Our services are geared toward helping our clients derive the most knowledge out of their data and information so they can focus on timely, productive decision-making. We achieve this with a scientific approach, backed by more than 20 years of experience in decision-making support, and a seasoned, results-centred defence pedigree.
Data fusion is the process of combining data to refine state estimates and predictions.
In practice, the goal is for the final picture to provide more meaningful information than the sum of its parts and a better understanding of the meaning and evolution of this picture in the context of the observer's goals. Although Fusion originated as an applied defence process, the tools developed and the ground already covered, intersect with many application fields, and we have the proven experience to efficiently apply the technology with palpable results.
OODA is a leader in the field of fusion, chairing the Board of Directors of the International Society for Information Fusion (ISIF), organizing studies and workshops, and promoting its use.
 Steinberg, A.N., Bowman, C.L., and White, F.E.,“Revisions to the JDL Data Fusion Model”, in Sensor Fusion: Architectures, Algorithms, and Applications, Proceedings of the SPIE, Vol. 3719, 1999
An essential element of knowledge discovery is to unearth unknown models of activity or behaviour from large volumes of data.
With products under its belt like Bullbear and Marketsense, OODA is particularly competent in performing natural language processing on vast quantities of documents and massive databases to extract meaningful information.
OODA builds large relational databases that acquire data from multiple sources in real-time and are easily extensible. We have pushed the speed limits of PostgreSQL centralized databases allowing them to comfortably handle terabytes of data.
A strong database architecture is often the backbone of data analysis and OODA has the capability to tailor it to our client’s specific needs.
OODA’s experience extends to producing clear visual representations of data allowing organizations to see the overall picture, often making use of geographic information systems (GIS) to provide more insight into data.
Decision support systems provide automated support to human decision making processes to facilitate and accelerate the decision cycles.
Automated decision support technologies are increasingly being integrated in military command and control (C2) systems and we have actively participated in that effort since our beginnings. We also apply our strong expertise in military decision support to diverse domains such as: finance, urban planning, pollution emergency management, etc.
At OODA, we believe that true understanding is a process of unwearying refinement, and the willingness to pursue the result to its fruition. This view is part of our work ethic, our projects and our name.
Elisa is a subject matter expert in the areas of data & information fusion, command and control systems, tactical datalinks and decision support systems. She does business development nationally and internationally identifying opportunities for project pursuits in decision support technologies.
Elisa is very active in national and international organizations involved in and promoting R&D in data/information fusion, decision support and applied mathematics: 2009 Chair and Member of the Board of Directors of the International Society for Information Fusion (ISIF)(2002-current); organized NATO Science for Peace study institutes and workshops and supported NATO Research and Technology Organisation (RTO) events (2000-current); Associate Member of the Centre de Recherches Mathématiques (CRM) at Université de Montréal (1995-current); and has been member of the Board of Directors (2000-2010) and Research Management Committee (2000-2011) for Mathematics of Information Technology and Complex Systems (MITACS) Network Centre of Excellence (NCE), Canada.
Marie-Odette is an expert in the application of mathematics and statistics to solve operational problems in diverse sectors including, but not limited to, maritime surveillance, finance, social networks and intelligence. Marie-Odette is manager and technical lead on several projects related to the exploitation of maritime information, including the development of an infrastructure to manage maritime real-time big data. She is also the co-creator of BullBear.
Eric is a maritime domain awareness, data fusion, sensor systems and systems engineering specialist. Eric has many years of experience in project management from capture to closure. He was also responsible for developing the C4ISR segment of an R&D group, including such tasks as: participating actively in proposal writing, building strong relationships with external customers and presenting at national and international events.
Yannick is an expert in algorithm development for information fusion and satellite image analysis as well as in the design and implementation of image processing, information fusion and decision support systems. He was technical lead and project manager on image analysis and information fusion projects funded by the Canadian Space Agency. Yannick was lead developer int the team that previously developed the first Canadian application segment for the US Common Operating Environment, which was successfully subjected to independent verification and validation by the Space and Naval Warfare Systems Command.
Michel is a Jack-of-all-trades in the best sense of the word. From network administration to large database modelling, Michel can investigate, analyze and solve problems in various technological and R&D areas. A very quick and resourceful learner, he can even tackle situations in very recent scientific fields. Michel's past and still ongoing experience includes system architecture, multi-agent and blackboard frameworks, distributed computing, software architecture, data fusion, sensor modeling, simulation and systems engineering. He is particularly interested in AI technology and data mining.
Dan is a seasoned software designer and an expert in big data and large network visualization and analysis. He has applied his knowledge in flexible framework design and large database management systems in the maritime surveillance domain where ship data from all over the world can be merged and stored in a single format. He contributed to the final push on the Halifax Class Modernization program at Lockheed Martin with work on data fusion database alignment as well as user interface development. He also developed several applications for social network analysis and visualization.
Mathieu Lavallée is a dynamic full-stack engineer, expert in web development and in data visualization. Mathieu is a fast learner able to adapt to ever-changing situation, but also identify and troubleshoot critical issues. From requirement analysis to deployment or from database to user interface, Mathieu will get the job done having the big picture in mind. Among others, he created an interactive web application for geomarketing data visualization.
Human operators trying to establish individual or collective maritime situational awareness often find themselves overloaded by huge amounts of information obtained from multiple and possibly dissimilar sources. This paper explores potential use of open source data mining tools, in particular R software, to enable discovery of maritime traffic patterns. It also presents an assessment of R software as a data mining tool using spatio-temporal maritime traffic data such as from the Automatic Identification System (AIS), and includes scenarios of potential interest to the maritime environment.
Rogova G. and Hadzagic M. and St-Hilaire M.-O. and Valin P. and Florea M., Context-Based Reliability in Information Fusion for Decision Making. Third IEEE Inter-Disciplinary CogSIMA 2013, San Jose, United States, February 2013.
The context-dependent nature of information quality has been well recognized in the literature. Depending on the context, the overall information quality may relate to a single quality attribute, a combination of several, or all the attributes. Selection of quality attributes, methods for their combination as well as quality control strategies require consideration of decision makers’ objectives and functions together with context parameters such as characteristics of the environment and current situations. This paper considers such context-dependent information quality attributes, namely, credibility, reliability, and timeliness, and presents a method of incorporating their combination into sequential decision making for pattern recognition. The context is represented by the time-dependent distance between an observed target and a sensor, and a situation-based time-dependent threshold on credibility. A case study involving an evidential learning system for sequential decision making designed for a quantitative evaluation of the described method, is also presented.
Shahbazian E., Pigeon L., Kraft J., Bosse E., Building Technology Enabled Decision Support Applications, Proceedings of NATO Science for Peace and Security Series E: Human and Societal Dynamics - Vol. 109, IOS Press, The Netherlands, Mar. 2013
Development of advanced technology (e.g. High Level Information Fusion (HLIF), Data Mining, Artificial Intelligence (AI), human factors, etc.) enabled decision support for prediction and recognition of piracy activity is an extremely challenging problem. It requires the consideration of multiple interrelated factors and constraints for numerous aspects of the system, including legal, jurisdictional, information sharing and exchange, information availability and quality, user roles and hierarchy, application of hardware and software technologies, etc. which do not lend themselves to simple analyses. There are numerous stakeholders, and no one has either an understanding of, or access to, the entire knowledge to be able to resolve any of these challenges by themselves. There are too many unknowns for a traditional top-down design methodology to be feasible to use. Because scenarios are task-based and descriptive, it was judged that the Scenario Based Design (SBD) approach needs to be explored for advanced technology enabled systems where the human is part of the system. This chapter presents an initial analysis of how SBD methodology could be applied to incrementally develop advanced decision support capabilities for prediction and recognition of piracy activity in collaboration with the stakeholders.
St-Hilaire M.-O. and Isenor A. W., Determining the Consistency of Information Between Multiple Subsystems used in Maritime Domain Awareness, Proceedings of NATO Science for Peace and Security Series E: Human and Societal Dynamics - Vol. 109, IOS Press, The Netherlands, Mar. 2013
Multiple public web sites can be used to obtain ship-related information that is relevant to Maritime Domain Awareness (MDA). However, the quality or timeliness of this information can be suspect. In this paper a software application, called the consistency application, is presented for the comparison of ship-related information from multiple data sources. The consistency application allows a user to compare the information from these multiple data sources and generate a comparison score for the specific ship information supplied by the data source, and also for the data source as a unit. This allows the user to assess the consistency of the information provided from the data source as compared to other sources. The consistency application presently links four MDA-related information sources.
Hadzagic M. and St-Hilaire M.-O. and Valin P. and Shahbazian E., Reliability and Relevance in the Thresholded Dempster-Shafer Algorithm for ESM Data Fusion. Proceeding of the 15th International Conference on Information Fusion, Singapore, July 2012.
The volume, and imperfect and heterogeneous nature of data to be processed under time-critical conditions pose a significant challenge for the design of future Command and Control (C2) Systems for decision support. The effectiveness of a multi-source information fusion process used in such systems highly depends on the quality of information that is received and processed. This paper addresses two attributes of information quality, namely, reliability defined as relative stability and relevance, and provides a quantitative assessment of their impact on the Electronic Support Measure (ESM) data fusion process employing the Thresholded Dempster-Shafer algorithm. The fusion algorithm’s performance is evaluated using several statistical measures and ground truth scenarios.
Maritime surveillance of coastal regions requires the processing of data from a large number of heterogeneous surveillance sources. The generation of effective Maritime Domain Awareness requires that the tracks from these sources must be fused. An automated fusion process that supports Maritime Domain Awareness requires that the tracks from heterogeneous sources be modeled in such a way as to support an automated track association decision process. Collection of sensor contact reports are therefore grouped into tracks and a mathematical model of these tracks is used to support track correlation which leads to track association. This paper presents the approach for track modeling based on genetic algorithm optimization and discusses its performance.
For the past five years, Lockheed Martin Canada has been developing an advanced image interpretation system integrating classification tools and target detection operators for multispectral, hyperspectral and polarimetric SAR imagery. Improved classification maps and superior object detection and identification performances have been obtained with a pixel-driven evidential fusion of textural measurements, and end-members respectively extracted from polSAR and HIS imagery. As in many applications the objects to be extracted occupy several pixels in the image, the objective of current developments is to supplement the pixel-based fusion already performed with radiometric, textural and spectral features, by adding specific object topological attributes (shape, size) and contextual description (spatial relationships). This paper presents a description of the new primitives and the results obtained on Ikonos imagery.