Search

Narratological studies have quite frequently focused upon linguistic structures, considered to be paradigmatic cases of narrativity, whereas pictorial signs (such as icons and symbols) or indices have received comparably much less attention. In this paper, however, I intend to outline some basic and regularly occurring narrative aspects of pictures and non-pictorial objects. As a point of departure, I shall suggest (influenced by approaches from cognitive psychology, e.g. the work of Roger Schank) that cognition basically consists of the storage and retrieval of action scripts or schemata (i.e. narrative structures) which may occur on various levels of abstraction. These schemas incorporate generalized knowledge about event sequences (e.g. the order in which specific events will take place; causal, enabling, or conventionalized relations between these events, and what kind of events occur at all in certain action sequences). Moreover, there are also scene schemas which are rather characterized by spatial than temporal relations. This means that we have mentally stored inventory information, i.e. what kinds of objects normally appear in such situations, as well as spatial-relation information, i.e. concerning the usual spatial layout of a scene. Through previous experiences we acquire a large amount of such culturally based event and scene stereotypes (along with idiosyncratic variations), either due to our previously acquired, direct familiarity with instances of events, or due to our acquaintance with written, oral, and of course pictorial descriptions of them (e.g. religious or mythological tales). They include settings, sub-goals, and actions in attempting to reach specific goals. The production and comprehension of pictorial signs, as I will claim, is frequently based upon the existence and activation of such mentally stored action and scene schemas on part of the beholders. Actually, even things in general, whether artificial or natural objects, are capable of

We outline a novel framework for automating steganographic and watermarking processes. Then, using an implementation built on this framework, we demonstrate the effect of a particular steganographic algorithm on picture quality and show how the framework may be used to generate acceptable, robust steganographic images without human intervention. The final system can serve to objectively compare the performance of various steganographic techniques.

This document consists of a one page picture of the patch of the Integrated Powerhead Demonstration (USAF Rocketdyne Aerojet NASA)., This document contains a one page picture of the IPD (Integrated Powerhead Demonstration) program patch of USAF Rocketdyne Aerojet NASA.

Tools have been developed which allow tactical data to be exchanged over Internet Protocol networks, but the quality of service necessary to operate these tools is not available for most Naval vessels at this time. The objective of this thesis is to show that using Multicast IP, distributing data in layers using an efficient protocol, and sending data with no inherent mechanism to ensure that packets arrive at their destinations will allow data to be exchanged over IP networks at much lower bandwidths than is required today while still maintaining a common tactical picture. Software was developed which interfaces to GCCS-M and exchanges data over a multicast network. This software was tested in a laboratory which simulated a Naval environment. The results of testing demonstrate the potential of using the characteristics of the track data being exchanged in a true multicast architecture to develop a efficient tactical data distribution system for users operating in the Naval environment.

The VEIL (Vision Environment Integrating Loom) project integrated advanced Knowledge Representation (KR) technology (Loom) with image understanding technology. The major innovations are as follows: 1) Use of declarative knowledge (as opposed to procedural knowledge) in vision systems made extending the recognition capabilities of the software easier. 2) Facilitated integrating high-level vision routines (recognizing sequences of scenes) with low-level routines that recognize picture elements. 3) Declarative knowledge enabled interaction with the system at a level of abstraction appropriate to the domain task. This included associating collateral information with the objects recognized by low-level image understanding programs. Loom is a powerful tool incorporating strong, frame-based representation capabilities, explicit term subsumption, and several powerful reasoning paradigms (including logical deduction, object-oriented methods, and production rules). Loom also provides knowledge representation integrity through consistency checking and truth maintenance. Infusing these facilities into the vision problem area, where strong KR capabilities have not yet been developed, significantly altered and improved the methodology for constructing vision systems. Finally, Loom was interfaced to a variety of vision processing elements to provide a new tool of extended capabilities. The result is a powerful software environment for the development of vision systems.

This grant has supported the research training of four graduate students and one undergraduate student in the area of human experimental cognition. Its primary purpose has been to broaden the range of the parent grant, Facilitation and Interference in Identification of Pictures and Words, by exploring more intensively an unexpected finding from the parent research---namely, that both explicit memory (recognition) and implicit memory (fragment completion and picture naming) show identical effects of similarity between study and test forms of the picture stimuli. We have explored two areas of research connected with this question. In the first, we have explored the question of whether the similarity effects occur because of stimulus similarity or process simularity between two sets of items. We have done this by varying whether the fragments presented at study and test are the same or not, and by whether the level of fragmentation between study and test is the same or not. This line of research has culminated in a paper published in Memory and Cognition (Snodgrass, Hischman and Fan).

This monograph looks for historical examples of information warfare in order to gain insight into its current practice. It first describes key elements of the concept of information operations, particularly as they relate to battle command. It then explores how George S. Patton and his Third Army Information Service demonstrated those ideas, and how their example offers direction for current developments in information warfare. Key sources used in research included emerging doctrinal literature on information warfare, biographical information on the professional development and command qualities of Patton, and after action reports of the Third Army and 6th Cavalry Group, the unit that constituted the Army Information Service. This monograph found that Patton aggressively sought information advantage as a battle commander, and that he demonstrated the key qualities of vision and intuition. The Third Army Information Service developed a relevant common picture of the battlefield by the expanding the instrument of directed liaison. What needs emphasis in current concept of information warfare is the improving the ability of commanders and staffs to process information. We must reemphasize the human dimensions of information operations through refined professional development.

This research is concerned with long-term facilitation and short-term interference and facilitation in identification of pictures and words. The long- term facilitation occurs when subjects are exposed to some representation of the item during a study episode, and then show improved identification of that item during a retention test. This type of facilitation is known as priming (or long- term priming) and the retention test is known as an implicit or indirect test because subjects are not instructed to think back to the prior study episode during the test. Much of our recent research has concerned the relationship between performance on the implicit test of picture fragment completion and the explicit test of recognition memory. Our major interest has been on the importance of maintaining the same surface features between study and test on performance in both implicit and explicit tests. Contrary to previous findings that explicit tests are impervious to surface changes and only sensitive to changes in meaning, we have found performance decrements from changes in surface features in explicit as well as implicit tests. These surface changes have been as subtle as differences in the level of fragmentation between study and test and as extreme as differences in the form of item (picture vs. word) between study and test. The research carried out under the grant has exploited this similarity between explicit and implicit tests within a components-of- information model of memory which accommodates both associations and dissociations between the two classes of tests.

Effective management of recruiting resources requires information on the relative cost-effectiveness of alternative methods of enhancing enlisted supply. Two key recruiting policy options are enlistment bonuses and post-service educational benefits. While numerous studies have examined the effects of these programs on enlistment outcomes, less is known about how these options affect the service history of recruits. The incentives embedded in the programs, the structure of the programs, and the types of recruits systematically drawn in by these programs may result in these benefits' significantly altering the attrition, retention, and reserve component accession rates of recruits. Such effects could dramatically influence estimated per-man-year costs of these recruiting programs. As a result, cost- effectiveness comparisons based on the enlistment effects of each program could provide only a partial picture of the programs' effects. For example, educational benefits could induce enlistees to leave after their first term but could also induce them to become part of the reserve force later. An examination of enlistment effects only rather than total force effects would provide a misleading comparison between educational benefits and bonuses.

This research is concerned with perception and memory of pictures. The theoretical motivations behind the experiments vary from area to area: in some cases, we want to test predictions of a connectionist model for picture recognition; in other we want to compare pictures with words to determine whether the two surface forms are understood at the same rate; in still other, the pictures are used as a vehicle to study questions about implicit memory. In the area of perception, interference in identification of a degraded image occurs when even more degraded images of the same object precede it. We tested, and rejected, the explanation proposed by Bruner and Potter that erroneous hypotheses about the object's identity interfere with subsequent recognition in favor of the explanation generated by our connectionist model. This explanation holds that transient activation of perceptual features common to the target and its distractors reduces the signal-to-noise ratio and causes interference. We were able to eliminate interference by having subjects solve math problems between presentations of the more degraded images. In the area of implicit memory, we found that the best priming stimulus for subsequent identification was a moderately fragmented stimulus, as compared to a very fragmented or almost complete stimulus. We developed the perceptual closure hypothesis to account for this effect--it says that the more difficult perceptual closure or completion of the fragmented figure is to achieve, the more priming occurs, as along as closure is finally achieved.

The guide presents interior commissary store decor schemes to be used by commissary officers and facilities engineers when improving and modernizing Army commissary facilities. These decor schemes are intended to create a shopping atmosphere by using 'functional decor'; that is, decor that improves the overall appearance of the commissary facility and makes shopping more efficient and satisfying. The decor schemes accomplish this by using color to create and emphasize the character of each department within the commissary store. The decor schemes tend to make the total store more understandable and facilitate identification of the departments, allowing customers to locate and select desired merchandise more quickly. By simplifying the shopping environment, the schemes allow the merchandise to be the focal point of the store. The designs focus on improvement of existing facilities, but the concepts and items listed are adaptable and can be used effectively to decorate new facilities.

This paper reviews some of the milestones in the development of techniques for the computer analysis of pictures and scenes. Topics covered include preprocessing (noise cleaning, deblurring, filtering); segmentation (region growing, decomposition of regions into parts, grouping of regions into 'objects'); property measurement (invariant properties, textural properties); shape analysis (local shape features, boundary and skeleton representations); and structural analysis (syntactic analysis, model matching). The emphasis is on describing representative ideas that are of historical importance, rather than on giving a systematic treatment of the subject.

This report describes project work for the Naval Environmental Prediction Research Facility (NEPRF) of the U.S. Navy. The unique SRI Objective Cloud Tracking System (also known as ISODATA) has been converted for use on the EPRF's computer hardware -- a NOVA computer. The original programs were written in FORTRAN for a CDC 6400 computer. The conversion has been carried out partly by means of an automatic conversion program and partly by mutual conversion. The report not only lists the conversion operations in detail, but also documents the programs so that future improvements and adaptations to interactive use with displays can begin from a clear base.

The theory of histogram modification of continuous real-valued pictures is developed. It is shown that the transformation of gray levels taking a picture's histogram to a desired histogram is unique under the constraint that the transformation be monotonic increasing. Algorithms for implementing this solution on digital pictures are discussed. A gray level transformation is useful for increasing visual constrast, but may destroy some of the information content. It is shown that solutions to the problem of minimizing the sum of the information loss and the histogram discrepancy are solutions to certain differential equations, which can be solved numerically.

Local gradients yield valuable directional information, which is useful in edge detection in digital pictures. Parallel iterative algorithms can be devised to process the edge vectors derived from the gradient. The resulting output consists of thinned edges which may be tracked very quickly. Examples of pictures processed by this method are given, and suggestions for further modifications are made.

Developments in the field of picture processing by computer during 1974 are surveyed, and a bibliography of nearly 350 references is given. The topics covered include picture compression; image enhancement and reconstruction; picture matching, edge and curve detection; picture processing implementations; pictorial pattern recognition; picture properties; picture parts and picture description; picture automata and grammars., See also report dated Jan 74, AD-773 612.

A method of detecting step edges in noisy one-dimensional input data is described. The method involves examination of differences in average gray level over ranges of positions and sizes. Unlike previously described methods, it remains reliable when edges occur close to one another.

The Knowledge Acquisition Tool is a graphics oriented user interfaces to the system under development. It provides a pictorial representation of the network and allows the user to add information in a much simplified manner. The tool does error checking for simple errors and has a facility for specifying dependency relations. These relations are graphically displayed as a set of nodes and edges. The interface to the underlying Bayesian Reasoning Tool is actually spread throughout the code, hence is somewhat inseparable from it. However, enclosed is the source code for the major functions used by the KA Tool for much of the graphics, etc. At this writing, this code is, and will continue to be, quite dynamic in structure as new ideas/different approaches are tried. The code is self-documenting. It also responds to the describe function available on the LISP machines.

The validity of the display evaluative index technique was investigated in two separate studies. Both studies indicated adequate predictive validity for the technique. The ability of the technique to predict 'decision making' was found to be superior to its ability to predict 'information extraction' processes. Generally, it was found that pictorial type displays are superior to tabular displays for 'decision making;' no difference seemed manifest between the two display types for 'information extraction.' Adding appropriate color coding seemed to augment the ability of the user to employ display information. As fact density increased, display decision making scores decreased.

The report discusses the study and evaluation of the tetrahedron beam geometry for application in microwave tubes: the plane of adjacent focii are orthogonal to one another allowing correction of spherical aberration, minimizing high frequency cross-modulation allowing chromatic abberation correction of the magnetic focusing by a compensating velocity cross-modulation obtained in curved control gaps. A demountable three cavity klystron of this kind has been designed and partially constructed. Tuning, coupling and loading mechanisms as well as cavities drift tubes and quadrupoles are described and illustrated. (Author)

Stock shots: images of people - individuals and small groups, children and adults, including 'quirky' shots.