Built by researchers for researchers, IDRISI is designed to support the analytical requirements of the most challenging problems confronted in our stewardship of the environment as well as provide day-to-day support for the common tasks of the GIS and Image Processing community. Clark Labs has the largest proportional research and development (R&D) budget in the industry devoted to the analytical development of geographic information technology.
In order to provide a real-world basis for the testing and development of new analytical approaches, Clark Labs engages in application research. Clients have included The Gordon and Betty Moore Foundation, Google.org, Conservation International, the US Department of Agriculture, the United Nations Environment Programme, and the World Food Programme. Recent projects include the detection of teleconnections of earth trends such as the spatial and temporal analysis of climate cycles (El Nino/La Nina), carbon modeling for REDD programs, the detection of diseased trees using hyperspectral imagery, and the predictive modeling of invasive species using neural networks.
A direct result of this applied research is that IDRISI remains current and relevant to the community it serves. The areas of significant research and development include:
Climate and Ecosystem Dynamics
At the forefront of research at Clark Labs is earth system science and it is most evident in the vertical application: Earth Trends Modeler. The Earth Trends Modeler is a major addition to the IDRISI analytical system and provides a critical set of tools for both demonstrating earth system dynamics (crucial for teaching) and monitoring trends such as land and climate change.
Earth Trends Modeler includes the standard time series tools such as a viewer for animating time series in a space-time cube, analyzing variability across varying temporal scales, and graphical and analytical techniques for analyzing long term trends. It includes also some unique features:
Principal components analysis in both T-mode and S-mode.
Multichannel singular spectrum analysis and canonical correlation analysis.
The implementation of empirical orthogonal teleconnection (EOT) analysis for uncovering characteristic patterns of variability over space and time including a multichannel empirical orthogonal teleconnection analysis.
A seasonal trend analysis that can be applied to any dataset that exhibits a seasonal response to environmental conditions.
A tool for exploring the presence of cycles in image time series utilizing a Fourier-PCA technique.
Earth Trends Modeling with Earth Trends Modeler
Climate Change Analysis with Earth Trends Modeler
Land Change Analysis
The dynamics of earth trends is most evident by our changing landscape. IDRISI provides the most comprehensive and advanced set of change analysis procedures for measuring our changing landscape and assessing the impacts of this change at both local and global scales.
Clark Labs worked with Conservation International over a period of several years to develop a modeling environment that could be used for a variety of land change scenarios and contexts. This cutting-edge tool, the Land Change Modeler for Ecological Sustainability, was released within the IDRISI software in 2006. In 2007, Clark Labs developed the Land Change Modeler as an extension for ArcGIS, broadening the accessibility of this important tool for users concerned with land change, conservation and biodiversity.
The distinctive tools for land change analysis within IDRISI include:
Image comparison tools for image differencing, image ratioing, regression differencing, change vector analysis, and qualitative data analysis.
A suite of tools for predictive land cover change modeling including Markov Chain Analysis, Cellular Automata, logistical regression and multinomial logistical regression, GEOMOD, and artificial neural networks.
The Land Change Modeler, one of the most comprehensive land change and impact analysis tools on the market. This integrated application within IDRISI (available also as an extension to ArcGIS) provides tools for modeling land cover change and its impacts on biodiversity. It includes tools for: the rapid assessment of land cover change, the identification of driving forces of change, and the use of that information to predict future scenarios, the assessment of the implications of change on species-specific habitat, the detection of change in habitat status, gap analysis, land cover pattern and change process analysis, biodiversity assessment and species distribution modeling, species corridor analysis and reserve planning.
Land Change Modeling with IDRISI
Clark Labs plays a key role in REDD (Reducing Emissions from Deforestation and Forest Degradation) initiatives through training, advising and software development. Robust modeling tools such as Land Change Modeler and GEOMOD give practitioners the ability to address the complexities inherent in REDD projects. Fully integrated into Land Change Modeler is support for REDD forest project planning aimed at Reducing Emissions from Deforestation and Forest Degradation. The REDD tool facilitates the estimation of baseline emissions from various carbon pools and allows the calculation of deferred emissions and carbon credits.
The IDRISI software includes a complete land analysis toolkit, compatible with international requirements, for mapping historical baselines and modeling future scenarios, as well as a REDD modeling facility to estimate and monitor GHG emission reductions due to REDD project implementation. Its integrated and comprehensive features allow you to process your satellite data for the land cover mapping component as well as manage and conserve forest carbon, biodiversity, and related ecosystem services. The participation of local stakeholders is also possible since IDRISI includes a unique suite of decision support tools.
Machine Learning and Neural Networks
Clark Labs pioneered the introduction of integrated neural networks and has become the leader in the development of the first ever machine learning procedures in a GIS and Image Processing system. Neural networks and related machine learning approaches are so important since they do not depend upon restrictive assumptions about the underlying character of class distributions and are capable of learning complex patterns with limited data. IDRISI is the premier system for integrated neural network and machine learning solutions with the introduction of:
An advanced Multi-Layer Perceptron (MLP) neural network classifier with the first ever automatic mode supervised training, progressive learning rate adjustment and hidden layer mapping with linear output option. The first two options offer the ability to run the MLP without the need for an in-depth understanding of how it works. The third option is one of a range provided for those actively researching the potential of neural networks as analytical tools and want to know more about how they work.
A Self-Organizing Feature Map (SOM) neural network which uses a two-dimensional neuron topology with both supervised and unsupervised output options.
A Fuzzy ARTMAP neural network with both supervised and unsupervised output options.
A Radial Basis Function neural network classifier.
An integrated Decision Tree machine learning classifier based on the ID3/C4.5 algorithm.
The first implementation of the K-Means unsupervised classification procedure as a machine learning algorithm with dynamic feedback and direct training intervention.
IDRISI includes the most extensive set of soft classifiers in the industry. Soft classifiers express the degree of support for each of a set of potential land cover classes at each pixel location. Thus, rather than a single map of most likely class membership, a set of images (one for each class) is produced expressing the degree of support. Soft classifiers can be used for a variety of purposes including uncertainty management (i.e., Why is the classifier having difficulty classifying this pixel?) and sub-pixel classification (i.e., What are the proportions of cover types mixed into this pixel?) Specific innovations developed by Clark Labs include:
First-of-its-kind, innovative solutions to the band limitations of linear spectral unmixing. Normally, the number of parent classes (end members) in sub-pixel classification is limited to the number of input bands. IDRISI has removed this restriction through a logical pairing of soft classifier approaches.
The first introduction of soft classifiers based on Bayesian, Mahalanobis Typicality, Dempster-Shafer Belief and Plausibility, and Fuzzy Set membership metrics.
Multi-Criteria / Multi-Objective Decision Making
In 1993, IDRISI introduced the first instance of Multi-Criteria and Multi-Objective decision making tools in GIS. Eighteen years later, IDRISI is still the industry leader, responsible for:
The first implementation of the Ordered-Weighted Average for multi-criteria evaluation that allows one to balance the relative amount of tradeoff between criteria with decision risk in balancing discordant information.
The first implementation of the MOLA heuristic for multi-objective land allocation.
The first GIS software implementation of Saatys Analytical Hierarchy Process (AHP).
The first great horizon for GIS was conquering complexity. Computers and software have done that exceptionally well. The next great horizon is the conquest of uncertainty. Clark Labs has taken a pioneering role in this area with the following selective developments in IDRISI:
The first ever implementation of a Dempster-Shafer evidence aggregation procedure in GIS.
The first soft reclassification procedure (PCLASS) that allows one to map the probability of a location being above or below a threshold (such as sea level rise).
A first ever module to generate normal and rectilinear distributions for uncertainty analysis such as Monte Carlo.
The only implementation of spatial prior probabilities for Maximum Likelihood classification.
IDRISI has always been recognized as a pioneer in the analysis and modeling of spatial processes. Specific innovations include:
IDRISI is the only system that has implemented dynamic modeling using a graphical interface. IDRISI’s Macro Modeler interface provides, without question, the premier modeling interface in the industry complete with feedback loops and dynamic layer groups for batch processing. It is so advanced that it is a primary tool in our own development of new analytical modules.
Below are some recent research projects:
Provision of Training and Production of Spatial Models of Future Deforestation in Suriname
Clark Labs of Clark University and Imazon have partnered to assist Conservation International (CI) and the government of Suriname in its planning for REDD (Reducing Emission from Deforestation and Degradation). The work includes technical capacity building and the creation of data products needed to estimate a national reference level of greenhouse gas emissions. As part of this project, Clark Labs and Imazon provided advanced training on spatial modeling and produced a spatial model of future deforestation in Suriname. KFW, the German donor agency, funded this project. This was the third training Clark Labs has provided with Conservation International for the Surinamese government.
Development of Early Warning System for Food, Health and Ecosystem Security
Clark Labs received a grant of over $1.2 million to research the potential for a climate teleconnection-based early warning system for food, health and ecosystem security. Jointly funded by the Gordon and Betty Moore Foundation and Google.org, the project investigates the relationship between climate teleconnections and problematic climatic episodes that lead to crop failures, infectious disease outbreaks and ecosystem disruptions such as fire. Teleconnections refer to a linkage between climate changes over widely separated regions of the earth.
Development and Modeling of Carbon Emissions Baselines
Clark Labs worked with Conservation International (CI) to assist in the development of site-level carbon emissions baselines for a subset of their projects in their Reduced Emissions from Deforestation and Degradation (REDD) program. CI’s Regional Programs Department (RPD) and the Center for Applied Biodiversity (CABS) is charged with developing a portfolio of site-based carbon-offset projects. This task required developing the expertise of its regional programs to conduct REDD project development, including the technical capacity to estimate carbon-emissions baselines that are transparent, and validated to a level to achieve confidence from the carbon-investment community.
Tropical Biodiversity Trend Analysis
Clark Labs received a grant from the Gordon and Betty Moore Foundation to develop a system for analyzing time series of remotely sensed images to infer changes in biodiversity and ecosystem function. The result is a protocol and software system that permits scientists and conservationists to monitor ecosystem responses to a wide range of anthropogenic disturbances such as land cover conversion and climate change.
Species Distribution and Biodiversity Modeling for the Andean Region
Clark Labs worked with Conservation International to explore the use of species range polygons (drawn by experts) as inputs to species distribution modeling procedures based on environmental variables. Procedures were developed for the semi-automated processing of range polygon archives to develop new species distribution maps and maps of biodiversity at a 1 km resolution for all of South America.
Pest Infestation Risk Mapping
Clark Labs worked with USDA to construct maps of potential gypsy moth establishment and damage in uninfested areas of the United States. The current gypsy moth management program strongly relies on survey data acquired from their 350,000 pheromone-based traps that are deployed annually. The main risk factor considered in the trap distribution formula is the proximity of humans and suitable gypsy moth hosts. Incorporating other identified risk factors, such as climatic conditions, host plant quality and the degree of human activity, predictive spatial modeling techniques were used to create a risk-based model of current gypsy moth distribution in the United States.