Corneal pathologies are leading causes of blindness and represent a world health problem according to the world health organization. Early detection of corneal diseases is necessary to prevent blindness. In this paper, we use transfer learning with pretrained deep learning networks to diagnose three common corneal diseases, namely, dry eye, Fuchs' endothelial dystrophy, and keratoconus as well as healthy eyes using only optical coherence tomography (OCT) images. Corneal OCT scans were obtained from 413 eyes of 269 patients and used to train, validate, and test the networks. All networks achieved all-category accuracy values > 99%, categorical area under curve values > 0:99, categorical specificity values > 99%, and categorical sensitivity values > 99% on the training, validation, and testing, respectively. The work in this paper has clinical significance and can potentially be applied in clinical practice to potentially solve a significant world health problem.
Various common corneal eye diseases, such as dry eye, Fuchs endothelial dystrophy, Keratoconus and corneal graft rejection, can be diagnosed based on the changes in the thickness of corneal microlayers. Optical Coherence Tomography (OCT) technology made it possible to obtain high resolution corneal images that show the microlayered structures of the cornea. Manual segmentation is subjective and not feasible due to the large volume of obtained images. Existing automatic methods, used for segmenting corneal layer interfaces, are not robust and they segment few corneal microlayer interfaces. Moreover, there is no large annotated database of corneal OCT images, which is an obstacle towards the application of powerful machine learning methods such as deep learning for the segmentation of corneal interfaces. In this paper, we propose a novel segmentation method for corneal OCT images using Graph Search and Radon Transform. To the best of our knowledge, we are the first to develop an automatic segmentation method for the six corneal microlayer interfaces. The proposed method involves a novel image denoising method and an inner interfaces localization method. The proposed method was tested on 15 corneal OCT images. The images were randomly selected and manually segmented by two operators. Experimental results show that our method has a mean segmentation error of 3.87 ± 5.21 pixels (i.e. 5.81 ± 7.82μm) across all interfaces compared to the segmentation of the manual operators. The two manual operators have mean segmentation difference of 4.07 ± 4.71 pixels (i.e. 6.11 ± 7.07μm). The mean running time to segment all the corneal microlayer interfaces is 6.66 ± 0.22 seconds.
Measuring the thickness of different corneal microlayers is important for the diagnosis of common corneal eye diseases such as dry eye, keratoconus, Fuchs endothelial dystrophy, and corneal graft rejection. High resolution corneal images, obtained using optical coherence tomography (OCT), made it possible to measure the thickness of different corneal microlayers in vivo. The manual segmentation of these images is subjective and time consuming. Therefore, automatic segmentation is necessary. Several methods were proposed for segmenting corneal OCT images, but none of these methods segment all the microlayer interfaces and they are not robust. In addition, the lack of a large annotated database of corneal OCT images impedes the application of machine learning methods such as deep learning which proves to be very powerful. In this paper, we present a new corneal OCT image segmentation algorithm using Randomized Hough Transform. To the best of our knowledge, we developed the first automatic segmentation method for the six corneal microlayer interfaces. The proposed method includes a robust estimate of relative distances of inner corneal interfaces with respect to outer corneal interfaces. Also, it handles properly the correct ordering and the non-intersection of corneal microlayer interfaces. The proposed method was tested on 15 corneal OCT images that were randomly selected. OCT images were manually segmented by two trained operators for comparison. Comparison with the manual segmentation shows that the proposed method has mean segmentation error of 3.77±4.25 pixels across all interfaces which corresponds to 5.66 ± 6.38μm. The mean segmentation error between the two manual operators is 4.07 ± 4.71 pixels, which corresponds to 6.11 ± 7.07μm. The proposed method takes a mean time of 2.59 ± 0.06 seconds to segment six corneal interfaces.
KEYWORDS: Video, Video surveillance, Data modeling, Analytical research, Information visualization, Image segmentation, Systems modeling, FDA class I medical device development, Visualization, Semantic video
In this paper, we present a content-adaptive audio texture based method to segment video into audio scenes. The audio
scene is modeled as a semantically consistent chunk of audio data. Our algorithm is based on "semantic audio texture
analysis." At first, we train GMM models for basic audio classes such as speech, music, etc. Then we define the
semantic audio texture based on those classes. We study and present two types of scene changes, those corresponding to
an overall audio texture change and those corresponding to a special "transition marker" used by the content creator,
such as a short stretch of music in a sitcom or silence in dramatic content. Unlike prior work using genre specific
heuristics, such as some methods presented for detecting commercials, we adaptively find out if such special transition
markers are being used and if so, which of the base classes are being used as markers without any prior knowledge about
the content. Our experimental results show that our proposed audio scene segmentation works well across a wide variety
of broadcast content genres.
In this paper, we measure the effect of the lighting direction in facial images on the performance of 2 well-known face recognition algorithms, an appearance based method and a facial feature based method. We collect hundreds/thousands of facial images of subjects with a fixed pose and under different lighting conditions through a unique facial acquisition laboratory designed specifically for this purpose. Then we present a methodology for automatically detecting the lighting direction of different face images based on statistics derived from the image. We also detect if there is any glare regions in some lighting directions. Finally we determine the most reliable lighting direction that will lead to a good quality/high performance facial image from both techniques based on our experiments with the acquired data.
Forensic odontology has long been carried out by forensic experts of law enforcement agencies for postmortem identification. We address the problem of developing an automated system for postmortem identification using dental records (dental radiographs). This automated dental identification system (ADIS) can be used by law enforcement agencies as well as military agencies throughout the United States to locate missing persons using databases of dental x rays of human remains and dental scans of missing or wanted persons. Currently, this search and identification process is carried out manually, which makes it very time-consuming in mass disasters. We propose a novel architecture for ADIS, define the functionality of its components, and describe the techniques used in realizing these components. We also present the performance of each of these components using a database of dental images.
This paper presents an automated system for human identification using dental radiographs. The goal of the system is to automatically archive dental images and enable identification based on shapes of the teeth in bitewing images. During archiving, the system builds the antemortem (AM) database, where it segments the teeth and the bones, separates each tooth into crown and root, and stores the contours of the teeth in the database. During retrieval, given a dental image of a postmortem (PM), the proposed system identifies the person from the AM database by automatically matching extracted teeth contours from the PM image to the teeth contours in the AM database. Experiments on a small database show that our method is effective for teeth segmentation, separation of teeth into crowns and roots, and matching.
With the increase in the amount of digital content that consumers have to deal with, both broadcast content and home captured video and images, there is a real need for tools that help manage multimedia content. In this paper we will describe a system that we are building to manage digital content at home. The system is designed for managing both still images and collections of video, where video can be either video captured by the consumer or clips that are captured from TV broadcast.
Histograms are the most prevalently used representation for the color content of images and video. An elaborate representation of the histograms requires specifying the color centers of the histogram bins and the count of the number of image pixels with that color. Such an elaborate representation, though expressive, may not be necessary for some tasks in image search, filtering and retrieval. A qualitative representation of the histogram is sufficient for many applications. Such as representation will be compact and greatly simplify the storage and transmission of the image representation. It will also reduce the computational complexity of search and filtering algorithms without adversely affecting the quality. We present such a compact binary descriptor for color representation. This descriptor is the quantized Haar transform coefficients of the color histogram. We show the use of this descriptor for browsing large image databases without the need for computationally expensive clustering algorithms. The compact nature of the descriptor and the associated simple similarity measure allows searching over a database of about four hours of video in less than 5-6 seconds without the use of any sophisticated indexing scheme.
KEYWORDS: Video, Image compression, Image retrieval, Digital watermarking, Databases, Video surveillance, Multimedia, Binary data, Video compression, Internet
The use of digital images and video is growing on the Internet and on consumer devices. Digital images and video are easy to manipulate, but this ease of manipulation makes tampering with digital content possible. Examples of the misuse of digital content include violating copyrights of the content and tampering with important material such as contents of video surveillance. In this paper we present an algorithm that extracts a binary signature from an image. This approach can be used to search for possible copyright violations by finding images with signatures close to that of a given image. The experimental results show that the algorithm can be very effective in helping users to retrieve sets of almost identical images from large collections of images. The signature can also be used for tamper detection. We will show that the signatures we extract are immune to quantization errors that result from compression and decompression.
Image retrieval systems, which compare the query image exhaustively with each individual image in the database, are not scalable to large databases. A scalable search system should ensure that the search time does not increase linearly with the number of images in the database. We present a clustering based indexing technique, where the images in the database are grouped into clusters of images, with similar color content using a hierarchical clustering algorithm. At search time, the query image is not compared with all the images in the database, but only with a small subset. Experiments show that this clustering-based approach offers a superior response time with high retrieval accuracy. Experiments with different database sizes indicate that for a given retrieval accuracy, the search time does not increase linearly with the database size.
KEYWORDS: Video, Databases, Video compression, Video processing, Distance measurement, Video coding, Genetic algorithms, Image compression, Signal processing, Human-machine interfaces
This paper presents a novel approach for video retrieval from a large archive of MPEG or Motion JPEG compressed video clips. We introduce a retrieval algorithm that takes a video clip as a query and searches the database for clips with similar contents. Video clips are characterized by a sequence of representative frame signatures, which are constructed from DC coefficients and motion information (`DC+M' signatures). The similarity between two video clips is determined by using their respective signatures. This method facilitates retrieval of clips for the purpose of video editing, broadcast news retrieval, or copyright violation detection.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.