Adam W. Hoover

An experimental comparison of range image segmentation algorithms

A methodology for evaluating range image segmentation algorithms is proposed. This methodology involves (1) a common set of 40 laser range finder images and 40 structured light scanner images that have manually specified ground truth and (2) a set of defined performance metrics for instances of correctly segmented, missed, and noise regions, over- and under-segmentation, and accuracy of the recovered geometry. A tool is used to objectively compare a machine generated segmentation against the specified ground truth. Four research groups have contributed to evaluate their own algorithm for segmenting a range image into planar patches.

Locating the optic nerve in a retinal image using the fuzzy convergence of the blood vessels

We describe an automated method to locate the optic nerve in images of the ocular fundus. Our method uses a novel algorithm we call fuzzy convergence to determine the origination of the blood vessel network. We evaluate our method using 31 images of healthy retinas and 50 images of diseased retinas, containing such diverse symptoms as tortuous vessels, choroidal neovascularization, and hemorrhages that completely obscure the actual nerve. On this difficult data set, our method achieved 89% correct detection. We also compare our method against three simpler methods, demonstrating the performance improvement. All our images and data are freely available for other researchers to use in evaluating related methods.

Detecting Periods of Eating During Free-Living by Tracking Wrist Motion

This paper is motivated by the growing prevalence of obesity, a health problem affecting over 500 million people. Measurements of energy intake are commonly used for the study and treatment of obesity. However, the most widely used tools rely upon self-report and require a considerable manual effort, leading to under reporting of consumption, noncompliance, and discontinued use over the long term. The purpose of this paper is to describe a new method that uses a watch-like configuration of sensors to continuously track wrist motion throughout the day and automatically detect periods of eating. Our method uses the novel idea that meals tend to be preceded and succeeded by the periods of vigorous wrist motion. We describe an algorithm that segments and classifies such periods as eating or noneating activities. We also evaluate our method on a large dataset (43 subjects, 449 total h of data, containing 116 periods of eating) collected during free-living. Our results show an accuracy of 81% for detecting eating at 1-s resolution in comparison to manually marked event logs of periods eating. These results indicate that vigorous wrist motion is a useful indicator for identifying the boundaries of eating activities, and that our method should prove useful in the continued development of body-worn sensor tools for monitoring energy intake.

Drusen Detection in a Retinal Image Using Multi-level Analysis

This paper concerns a method to automatically detect drusen in a retinal image without human supervision or interaction. We use a multi-level approach, beginning with classification at the pixel level and proceeding to the region level, area level, and then image level. This allows the lowest levels of classification to be tuned to detect even the faintest and most difficult to discern drusen, relying upon the higher levels of classification to use an ever broadening context to refine the segmentation. We test our methods on a set of 119 images containing all types of drusen as well as images containing no drusen or other potentially confusing lesions. Our overall correct detection rate is 87%.

Slowing Bite-Rate Reduces Energy Intake: An Application of the Bite Counter Device

Slow eating may be associated with reduced energy intake. A device that counts bites can provide bite-rate feedback to the user. The purpose of this study was to explore the bite counters utility for slowing bite-rate and reducing energy intake. The study was a within-participants design with three conditions. From February to April 2009, university students (N=30) ate three meals in the laboratory: a baseline meal without feedback (Baseline), a meal during which participants received bite-rate feedback (Feedback), and a meal during which participants followed a 50% slower bite-rate target (Slow Bite-Rate). Kilocalories of food consumed, ratings of satiation and food-liking, and milliliters of water consumed were statistically compared across conditions using repeated-measures analyses of variance. Overall, participants ate 70 kcal fewer during the Slow Bite-Rate condition compared with the Feedback condition. In addition, when baseline energy consumption was added post hoc as a grouping variable, participants who ate more than 400 kcal at baseline (n=11) ate 164 kcal fewer during the Slow Bite-Rate condition compared to Baseline, and 142 kcal fewer in the Feedback condition compared with Baseline. However, the Slow Bite-Rate condition did not significantly affect participants who ate fewer than 400 kcal at baseline (n=19). Therefore, it seems that slowing bite-rate with the bite counter may be most effective for individuals who consume larger amounts of food. Future research should explore more foods and drinks, more diverse groups of individuals, potential moderating variables, and additional applications of the bite counter.

Real-time detection of workload changes using heart rate variability

Abstract This work presents a novel approach to detecting real-time changes in workload using heart rate variability (HRV). We propose that for a given workload state, the values of HRV vary in a sub-range of a Gaussian distribution. We describe methods to monitor a HRV signal in real-time for change points based upon sub-Gaussian fitting. We tested our method on subjects sitting at a computer performing a low workload surveillance task and a high workload video game task. The proposed algorithm showed superior performance compared to the classic CUSUM method for detecting task changes.

A real-time occupancy map from multiple video streams

We describe an algorithm to fuse intensity data from multiple video cameras to create a spatial-temporal occupancy map. The camera layout is assumed to resemble a security video network. The occupancy map is a two-dimensional raster image, uniformly distributed in the floor-plane. Each map pixel contains a binary value, signifying whether the designated floorspace is empty or occupied. Our algorithm requires only one difference and one look-up table operation to determine each pixels effect upon the map. This brevity of operations allows the spatial occupancy map to be temporally computed at real-time video rates. We demonstrate our algorithm operating in several dynamic scenarios.

Sensor network perception for mobile robotics

The dominant architecture for mobile robot perception uses sensors on-board the robot, providing only a first-person perspective on the environment. This work describes a novel mobile robot system that uses an environment-based sensor network, providing a powerful third-person perspective on the environment. In our previous work (1999), we described an algorithm that computes a real-time spatial-temporal occupancy map from the multiple video streams of the sensor network. We also described a novel path-planning algorithm based upon this system. In this work we describe a novel motion control loop that is based upon tracking the mobile robot in the occupancy map. Tracking in the fused perceptual space of the sensor network provides several advantages over tracking individually in a set of raw sensor spaces. We demonstrate a prototype system operating in several dynamic scenarios.

Examining the Utility of a Bite-Count–Based Measure of Eating Activity in Free-Living Human Beings

The obesity epidemic has triggered a need for novel methods for measuring eating activity in free-living settings. Here, we introduce a bite-count method that has the potential to be used in long-term investigations of eating activity. The purpose of our observational study was to describe the relationship between bite count and energy intake and determine whether there are sex and body mass index group differences in kilocalories per bite in free-living human beings. From October 2011 to February 2012, 77 participants used a wrist-worn device for 2 weeks to measure bite count during 2,975 eating activities. An automated self-administered 24-hour recall was completed daily to provide kilocalorie estimates for each eating activity. Pearsons correlation indicated a moderate, positive correlation between bite count and kilocalories (r=0.44; P<0.001) across all 2,975 eating activities. The average per-individual correlation was 0.53. A 2 (sex)×3 (body mass index group: normal, overweight, obese) analysis of variance indicated that men consumed 6 kcal more per bite than women on average. However, there were no body mass index group differences in kilocalories per bite. This was the longest study of a body-worn sensor for monitoring eating activity of free-living human beings to date, which highlights the strong potential for this method to be used in future, long-term investigations.

Function-based recognition from incomplete knowledge of shape

The paper describes a function-based reasoning system that analyzes incomplete object shape descriptions of the type that can be acquired from a single view or from a sequence of views in which not all of the object is seen. Results are presented for the analysis of the partial shape descriptions extracted from over 200 real laser range finder images taken by a mobile robot. The work represents the first function-based recognition system to reason from incomplete knowledge of 3D object shape and also the first to analyze 3D shape descriptions acquired from real image data. Results demonstrate that function based recognition is a feasible and practical capability to pursue for autonomous robotic systems. >