Regina K. Ferrell

Method for the measurement of the modulation transfer function of sampled imaging systems from bar-target patterns

A rigorous and simple method for the determination of the modulation transfer function (MTF) of a sampled imaging system is presented. One calculates the MTF by imaging bar patterns and calculating the reduction in amplitude of the fundamental frequency components. The optimal set of bar-pattern frequencies that reduce errors from aliased frequency components is derived. Theoretical and experimental data are presented.

Quality of Big Data in health care

PURPOSE The current trend in Big Data analytics and in particular health information technology is toward building sophisticated models, methods and tools for business, operational and clinical intelligence. However, the critical issue of data quality required for these models is not getting the attention it deserves. The purpose of this paper is to highlight the issues of data quality in the context of Big Data health care analytics. DESIGN/METHODOLOGY/APPROACH The insights presented in this paper are the results of analytics work that was done in different organizations on a variety of health data sets. The data sets include Medicare and Medicaid claims, provider enrollment data sets from both public and private sources, electronic health records from regional health centers accessed through partnerships with health care claims processing entities under health privacy protected guidelines. FINDINGS Assessment of data quality in health care has to consider: first, the entire lifecycle of health data; second, problems arising from errors and inaccuracies in the data itself; third, the source(s) and the pedigree of the data; and fourth, how the underlying purpose of data collection impact the analytic processing and knowledge expected to be derived. Automation in the form of data handling, storage, entry and processing technologies is to be viewed as a double-edged sword. At one level, automation can be a good solution, while at another level it can create a different set of data quality issues. Implementation of health care analytics with Big Data is enabled by a road map that addresses the organizational and technological aspects of data quality assurance. PRACTICAL IMPLICATIONS The value derived from the use of analytics should be the primary determinant of data quality. Based on this premise, health care enterprises embracing Big Data should have a road map for a systematic approach to data quality. Health care data quality problems can be so very specific that organizations might have to build their own custom software or data quality rule engines. ORIGINALITY/VALUE Today, data quality issues are diagnosed and addressed in a piece-meal fashion. The authors recommend a data lifecycle approach and provide a road map, that is more appropriate with the dimensions of Big Data and fits different stages in the analytical workflow.

Semantic information extraction from multispectral geospatial imagery via a flexible framework

Identification and automatic labeling of facilities in high-resolution satellite images is a challenging task as the current thematic classification schemes and the low-level image features are not good enough to capture complex objects and their spatial relationships. In this paper we present a novel algorithm framework for automated semantic labeling of large image collections. The framework consists of various segmentation, feature extraction, vector quantization, and Latent Dirichlet Allocation modules. Initial experimental results show promise as well as the challenges in semantic classification technology development for nuclear proliferation monitoring.

Detection of semiconductor defects using a novel fractal encoding algorithm

This paper introduces a new non-referential defect detection (NRDD) algorithm for application to digital images of semiconductor wafers required during the manufacturing process. This new algorithm is composed to two major steps: (1) defect detection via the use of fractal image encoding and (2) enhanced defect boundary delineation using active contours. One primary application for this technology is the redetection of defects within archived databases of historical defect imagery. Defect images are commonly stored by semiconductor manufacturers for future diagnostic purposes, but reference images are usually unavailable. The ability to automatically redetect a defect is crucial in an automated diagnostic system that uses the historical defect images for defect sourcing. Results are presented for four large data sets of semiconductor images. Three of these data sets are composed of scanning-electron microscope (SEM) images and the fourth contains optical microscope images. Performance criteria were created that score the NRDD segmentation result as a percentage based on a comparison to a manually outlined version of the defect. The overall NRDD score across all four databases ranged from 50 percent to 84 percent using the same set of manually-determined parameters on all image within each database. By using an automated parameter setting algorithm these performance values improved to 57 percent to 92 percent. The NRDD algorithm performance depends, in part, on the size of the defect and the level of complexity of the background of the semiconductor image.

Content-based image retrieval for semiconductor process characterization

Image data management in the semiconductor manufacturing environment is becoming more problematic as the size of silicon wafers continues to increase, while the dimension of critical features continues to shrink. Fabricators rely on a growing host of image-generating inspection tools to monitor complex device manufacturing processes. These inspection tools include optical and laser scattering microscopy, confocal microscopy, scanning electron microscopy, and atomic force microscopy. The number of images that are being generated are on the order of 20,000 to 30,000 each week in some fabrication facilities today. Manufacturers currently maintain on the order of 500,000 images in their data management systems for extended periods of time. Gleaning the historical value from these large image repositories for yield improvement is difficult to accomplish using the standard database methods currently associated with these data sets (e.g., performing queries based on time and date, lot numbers, wafer identification numbers, etc.). Researchers at the Oak Ridge National Laboratory have developed and tested a content-based image retrieval technology that is specific to manufacturing environments. In this paper, we describe the feature representation of semiconductor defect images along with methods of indexing and retrieval, and results from initial field-testing in the semiconductor manufacturing environment.

Image retrieval in the industrial environment

The ability to manage large image databases has been a topic of growing research over the past several years. Imagery is being generated and maintained for a large variety of applications including remote sensing, art galleries, architectural and engineering design, geographic information systems, weather forecasting, medical diagnostics, and law enforcement. Content-based image retrieval (CBIR) represents a promising and cutting-edge technology that is being developed to address these needs. To date, little work has been accomplished to apply these technologies to the manufacturing environment. Imagery collected from manufacturing processes have unique characteristics that can be used in developing a manufacturing-specific CBIR approach. For example, a product image typically has an expected structure that can be characterized in terms of its redundancy, texture, geometry, or a mixture of these. Defect objects in product imagery share a number of common traits across product types and imaging modalities as well. For example, defects tend to be contiguous, randomly textured, irregularly shaped, and they disrupt the background and the expected pattern. We will present the initial results of the development of a new capability for manufacturing-specific CBIR that addresses defect analysis, product quality control, and process understanding in the manufacturing environment. Image data from the semiconductor-manufacturing environment will be presented.

Application of fractal encoding techniques for image segmentation

Fractal encoding is the first step in fractal based image compression techniques, but this technique can also be useful outside the image compression field. This paper discusses a fractal encoding technique and some of its variations adapted to the concept of segmenting anomalous regions within an image. The primary goal of this paper is to provide background information on fractal encoding and show application examples to equip the researcher with enough knowledge to apply this technique to other image segmentation applications. After a brief overview of the algorithm, important parameters for successful implementation of fractal encoding are discussed. Included in the discussion is the impact of image characteristics on various parameters or algorithm implementation choices in the context of two applications that have been successfully implemented.

Circumference imaging for optical-based identification of cylindrical and conical objects

Inspection and identification of cylindrical or conical shaped objects presents a unique challenge for a machine vision system. Due to the circular nature of the objects it is difficult to image the whole object using traditional area cameras and image capture methods. This work describes a unique technique to acquire a 2D image of the entire surface circumference of a cylindrical/conical shaped object. The specific application of this method is the identification of large caliber ammunition rounds in the field as they are transported between or within vehicles. The proposed method utilizes a line scan camera in combination with high speed image acquisition and processing hardware to acquire images from multiple cameras and generate a single, geometrically accurate, surface image. The primary steps involved are the capture of multiple images as the ammunition moves by on the conveyor followed by warping to correct for the distortion induced by the curved projectile surface. The individual images are then tiled together to form one 2D image of the complete circumference. Once this image has been formed an automatic identification algorithm begins the feature extraction and classification process.

Content-based image retrieval for semiconductor manufacturing

In the semiconductor manufacturing environment, defect imagery is used to diagnose problems in the manufacturing line, train automatic defect classification systems, and examine historical data for trends. Image management in semiconductor yield management systems is a growing cause of concern since many facilities collect 3000 to 5000 images each month, with future estimates of 12,000 to 20,000. Engineers at Oak Ridge National Laboratory (ORNL) have developed a semiconductor- specific content-based image retrieval architecture, also known as Automated Image Retrieval (AIR). We review the AIR system approach including the application environment as well as details on image interpretation for content-based image retrieval. We discuss the software architecture that has been designed for flexibility and applicability to a variety of implementation schemes in the fabrication environment. We next describe details of the system implementation including image processing and preparation, database indexing, and image retrieval. The image processing and preparation discussion includes a description of an image processing algorithm which enables a more accurate description of the semiconductor substrate (non-defect area). We also describe the features used that identify the key areas of the defect imagery. The feature indexing mechanisms are described next, including their implementation in a commercial database. Next, the retrieval process is described, including query image processing. Feedback mechanisms, which direct the retrieval mechanism to favor specified retrieval results, are also discussed. Finally, experimental results are shown with a database of over 10,000 images obtained from various semiconductor manufacturing facilities. These results include subjective measures of system performance and timing details for our implementation.

Baseline face detection, head pose estimation, and coarse direction detection for facial data in the SHRP2 naturalistic driving study

Keeping a driver focused on the road is one of the most critical steps in insuring the safe operation of a vehicle. The Strategic Highway Research Program 2 (SHRP2) has over 3,100 recorded videos of volunteer drivers during a period of 2 years. This extensive naturalistic driving study (NDS) contains over one million hours of video and associated data that could aid safety researchers in understanding where the drivers attention is focused. Manual analysis of this data is infeasible; therefore efforts are underway to develop automated feature extraction algorithms to process and characterize the data. The real-world nature, volume, and acquisition conditions are unmatched in the transportation community, but there are also challenges because the data has relatively low resolution, high compression rates, and differing illumination conditions. A smaller dataset, the head pose validation study, is available which used the same recording equipment as SHRP2 but is more easily accessible with less privacy constraints. In this work we report initial head pose accuracy using commercial and open source face pose estimation algorithms on the head pose validation data set.