This PDF file contains the front matter associated with SPIE Proceedings Volume 9851, including the Title Page, Copyright information, Table of Contents, Introduction (if any), and Conference Committee listing.

Lockheed Martin Advanced Technology Laboratories (LM ATL) is researching methods, representations, and processes for human/autonomy collaboration to scale analysis and hypotheses substantiation for intelligence analysts. This research establishes a machinereadable hypothesis representation that is commonsensical to the human analyst. The representation unifies context between the human and computer, enabling autonomy in the form of analytic software, to support the analyst through proactively acquiring, assessing, and organizing high-value information that is needed to inform and substantiate hypotheses.

Intelligence analysts demand rapid information fusion capabilities to develop and maintain accurate situational awareness and understanding of dynamic enemy threats in asymmetric military operations. The ability to extract relationships between people, groups, and locations from a variety of text datasets is critical to proactive decision making. The derived network of entities must be automatically created and presented to analysts to assist in decision making.

DECISIVE ANALYTICS Corporation (DAC) provides capabilities to automatically extract entities, relationships between entities, semantic concepts about entities, and network models of entities from text and multi-source datasets. DAC’s Natural Language Processing (NLP) Entity Analytics model entities as complex systems of attributes and interrelationships which are extracted from unstructured text via NLP algorithms. The extracted entities are automatically disambiguated via machine learning algorithms, and resolution recommendations are presented to the analyst for validation; the analyst’s expertise is leveraged in this hybrid human/computer collaborative model. Military capability is enhanced by these NLP Entity Analytics because analysts can now create/update an entity profile with intelligence automatically extracted from unstructured text, thereby fusing entity knowledge from structured and unstructured data sources. Operational and sustainment costs are reduced since analysts do not have to manually tag and resolve entities.

Within operational environments decisions must be made quickly based on the information available. Identifying an appropriate knowledge base and accurately formulating a search query are critical tasks for decision-making effectiveness in dynamic situations. The spreading of graph data management tools to access large graph databases is a rapidly emerging research area of potential benefit to the intelligence community. A graph representation provides a natural way of modeling data in a wide variety of domains. Graph structures use nodes, edges, and properties to represent and store data. This research investigates the advantages of information search by graph query initiated by the analyst and interactively refined within the contextual dimensions of the answer space toward a solution. The paper introduces SLQ, a user-friendly graph querying system enabling the visual formulation of schemaless and structureless graph queries. SLQ is demonstrated with an intelligence analyst information search scenario focused on identifying individuals responsible for manufacturing a mosquito-hosted deadly virus. The scenario highlights the interactive construction of graph queries without prior training in complex query languages or graph databases, intuitive navigation through the problem space, and visualization of results in graphical format.

In recent years, many new interaction technologies have been developed that enhance the usability of computer systems and allow for novel types of interaction. The areas of application for these technologies have mostly been in gaming and entertainment. However, in professional environments, there are especially demanding tasks that would greatly benefit from improved human machine interfaces as well as an overall improved user experience. We, therefore, envisioned and built an image-interpretation-workstation of the future, a multi-monitor workplace comprised of four screens. Each screen is dedicated to a complex software product such as a geo-information system to provide geographic context, an image annotation tool, software to generate standardized reports and a tool to aid in the identification of objects. Using self-developed systems for hand tracking, pointing gestures and head pose estimation in addition to touchscreens, face identification, and speech recognition systems we created a novel approach to this complex task. For example, head pose information is used to save the position of the mouse cursor on the currently focused screen and to restore it as soon as the same screen is focused again while hand gestures allow for intuitive manipulation of 3d objects in mid-air. While the primary focus is on the task of image interpretation, all of the technologies involved provide generic ways of efficiently interacting with a multi-screen setup and could be utilized in other fields as well. In preliminary experiments, we received promising feedback from users in the military and started to tailor the functionality to their needs

Army staffs at division, brigade, and battalion levels often plan for contingency operations. As such, analysts consider the impact and potential consequences of actions taken. The Army Military Decision-Making Process (MDMP) dictates identification and evaluation of possible enemy courses of action; however, non-state actors often do not exhibit the same level and consistency of planned actions that the MDMP was originally designed to anticipate. The fourth MDMP step is a particular challenge, wargaming courses of action within the context of complex social-cultural behaviors. Agent-based Modeling (ABM) and its resulting emergent behavior is a potential solution to model terrain in terms of the human domain and improve the results and rigor of the traditional wargaming process.

Existing information extraction techniques can only partially address the problem of exploiting unreadable-large amounts text. When discussion of events and relationships is limited to simple, past-tense, factual descriptions of events, current NLP-based systems can identify events and relationships and extract a limited amount of additional information. But the simple subset of available information that existing tools can extract from text is only useful to a small set of users and problems. Automated systems need to find and separate information based on what is threatened or planned to occur, has occurred in the past, or could potentially occur. We address the problem of advanced event and relationship extraction with our event and relationship attribute recognition system, which labels generic, planned, recurring, and potential events. The approach is based on a combination of new machine learning methods, novel linguistic features, and crowd-sourced labeling. The attribute labeler closes the gap between structured event and relationship models and the complicated and nuanced language that people use to describe them. Our operational-quality event and relationship attribute labeler enables Warfighters and analysts to more thoroughly exploit information in unstructured text. This is made possible through 1) More precise event and relationship interpretation, 2) More detailed information about extracted events and relationships, and 3) More reliable and informative entity networks that acknowledge the different attributes of entity-entity relationships.

Intelligence analysts and military decision makers are faced with an onslaught of information. From the now ubiquitous presence of intelligence, surveillance, and reconnaissance (ISR) platforms providing large volumes of sensor data, to vast amounts of open source data in the form of news reports, blog postings, or social media postings, the amount of information available to a modern decision maker is staggering. Whether tasked with leading a military campaign or providing support for a humanitarian mission, being able to make sense of all the information available is a challenge. Due to the volume and velocity of this data, automated tools are required to help support reasoned, human decisions. In this paper we describe several automated techniques that are targeted at supporting decision making. Our approaches include modeling the kinematics of moving targets as motifs; developing normalcy models and detecting anomalies in kinematic data; automatically classifying the roles of users in social media; and modeling geo-spatial regions based on the behavior that takes place in them. These techniques cover a wide-range of potential decision maker needs.

A key issue in mission planning for aerial reconnaissance is to use the sensor ressources in an appropriate way. The sensor mission planning requires knowledge, e.g., about the optimal sensor type (IR/EO) or the necessary flying altitude for a specific task. There are various types of task that can be part of the mission, e.g., to detect a vehicle or investigate a bridge. The goal of this work is to examine knowledge-based approaches like ontologies and the use of them to automatically derive all needed parameters for an optimal sensor mission planning based on the task. The task-oriented mission planning is processed on the tactical level. Based on the task the aerial image analyst defines the specific evaluation conditions. Various parameters are part of the task-oriented mission planning. For example there are the scene, the flight mode, the sensor, the system and the image analysis parameters. We introduce an idea to represent the sensor mission planning task at a relatively high-level knowledge-based approach. We want to create a representation of a useful sensor mission planning where all aspects and its components are considered, what these parts do, how they relate to each other and define the rules and constraints within. The main framework is based on the definitions of target categories and purpose codes (STANAG 3596) and NIIRS level.

Future unmanned vehicle operations will see more responsibilities distributed among fewer pilots. Current systems typically involve a small team of operators maintaining control over a single aerial platform, but this arrangement results in a suboptimal configuration of operator resources to system demands. Rather than devoting the full-time attention of several operators to a single UAV, the goal should be to distribute the attention of several operators across several UAVs as needed. Under a distributed-responsibility system, operator task load would be continuously monitored, with new tasks assigned based on system needs and operator capabilities. The current paper sought to identify a set of metrics that could be used to assess workload unobtrusively and in near real-time to inform a dynamic tasking algorithm. To this end, we put 20 participants through a variable-difficulty multiple UAV management simulation. We identified a subset of candidate metrics from a larger pool of pupillary and behavioral measures. We then used these metrics as features in a machine learning algorithm to predict workload condition every 60 seconds. This procedure produced an overall classification accuracy of 78%. An automated tasker sensitive to fluctuations in operator workload could be used to efficiently delegate tasks for teams of UAV operators.

Combining results from mixed integer optimization, stochastic modeling and queuing theory, we will advance the interdisciplinary problem of efficiently and effectively allocating centrally managed resources. Academia currently fails to address this, as the esoteric demands of each of these large research areas limits work across traditional boundaries. The commercial space does not currently address these challenges due to the absence of a profit metric. By constructing algorithms that explicitly use inputs across boundaries, we are able to incorporate the advantages of using human decision makers. Key improvements in the underlying algorithms are made possible by aligning decision maker goals with the feedback loops introduced between the core optimization step and the modeling of the overall stochastic process of supply and demand. A key observation is that human decision-makers must be explicitly included in the analysis for these approaches to be ultimately successful. Transformative access gives warfighters and mission owners greater understanding of global needs and allows for relationships to guide optimal resource allocation decisions. Mastery of demand processes and optimization bottlenecks reveals long term maximum marginal utility gaps in capabilities.

We have seen significant change in the study and practice of human reasoning in recent years from both a theoretical and methodological perspective. Ubiquitous communication coupled with advances in computing and a plethora of analytic support tools have created a push for instantaneous reporting and analysis. This notion is particularly prevalent in law enforcement, emergency services and the intelligence community (IC), where commanders (and their civilian leadership) expect not only a birds’ eye view of operations as they occur, but a play-by-play analysis of operational effectiveness. This paper explores the use of Red Cell Analytics (RCA) as pedagogy to train the next-gen analyst. A group of Penn State students in the College of Information Sciences and Technology at the University Park campus of The Pennsylvania State University have been practicing Red Team Analysis since 2008. RCA draws heavily from the military application of the same concept, except student RCA problems are typically on non-military in nature. RCA students utilize a suite of analytic tools and methods to explore and develop red-cell tactics, techniques and procedures (TTPs), and apply their tradecraft across a broad threat spectrum, from student-life issues to threats to national security. The strength of RCA is not always realized by the solution but by the exploration of the analytic pathway. This paper describes the concept and use of red cell analytics to teach and promote the use of structured analytic techniques, analytic writing and critical thinking in the area of security and risk and intelligence training.

Citizen science projects are those which recruit volunteers to participate as assistants in scientific studies. Since these projects depend on volunteer efforts, understanding the motivation that drives a volunteer to collaborate is important to ensure its success.

One way to understand motivation is by interviewing the volunteers. While this approach may elicit detailed information on the volunteers' motivation and actions, it is restricted to a subset of willing participants. For web-based projects we could instead use logs of volunteers' activities, which measures which volunteer did what and when for all volunteers in a project.

In this work we present some metrics that can be calculated from the logs, based on a model of interaction. We also comment on the applicability of those metrics, describe an ongoing work that may yield more precise logs and metrics and comment on issues for further research.

In this paper, we demonstrate the use of eye-gaze metrics of unmanned aerial systems (UAS) operators as effective indices of their cognitive workload. Our analyses are based on an experiment where twenty participants performed pre-scripted UAS missions of three different difficulty levels by interacting with two custom designed graphical user interfaces (GUIs) that are displayed side by side. First, we compute several eye-gaze metrics, traditional eye movement metrics as well as newly proposed ones, and analyze their effectiveness as cognitive classifiers. Most of the eye-gaze metrics are computed by dividing the computer screen into “cells”. Then, we perform several analyses in order to select metrics for effective cognitive context classification related to our specific application; the objective of these analyses are to (i) identify appropriate ways to divide the screen into cells; (ii) select appropriate metrics for training and classification of cognitive features; and (iii) identify a suitable classification method.

Social Media (SM) is a relatively new phenomenon. Intelligence agencies have been struggling to understand how to exploit the social pulse that flows from this source. The paper starts with a brief overview of SM with some examples of how it is being used by adversaries and how we might be able to exploit this usage. Often treated as another form of open source intelligence (OSINT), we look at some of the differences with traditional OSINT compared to SM then outline the possible uses by military intelligence.

The next section looks at how SM fits into the different phases of the intelligence cycle: Direction, Collection, Processing and Dissemination. For the first phase, Direction, a number of questions are identified that can be answered typically by SM. For the second phase, the Collection, it is explained how SM, as an asset, transfers questions into methods and the use of different SM resources (e.g. marketer, cognitive behavioral psychologist) and sources to seek the required information.

SM is exploited as a multi-intelligence capability. For the Processing phase some aspects are described in how to deal with this capacity (e.g. enabling other intelligence sources) and also which techniques are used to be able to validate the SM sources used.

When sampling part of the enormous amounts of social media data it is important to consider whether the sample is representative. Any method of studying the sampled data is also prone to bias.

Sampling and bias aside the data may be generated with malicious intent, such as deception. Deception is a complicated (broad, situational, vague) concept. It seems improbable that an automated computer system would be able to find deception as such. Instead, we argue that the role of a system would be to aid the human analyst by detecting indicators, or clues, of (potential) deception. Indicators could take many forms and are typically neither necessary nor sufficient for there to be an actual deception. However, by using one or combining several of them a human may reach conclusions. Indicators are not necessarily dependent and will be added to or removed from the analysis depending on the circumstances. This modularity can help in counteracting/alleviating attacks on the system by an adversary. If we become aware that an indicator is compromised we can remove it from the analysis and/or replace it with a more sophisticated method that give us a similar indication.

Social media has become a dominating feature in modern life. Platforms like Facebook, Twitter, and Google have users all over the world. People from all walks of life use social media. For the intelligence services, social media is an element that cannot be ignored. It holds immense amount of information, and the potential to extract useful intelligence cannot be ignored. Social media has been around for sufficient time that most intelligence services recognize the fact that social media needs some form of attention. However, for the intelligence collector and analyst several aspects must be uncovered in order to fully exploit social media for intelligence purposes. This paper will present Project Avatar, an experiment in obtaining effective intelligence from social media sources, and several emerging analytic techniques to expand the intelligence gathered from these sources.

Social networks, in particular online social networks as a subset, enable the analysis of social relationships which are represented by interaction, collaboration, or other sorts of influence between people. Any set of people and their internal social relationships can be modelled as a general social graph. These relationships are formed by exchanging emails, making phone calls, or carrying out a range of other activities that build up the network. This paper presents an overview of current approaches to utilizing social media as a ubiquitous sensor network in the context of national and global security. Exploitation of social media is usually an interdisciplinary endeavour, in which the relevant technologies and methods are identified and linked in order ultimately demonstrate selected applications.

Effective and efficient intelligence is usually accomplished in a combined human and computer effort. Indeed, the intelligence process heavily depends on combining a human’s flexibility, creativity, and cognitive ability with the bandwidth and processing power of today’s computers. To improve the usability and accuracy of the intelligence analysis we will have to rely on data-processing tools at the level of natural language. Especially the collection and transformation of unstructured data into actionable, structured data requires scalable computational algorithms ranging from Artificial Intelligence, via Machine Learning, to Natural Language Processing (NLP). To support intelligence analysis on social media data, social media analytics is concerned with developing and evaluating computational tools and frameworks to collect, monitor, analyze, summarize, and visualize social media data. Analytics methods are employed to extract of significant patterns that might not be obvious.

As a result, different data representations rendering distinct aspects of content and interactions serve as a means to adapt the focus of the intelligence analysis to specific information requests.

Rapid processing and exploitation of open source information, including social media sources, in order to shorten decision-making cycles, has emerged as an important issue in intelligence analysis in recent years. Through a series of case studies and natural experiments, focussed primarily upon policing and counter-terrorism scenarios, we have developed an approach to information foraging and framing to inform decision making, drawing upon open source intelligence, in particular Twitter, due to its real-time focus and frequent use as a carrier for links to other media. Our work uses a combination of natural language (NL) and controlled natural language (CNL) processing to support information collection from human sensors, linking and schematising of collected information, and the framing of situational pictures. We illustrate the approach through a series of vignettes, highlighting (1) how relatively lightweight and reusable knowledge models (schemas) can rapidly be developed to add context to collected social media data, (2) how information from open sources can be combined with reports from trusted observers, for corroboration or to identify con icting information; and (3) how the approach supports users operating at or near the tactical edge, to rapidly task information collection and inform decision-making. The approach is supported by bespoke software tools for social media analytics and knowledge management.

The role of cyber forensics in criminal investigations has greatly increased in recent years due to the wealth of data that is collected and available to investigators. Physical forensics has also experienced a data volume and fidelity revolution due to advances in methods for DNA and trace evidence analysis. Key to extracting insight is the ability to correlate across multi-modal data, which depends critically on identifying a touch-point connecting the separate data streams. Separate data sources may be connected because they refer to the same individual, entity or event. In this paper we present a data source classification system tailored to facilitate the investigation of potential terrorist activity. This taxonomy is structured to illuminate the defining characteristics of a particular terrorist effort and designed to guide reporting to decision makers that is complete, concise, and evidence-based. The classification system has been validated and empirically utilized in the forensic analysis of a simulated terrorist activity. Next-generation analysts can use this schema to label and correlate across existing data streams, assess which critical information may be missing from the data, and identify options for collecting additional data streams to fill information gaps.

Current analysts are inundated with large volumes of data from which extraction, exploitation, and indexing are required. A future need for next-generation analysts is an appropriate balance between machine analytics from raw data and the ability of the user to interact with information through automation. Many quantitative intelligence tools and techniques have been developed which are examined towards matching analyst opportunities with recent technical trends such as big data, access to information, and visualization. The concepts and techniques summarized are derived from discussions with real analysts, documented trends of technical developments, and methods to engage future analysts with multiintelligence services. For example, qualitative techniques should be matched against physical, cognitive, and contextual quantitative analytics for intelligence reporting. Future trends include enabling knowledge search, collaborative situational sharing, and agile support for empirical decision-making and analytical reasoning.

Modern Soldiers rely upon accurate and actionable information technology to achieve mission objectives. While increasingly rich sensor networks for Areas of Operation (AO) can offer many directions for aiding Soldiers, limitations are imposed by current tactical edge systems on the rate that content can be transmitted. Furthermore, mission tasks will often require very specific sets of information which may easily be drowned out by other content sources.

Prior research on Quality and Value of Information (QoI/VoI) has aimed to define ways to prioritize information objects based on their intrinsic attributes (QoI) and perceived value to a consumer (VoI). As part of this effort, established ranking approaches for obtaining Subject Matter Expert (SME) recommendations, such as the Analytic Hierarchy Process (AHP) have been considered. However, limited work has been done to tie Soldier context – such as descriptions of their mission and tasks – back to intrinsic attributes of information objects.

As a first step toward addressing the above challenges, this work introduces an ontology-backed approach – rooted in Semantic Web publication practices – for expressing both AHP decision hierarchies and corresponding SME feedback. Following a short discussion on related QoI/VoI research, an ontology-based data structure is introduced for supporting evaluation of Information Objects, using AHP rankings designed to facilitate information object prioritization. Consistent with alternate AHP approaches, prioritization in this approach is based on pairwise comparisons between Information Objects with respect to established criteria, as well as on pairwise comparison of the criteria to assess their relative importance. The paper concludes with a discussion of both ongoing and future work.

The most representative and accurate data for testing and evaluating information extraction technologies is real-world data. Real-world operational data can provide important insights into human and sensor characteristics, interactions, and behavior. However, several challenges limit the feasibility of experimentation with real-world operational data. Realworld data lacks the precise knowledge of a “ground truth,” a critical factor for benchmarking progress of developing automated information processing technologies. Additionally, the use of real-world data is often limited by classification restrictions due to the methods of collection, procedures for processing, and tactical sensitivities related to the sources, events, or objects of interest. These challenges, along with an increase in the development of automated information extraction technologies, are fueling an emerging demand for operationally-realistic datasets for benchmarking. An approach to meet this demand is to create synthetic datasets, which are operationally-realistic yet unclassified in content. The unclassified nature of these unclassified synthetic datasets facilitates the sharing of data between military and academic researchers thus increasing coordinated testing efforts. This paper describes the expansion and augmentation of two synthetic text datasets, one initially developed through academic research collaborations with the Army. Both datasets feature simulated tactical intelligence reports regarding fictitious terrorist activity occurring within a counterinsurgency (COIN) operation. The datasets were expanded and augmented to create two military relevant datasets. The first resulting dataset was created by augmenting and merging the two to create a single larger dataset containing ground-truth. The second resulting dataset was restructured to more realistically represent the format and content of intelligence reports. The dataset transformation effort, the final datasets, and their applicability for research are presented.

In the domain of business intelligence, user-oriented software for conducting multidimensional analysis via Online- Analytical Processing (OLAP) is now commonplace. In this setting, datasets commonly have well-defined sets of dimensions and measures around which analysis tasks can be conducted. However, many forms of data used in intelligence operations – deriving from social networks, online communications, and text corpora – will consist of graphs with varying forms of potential dimensional structure. Hence, enabling OLAP over such data collections requires explicit definition and extraction of supporting dimensions and measures. Further, as Graph OLAP remains an emerging technique, limited research has been done on its user interface requirements. Namely, on effective pairing of interface designs to different types of graph-derived dimensions and measures.

This paper presents a novel technique for pairing of user interface designs to Graph OLAP datasets, rooted in Analytic Hierarchy Process (AHP) driven comparisons. Attributes of the classification strategy are encoded through an AHP ontology, developed in our alternate work and extended to support pairwise comparison of interfaces. Specifically, according to their ability, as perceived by Subject Matter Experts, to support dimensions and measures corresponding to Graph OLAP dataset attributes.

To frame this discussion, a survey is provided both on existing variations of Graph OLAP, as well as existing interface designs previously applied in multidimensional analysis settings. Following this, a review of our AHP ontology is provided, along with a listing of corresponding dataset and interface attributes applicable toward SME recommendation structuring. A walkthrough of AHP-based recommendation encoding via the ontology-based approach is then provided. The paper concludes with a short summary of proposed future directions seen as essential for this research area.

In this paper, we demonstrate the use of pupillary measurements as indices of cognitive workload. We analyze the pupillary data of twenty individuals engaged in a simulated Unmanned Aerial System (UAS) operation in order to understand and characterize the behavior of pupil dilation under varying task load (i.e., workload) levels. We present three metrics that can be employed as real-time indices of cognitive workload. In addition, we develop a predictive system utilizing the pupillary metrics to demonstrate cognitive context detection within simulated supervisory control of UAS. Further, we use pupillary data collected concurrently from the left and right eye and present comparative results of the use of separate vs. combined pupillary data for detecting cognitive context.

Some of the conventional metrics derived from gaze patterns (on computer screens) to study visual attention, engagement and fatigue are saccade counts, nearest neighbor index (NNI) and duration of dwells/fixations. Each of these metrics has drawbacks in modeling the behavior of gaze patterns; one such drawback comes from the fact that some portions on the screen are not as important as some other portions on the screen. This is addressed by computing the eye gaze metrics corresponding to important areas of interest (AOI) on the screen. There are some challenges in developing accurate AOI based metrics: firstly, the definition of AOI is always fuzzy; secondly, it is possible that the AOI may change adaptively over time. Hence, there is a need to introduce eye-gaze metrics that are aware of the AOI in the field of view; at the same time, the new metrics should be able to automatically select the AOI based on the nature of the gazes. In this paper, we propose a novel way of computing NNI based on continuous hidden Markov models (HMM) that model the gazes as 2D Gaussian observations (x-y coordinates of the gaze) with the mean at the center of the AOI and covariance that is related to the concentration of gazes. The proposed modeling allows us to accurately compute the NNI metric in the presence of multiple, undefined AOI on the screen in the presence of intermittent casual gazing that is modeled as random gazes on the screen.

SkyServer, the Internet portal for the Sloan Digital Sky Survey (SDSS) astronomic catalog, provides a set of tools that allows data access for astronomers and scientific education. One of SkyServer data access interfaces allows users to enter ad-hoc SQL statements to query the catalog. SkyServer also presents some template queries that can be used as basis for more complex queries. This interface has logged over 330 million queries submitted since 2001. It is expected that analysis of this data can be used to investigate usage patterns, identify potential new classes of queries, find similar queries, etc. and to shed some light on how users interact with the Sloan Digital Sky Survey data and how scientists have adopted the new paradigm of e-Science, which could in turn lead to enhancements on the user interfaces and experience in general. In this paper we review some approaches to SQL query mining, apply the traditional techniques used in the literature and present lessons learned, namely, that the general text mining approach for feature extraction and clustering does not seem to be adequate for this type of data, and, most importantly, we find that this type of analysis can result in very different queries being clustered together.

Tone mapping operators compress high dynamic range images to improve the picture quality on a digital display when the dynamic range of the display is lower than that of the image. However, tone mapping operators have been largely designed and evaluated based on the aesthetic quality of the resulting displayed image or how perceptually similar the compressed image appears relative to the original scene. They also often require per image tuning of parameters depending on the content of the image. In military operations, however, the amount of information that can be perceived is more important than the aesthetic quality of the image and any parameter adjustment needs to be as automated as possible regardless of the content of the image. We have conducted two studies to evaluate the perceivable detail of a set of tone mapping algorithms, and we apply our findings to develop and test an automated tone mapping algorithm that demonstrates a consistent improvement in the amount of perceived detail. An automated, and thereby predictable, tone mapping method enables a consistent presentation of perceivable features, can reduce the bandwidth required to transmit the imagery, and can improve the accessibility of the data by reducing the needed expertise of the analyst(s) viewing the imagery.