific Ocean, the Kingdom of Tonga extends over an area of 362,000 km2. With a population of 107.122 inhabitants in 2016, 58.4% of its population depends on agriculture and forestry as a primary source of income and a key driver for economic growth. Its most prominent agricultural products are bananas, coconuts, coffee beans, vanilla beans, and roots such as cassava, sweet potato, and taro2 (Halavatau and Halavatau, 2001).
Most of the countries in the Pacific region are exposed to high-risk disasters including cyclones, earthquakes, tsunami, storm surge, volcanic eruptions, landslides, and droughts, e.g., Tonga is affected by more than one tropical cyclone every four years. Theses recurrent disasters cause damage and losses to agriculture, food security and local economy. In the last years, according to the 2015 Report of the Secretary-General on the Implementation of the International Strategy for Disaster Reduction; disasters worldwide cost around USD 1.5 trillion in economic damage. The frequency and severity of natural disasters are increasing, revealing an urgent need to strengthen the resilience of food assessments and security (FAO, 2015).
To understand how local agriculture and food security were affected by a natural disaster, aerial imagery from the site and the succeeding mapping and classification of data are required. The field of Remote Sensing over the past decades has robustly investigated faster methods to collect, produce, classify, and map earth observation data. In recent years, the use of Unmanned Aerial Vehicles (UAV) to collect data has increased rapidly, mainly for their inexpensive hardware and rapidly deploy for the collection of imagery. In parallel, the development of new technics to detect objects in optical remote sensing imagery were actively explored3 by several scholars. In 1991 an automatic tree detection and delineation from digital imagery was performed by Pinz (1991) who proposed a Vision Expert System using aerial imagery. He was able to locate the center of trees crown and estimate their radius using local brightness maxima. In 1995 Gougeon (1995), launched a rule-based algorithm, that followed the valleys of shadows between tree crow in a ground sampled distance from digital aerial imagery. Hung et al. (2006) proposed a vision-based shadow algorithm for tree crowns to detect and classify imagery from UAV, using color and texture information to segment regions of interest. Hassaan et al. (2016) presented an algorithm to count trees in urban environments using image processing techniques for vegetation segmentation and tree counting. By applying a k-means clustering algorithm and setting threshold values to green clusters centers, the algorithm was able to segment out the green portion out of any image without any noise.
Today, the development of machine learning approaches provides researchers with a conceptual alternative to solve problems in the mentioned domains without predefining the rules for a specific task. Instead, models can learn the underlying features emerging from a large amount of data. One of the most prominent approaches comes from the field of image processing and computer vision named Convolutional Neural Network (CNN). The algorithm is based on an end-to-end learning process, from raw data to semantic labels, which is an essential advantage in comparison with previous state-of-the-art methods (Nogueira et al., 2017). This model outperforms all the other approaches in tasks like image classification, object recognition and localization, and pixel-wise semantic labeling. The early implementation of CNN by LeCun et al. (1998) achieved 99.2% of accuracy in handwriting digits recognition and led the boost of CNN based image processing in the following 20 years. In recent years, large online image repositories such as ImageNet (Deng et al., 2009), and high-performance computing platforms like GPU acceleration, have contributed significantly to the success of using CNN in a large-scale image and video recognition. Competitions and challenges like the ImageNet Challenge (Russakovsky et al., 2015) and Visual Object Classes Challenge (Everingham et al., 2015) attract many researchers and as a result, state-of-art CNN models such as AlexNet (Krizhevsky et al., 2012) and VGG-Net (Simonyan & Zisserman, 2014) respectively – both available online.
Moreover, researchers can directly use or train these models on their dataset with no need to design its architecture, e.g., YOLO model (Redmon et al., 2016) achieved an excellent performance on recognition and made the real-time object localization possible. In the meantime, Long et al. (2015) with their novel model FCN achieved 20% relative improvement in pixel-wise semantic segmentation in the PASCAL VOC challenge. Also, SegNet proposed by Badrinarayanan et al. (2015) also achieved competitive performance as it is designed to be efficient both in terms of memory and computational time during prediction – It is also significantly smaller in the number of trainable parameters than other competing architectures.
The use of deep learning4 in Remote Sensing has grown exponentially since it can effectively encode spectral and spatial information based on the data itself. During the last years, considerable efforts have been made to develop various methods for the detection of different types of objects in satellite and aerial images with CNN, such as road, vegetation, tree, water, buildings, cars, etc. – In the Conclusion section we address quantitative measures to support the effectiveness of the proposed approach compared to existing approaches: Chen et al., 2014, Luus et al., 2015, Lu et al., 2017, Kussul et al., 2017, Mortensen et al., 2016, Sørensen et al., 2017, Milioto et al., 2017.
In this paper, we aim to provide an approach that fulfills the necessities of rapid food security, assessment, planning, exploitation, and management of agricultural resources; we propose a framework to efficiently localize and classify four types of tropical fruits (coconut, banana, mango, and papaya). We pursue the latter by a method to automatically identify and segment roads, so that fastest and safest ways to transport crops to adjacent warehouses or security points can be detected.
To do so, we used two supervised deep CNNs; the first CNN model performs the task of object localization, to localize and classify the type of trees. The locations of the trees are not only used to control agricultural resources, but also in scenarios of natural disasters they can be compared with the previous state to have a better understanding on how local agriculture and food security were affected. This information can directly inform and accelerate subsequent relief efforts. Additionally, we propose a method to determine the density of each of these trees to improve productivity, based on the detection results of the first CNN and presented as Density Maps to quickly comprehend the condition of the agricultural site.
The second CNN model performs a semantic segmentation, that masks the streets from the aerial imagery to help identify local transportation infrastructure and, in the scenario of natural disasters, evaluates the damage, proposing a proper plan to distribute aid across affected areas. Ultimately, we introduce a method to optimize the harvesting process, based in specific sceneries, such as maximum time, path length, and location of the warehouse and security points.
Section snippets
Data for the first CNN: Object Localization model
For this experiment, we used UAVs high-resolution imagery over satellite images, the latter is easily affected by cloudy environments. Also, freely available satellite images have lower resolution than UAV imagery. The imagery was captured in October 2017 and was made available in early 2018 as part of an Open AI Challenge coordinated by WeRobotics, Pacific Flying Labs, OpenAerialMap and the World Bank UAVs for Disaster Resilience Program. We participated in this challenge that aim to
Classification and location of trees
This CNN model is trained with the training data described in the subchapter Data for the first CNN: Object Localization model. This model is able to classify and locate different trees species. The CNN takes one square RGB image of 256 × 256 × 3 as input and provides the corresponding prediction. At the end of the prediction process, the localization results are assembled. If the distance between two or more recognized trees – of the same species – is less than a predefined threshold, the
Results and discussion
The performance of the Tree Localization and Classification model was measured by evaluating how precise the classifier was to localize trees correctly. The average Euclidian distance between the center point of the original trees and the predicted trees is 8.86406 pixels (less than one meter). The classifier was able to count 16,457 trees, out of which 12,945 were correctly located. Considering the original 13,393 trees, the overall Localization accuracy of the model is 80%.
We draw a confusion
Conclusion
This paper has investigated the use of Convolutional Neural Networks to efficiently localize and transport four types of tropical trees using aerial imagery. This new approach reduces costs and time of inventory, mapping, harvesting, and management of agricultural resources, and assess the impact of disasters on food security.
We introduce a specific case where this method fulfills the necessities of rapid assessment after natural disasters. Together with two Convolutional Neural Networks
Outlooks
This approach can be used in localization, classification or transportation of resources; for instance, in the assessment of damage in buildings after a natural disaster, food supply chain, urban and regional planning, etc. Other potential uses could be informal settlements detection, and more specifically the monitoring of rooftop materials as a means determine localized socio-economical conditions.
Data and code
The data and code of this pipeline are open source and can be accessed via this link: https://github.com/guozifeng91/south-pacific-aerial-image.
Declarations of interest
None.
References (28)
- et al.
A survey on object detection in optical remote sensing images
ISPRS J. Photogramm. Remote Sens.
(2016) - et al.
Estimating babassu palm density using automatic palm tree detection with very high spatial resolution satellite images
J. Environ. Manage.
(2017) - et al.
Precision forestry: trees counting in urban areas using visible imagery based on an unmanned aerial vehicle
IFAC-PapersOnLine
(2016) - et al.
Deep learning in agriculture: a survey
Comput. Electron. Agric.
(2018) - et al.
Towards better exploiting convolutional neural networks for remote sensing scene classification
Pattern Recogn.
(2017) - Badrinarayanan, V., Kendall, A., Cipolla, R., 2015. Segnet: a deep convolutional encoder-decoder architecture for image...
- et al.
Deep learning-based classification of hyperspectral data
IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.
(2014) - et al.
Imagenet: a large-scale hierarchical image database
A note on two problems in connexion with graphs
Numer. Math.
(1959)- et al.
The pascal visual object classes challenge: A retrospective
Int. J. Comput. Vision
(2015)
Simulation of genetic systems by automatic digital computers I. Introduction
Aust. J. Biol. Sci.
A Crown-following approach to the automatic delineation of individual tree crowns in high spatial resolution aerial images
Can. J. Remote Sens.
Cited by (28)
Adoption of Unmanned Aerial Vehicle (UAV) imagery in agricultural management: A systematic literature review
2023, Ecological InformaticsEmerging trends in the agri-food sector: Digitalisation and shift to plant-based diets
2022, Current Research in Food ScienceCitation Excerpt :Big Data Analytics (BDA) is the process of using sophisticated analytics on BD (Ciccullo et al., 2022). The most current or potential applications of BD in vegetables, fruits and other plant-based sectors include optimal planting of fruit trees using data extracted from satellite and unmanned aerial vehicle imagery (Saldana Ochoa and Guo, 2019), characterisation of size and shape phenotypes of horticultural crops using high throughput imagery (Haque et al., 2021), improvement of controlled environment agriculture, such as soilless hydroponics and others for vegetable and fruit farming (Ragaveena et al., 2021), and mitigating post-harvest losses and managing fruit and vegetable quality through machine learning (Singh et al., 2022). There are also some applications of BD focused on driving demand and meeting consumer needs for plant-based foods, such as developing new smart fruit marketing models in e-commerce (Ma and Zhang, 2022), satisfying date consumers through an automatic image classification system based on 5G technology and deep learning (Hossain et al., 2018), and generating a healthy food recommendation for the end-user in a nutrition-based vegetable system (Ludena and Ahrary, 2016).
Computer vision technology in agricultural automation —A review
2020, Information Processing in AgricultureCitation Excerpt :Continuous crop monitoring plays a prominent role in precision agriculture [88,89]. Application of UAV can contribute to more sustainable agricultural automatic crop monitoring and provide a prominent support for agricultural decision-making [90,91]. In addition, the use of UAV is advantageous when constructing a scientific framework for agricultural resource management [77,92].
Identification of native palms (Arecaceae) in tropical forest areas based on Convolutional Neural Network with UAV Images
2023, Revista Brasileira de Geografia FisicaEnhanced aerial vehicle system techniques for detection and tracking in fog, sandstorm, and snow conditions
2023, Journal of Supercomputing
- 1
The two authors contributed equally to this work.