This article is the University of Saskatchewan, Canada: doctoral thesis (author Dennis Correa Duro), a total of 191.
In the next few decades, as the world's population continues to grow and the demand for Western food increases, a large number of agricultural production areas will face increasing pressure. Coupled with the potential consequences of climate change and the increasing costs involved in current energy-intensive agricultural production methods, achieving environmental and socio-economic sustainability goals will become increasingly challenging. At the very least, to achieve these goals, it is necessary to have a deeper understanding of the rate of change in time and space in order to correctly assess how current needs affect the needs of future generations.
Since agriculture is a fundamental part of modern society and the most common form of human-induced landscape changes on the planet, mapping and tracking these environmental changes is a critical first step towards achieving sustainability goals. In view of the increasing demand for consistent and timely information related to agricultural development, this paper proposes several advances in the field of geoinformatics, and has made specific contributions in the field of remote sensing and spatial analysis:
First, using pixel-based and object-based image analysis methods, the relative advantages of several supervised machine learning algorithms for remote sensing image classification are evaluated.
Second, the feature selection process based on the random forest classifier is applied to large data sets to reduce the total number of object-based predictors used by the classification model without sacrificing the overall classification accuracy.
Third, an object-based hybrid change detection method is introduced, which can process different image sources, generate change thresholds for each type, and minimize map update errors.
Fourth, the coarse-scale agricultural census data is spatially decomposed to show the agricultural development indicators of the 9,000 square kilometer watershed in southwestern Saskatchewan in a clear spatial manner, with a time span of several decades. The combination of methods used represents an overall analytical framework suitable for supporting the sustainable development of the agricultural environment.
Over the coming decades it is expected that the vast amounts of area currently in agricultural production will face growing pressure to intensify as world populations continue to grow, and the demand for a more Western‐based diet increases. Coupled with the potential consequences of climate change, and the increasing costs involved with current energy‐ intensive agricultural production methods, meeting goals of environmental and socioeconomic sustainability will become ever more challenging. At a minimum, meeting such goals will require a greater understanding of rates of change, both over time and space, to properly assess how present demand may affect the needs of future generations. As agriculture represents a fundamental component of modern society, and the most ubiquitous form of human induced landscape change on the planet, it follows that mapping and tracking changes in such environments represents a crucial first step towards meeting the goal of sustainability. In anticipation of the mounting need for consistent and timely information related to agricultural development, this thesis proposes several advances in the field of geomatics, with specific contributions in the areas of remote sensing and spatial analysis: First, the relative strengths of several supervised machine learning algorithms used to classify remotely sensed imagery were assessed using two image analysis approaches: pixel‐based and object‐based. Second, a feature selection process, based on a Random Forest classifier, was applied to a large data set to reduce the overall number of object‐based predictor variables used by a classification model without sacrificing overall classification accuracy. Third, a hybrid object‐ based change detection method was introduced with the ability to handle disparate image sources, generate per‐class change thresholds, and minimize map updating errors. Fourth, a spatial disaggregation procedure was performed on coarse scale agricultural census data to render an indicator of agricultural development in a spatially explicit manner across a 9,000 km2 watershed in southwest Saskatchewan for three time periods spanning several decades. The combination of methodologies introduced represents an overall analytical framework suitable for supporting the sustainable development of agricultural environments.
- Remote sensing and geographic information system methods for land cover and land use in agricultural environment
- Comparison of pixel-based and object-based image analysis and machine learning algorithms in SPOT-5 HRG image agricultural landscape classification
- Object-based multi-scale image analysis and random forest-based multi-sensor earth observation image feature selection
- A hybrid object-based change detection method (used for multi-sensor data sets in historical landscape reconstruction)
- Using remote sensing observations and spatial decomposition to explain changes in land cover and agricultural land intensity in a large watershed in southwestern Saskatchewan, Canada (1976-2005)
For more exciting articles, please pay attention to the public account: