Elisa H. Barney Smith

Towards Improved Paper-Based Election Technology

Elisa H. Barney Smith et al. — Tue, 31 Jan 2012 09:08:53 PST

Resources are presented for fostering paper-based election technology. They comprise a diverse collection of real and simulated ballot and survey images, and software tools for ballot synthesis, registration, segmentation, and ground-truthing. The grids underlying the designated location of voter marks are extracted from 13,315 degraded ballot images. The actual skew angles of sample ballots, recorded as part of complete ballot descriptions compiled with the interactive ground-truthing tool, are compared with their automatically extracted parameters. The average error is 0.1 degrees. These results provide a baseline for the application of digital image analysis to the scrutiny of electoral ballots.

Evaluation of Voting with Form Dropout Techniques for Ballot Vote Counting

Elisa H. Barney Smith et al. — Tue, 31 Jan 2012 09:08:50 PST

Vote counting accuracy has become a well-known issue in the vote collection process. Digital image processing techniques can be incorporated in the analysis of printed election ballots. Current image processing techniques in the vote collection process are heavily dependent on the anticipated, geometric positioning of the vote. These techniques don’t account for markings made outside of the requested field of input. Using various form dropout techniques, however, every mark on the form can be extracted and used by the machine to make an intelligent decision. Most methods will still miss a few marks and result in a few false alarms. This paper explores methods of voting between the results of the different mark extraction methods to improve recognition. To provide diversity a simple image subtraction technique is paired with a distance transform and a morphology based algorithm. The result has a higher detection rate and a lower false alarm rate.

Mark Detection from Scanned Ballots

Elisa H. Barney Smith et al. — Thu, 12 Jan 2012 08:04:29 PST

Analyzing paper-based election ballots requires finding all marks added to the base ballot. The position, size, shape, rotation and shade of these marks are not known a priori. Scanned ballot images have additional differences from the base ballot due to scanner noise. Different image processing techniques are evaluated to see under what conditions they are able to detect what sorts of marks. Basing mark detection on the difference of raw images was found to be much more sensitive to the mark darkness. Converting the raw images to foreground and background and then removing the form produced better results.

An Analysis of the Comparative Effectiveness of a Computerized Module in Probability Theory

Elisa H. Barney Smith et al. — Fri, 06 Jan 2012 12:09:39 PST

New Metric Describes Edge Noise in Bilevel Images

Elisa H. Barney Smith — Fri, 06 Jan 2012 11:54:24 PST

A new approach enables quantitative and qualitative characterization of varying edge noise even if the additive noise level is constant.

Model Reference Adaptive Control of Cardiac Output and Blood Pressure Through Two Drug Infusions

Elisa H. Barney Smith et al. — Fri, 06 Jan 2012 11:40:03 PST

Control of blood pressure and cardiac output, with dopamine and sodium nitroprusside has been achieved using a model reference adaptive controller. A procedure for selecting the adaptation weights for a 2×2 system has been developed. Satisfactory model following has occurred despite time delays that caused stability criteria to be violated. The structure of the plant prohibited zero steady state error in some cases, even when the parameters were known. However, the resulting plant response is allowable from a physician's viewpoint. The adaptive controller is compared with a nonadaptive controller, showing the motivation for adaptive control.

Characterization of Image Degradation Caused by Scanning

Elisa H. Barney Smith — Fri, 06 Jan 2012 09:14:49 PST

A single parameter value that represents the difference between the original and the digitized characters is determined from a binary scan of a test chart. It represents the combined effect of the point spread function diameter and the binary intensity threshold and is used to generate synthetic characters that are similar to characters scanned under the same conditions as the test chart.

Relating Statistical Image Differences and Degradation Features

Elisa Barney Smith et al. — Fri, 06 Jan 2012 08:46:51 PST

Document images are degraded through bilevel processes such as scanning, printing, and photocopying. The resulting image degradations can be categorized based either on observable degradation features or on degradation model parameters. The degradation features can be related mathematically to model parameters. In this paper we statistically compare pairs of populations of degraded character images created with different model parameters. The changes in the probability that the characters are from different populations when the model parameters vary correlate with the relationship between observable degradation features and the model parameters. The paper also shows which features have the largest impact on the image.

Engineering for Educators

Elisa Barney Smith et al. — Fri, 06 Jan 2012 08:46:49 PST

To increase the preparedness for and interest in Engineering, as well as general math-science skills in our youth, a new course has been developed in cooperation with the Education College at Boise State University called "Engineering for Educators". The goal is to reach more K-12 students for longer periods of time and at earlier ages than most outreach activities allow. The education students are mixed with the engineering students in the existing "Introduction to Engineering" course. The education students get an extra hour a week with an education professor to discuss how the engineering projects could be used in the K-12 classroom to meet state teaching standards for math and science as well as reading, writing and other non-technical subjects.

Combining the educators in with the engineers has three main benefits. First, the educators get to see what the engineering students actually learn without the material being watered down. Second, the teaching load is distributed by including more bodies in existing sections requiring the education professor to only prepare the discussion lecture and not the engineering projects. Third, by interacting with the engineering students, the educators can break down some of their stereotypical views of engineers and they may be more likely to recommend this field to their future K-12 students when they see a student with the appropriate interests.

Although this reports on the first semester that the course was offered, those involved have seen several improvements in student attitude, both engineering and education students. However, the most telling improvement is the confidence level of the education students. They have found that they can play a valuable part in any team project. Although they may not feel they have the necessary math skills they are able to organize, analyze and synthesize the ideas. Also, once the math is explained to them they realize that they do understand the concepts. The education students are also seeing numerous ways that the course projects could be used in the schools.

Plans are to continue to offer this course in the following semesters with expectations for higher enrollment including some practicing teachers. Discussion is beginning on making this course an acceptable substitute for the math & science methods course offered by the Education College.

Printer Modeling for Document Imaging

Margaret Norris et al. — Fri, 06 Jan 2012 08:46:47 PST

The microscopic details of printing often are unnoticed by humans, but can make differences that affect machine recognition of printed text. Models of the defects introduced into images by printing can be used to improve machine recognition. A probabilistic model used to generate images showing toner placement bears similarities to actual printed images. An equation derived for the average coverage of paper by toner particles having probabilistic placement is developed using geometric probability. Simulations show that averages of ‘printed images’ do have the same average coverage as the derived average coverage equations.

Modeling Image Degradations for Improving OCR

Elisa Barney Smith — Fri, 06 Jan 2012 08:46:45 PST

Clean documents are relatively easy to recognize. However, when digitizing collections of documents, the clean ones are rarely the documents that are encountered. The processes of printing and scanning documents introduce image degradations that interfere with the segmentation and recognition processes. Mathematical models of the degradation processes are presented. From these the types of degradations that are seen can be quantitatively and qualitatively described. Included in the discussion are sampling, edge spread, corner erosion, and edge noise. The relationship between these degradations and common OCR errors is described. By considering the degradation model, a theoretical foundation is available to improve the document recognition process.

Document Photography In Vitro

George Nagy et al. — Fri, 06 Jan 2012 08:46:43 PST

Experiments on evaluating and improving the mechanical configuration and target illumination in a prototype camera-based ballot counter are presented. The constraints on the mechanical design are gravity paper feed, portability, ease of use and low cost. The constraints on illumination are dictated by image processing requirements. Initial results are reported on the effects of transparent ballot cover plates, geometry, and light conditioning (color, diffusion, polarization).

The Power of External Mentors for Women Pursuing Academic Careers in Engineering and Science: Stories of MentorNet ACE and Its Protégés and Mentors

Carol B. Muller et al. — Fri, 06 Jan 2012 08:46:40 PST

"Finding my way through the completion of my doctorate and the start of my faculty career was challenging, but it was made more successful and more pleasant by the consistent mentorship I received."

"“My relationship with my mentor … has helped me with career altering decisions from a female perspective that I did not feel I could discuss with my male advisor and teachers."

"Why did I volunteer to be a mentor? … How to be successful as a PhD student and in the faculty job search shouldn't be a secret, but there is so little information on how to be successful, and having a book to read isn't the same as being able to ask questions of someone supportive and neutral willing to help guide you. What do I get from it on a day to day basis? I get to rethink how I do things, and what I would do."

While mentoring is frequently cited as a highly critical element of success for women in engineering and science, structured mentoring involving external mentors for those pursuing academic careers in these fields is relatively new. This paper considers the benefits of building cross-institutional mentoring networks to advance academic career progress for women in science and engineering.

The MentorNet Academic Career E-mentoring (ACE) program, matching graduate students, postdoctoral scholars, and early career faculty members in engineering and science fields pursuing academic careers with tenured faculty members external to their home institutions for structured e-mentoring relationships, involved 480 pairs of mentors and protégés during its first three years.

Education by Design: Connecting Engineering and Elementary Education

Rickie Miller et al. — Fri, 06 Jan 2012 08:46:39 PST

Camera-Based Ballot Counter

George Nagy et al. — Fri, 06 Jan 2012 08:46:37 PST

Portable ballot counters using camera technology and manual paper feed are potentially more reliable and less expensive than scanner based systems. We show that the spatial sampling rate, geometric linearity, point spread function, and photometric transfer function of off-the-shelf consumer cameras are acceptable for ballot imaging. However, scanner illumination is much more uniform than can be economically accomplished for variable size ballots. Therefore flat-field compensation must be designed into the image processing software. We illustrate the mechanical design of a prototype camera based ballot reader based on our comparative observations.

Statistical Image Differences, Degradation Features, and Character Distance Metrics

Elisa Barney Smith et al. — Fri, 06 Jan 2012 08:46:35 PST

Document image quality is degraded through processes such as scanning, printing, and photocopying. The resulting bilevel image degradations can be categorized based either on observable degradation features or on degradation model parameters. The image degradation features can be related mathematically to model parameters. In this paper we statistically compare pairs of populations of degraded character images created with different model parameters. The probability that the character populations were degraded by the same model parameters correlates with the relationship between observable degradation features and the model parameters. Two metrics of character difference are used: Hamming distance and moment feature distance. Knowledge about the conditions under which characters will be similar and when they will be different can influence the choice of parameters for future experiments.

Bilevel Image Degradations: Effects and Estimation

Elisa H. Barney Smith — Fri, 06 Jan 2012 08:46:34 PST

The two most significant parameters affecting degradations of bilevel images are the point spread function (PSF) width and the binarization threshold. Each pair of these values will affect an image differently. However, several combinations of these parameters will affect images in a similar fashion. This paper looks at two aspects of image degradation: the displacement of an edge, which determines stroke width, and the erosion of a corner, which affects crispness. The relationship between the PSF width and the binarization threshold and these two effects will be described. Sample characters, first with similar edge displacement and second with similar corner erosion, will show the effect of estimating the broader degradation versus the exact system parameters. Methods of estimating these degradations will also be briefly discussed.

Reports of the DAS02 Working Groups

Elisa Barney Smith et al. — Fri, 06 Jan 2012 08:46:32 PST

This document is a collection of four working group reports in the areas of digital libraries, document image retrieval, layout analysis, and Web document analysis. These reports were the outcome of discussions by participants at the Fifth IAPR International Workshop on Document Analysis Systems held in Princeton, NJ on 19-21 August 2002.

A Document Analysis System for Supporting Electronic Voting Research

Daniel Lopresti et al. — Fri, 06 Jan 2012 08:46:31 PST

As a result of well-publicized security concerns with direct recording electronic (DRE) voting, there is a growing call for systems that employ some form of paper artifact to provide a verifiable physical record of a voter's choices. In this paper, we present a system we are developing to support a multi-institution, cross-disciplinary research project examining issues that arise when paper ballots are used in elections. We survey the motivating factors behind our work, discuss the special constraints raised in processing ballots as opposed to more general document images, and describe the current status of our system.

Edge Noise in Document Images

Craig McGillivary et al. — Fri, 08 Jul 2011 08:17:27 PDT

A degradation model that describes many image degradations produced by desktop scanning is used to study the edge noise that is present in bilevel document images. The standard deviation of the additive noise does not adequately describe the noise present after the image is converted to a bilevel image. A measure of noise called Noise Spread is developed which describes the edge noise and is a function of the scanner parameters. If phase effects are removed this Noise Spread quantity is directly proportional to the expected value of the Hamming distance between scans with and without edge noise. The Noise Spread has also been correlated with the ability to accurately estimate edge locations. A simple method to estimate this quantity is proposed.