Abstract: In this paper, we present an application of Riemannian
geometry for processing non-Euclidean image data. We consider the
image as residing in a Riemannian manifold, for developing a new
method to brain edge detection and brain extraction. Automating this
process is a challenge due to the high diversity in appearance brain
tissue, among different patients and sequences. The main contribution, in this paper, is the use of an edge-based
anisotropic diffusion tensor for the segmentation task by integrating
both image edge geometry and Riemannian manifold (geodesic,
metric tensor) to regularize the convergence contour and extract
complex anatomical structures. We check the accuracy of the
segmentation results on simulated brain MRI scans of single
T1-weighted, T2-weighted and Proton Density sequences. We
validate our approach using two different databases: BrainWeb
database, and MRI Multiple sclerosis Database (MRI MS DB). We
have compared, qualitatively and quantitatively, our approach with
the well-known brain extraction algorithms. We show that using
a Riemannian manifolds to medical image analysis improves the
efficient results to brain extraction, in real time, outperforming the
results of the standard techniques.
Abstract: One of the tasks of optical surveillance is to detect
anomalies in large amounts of image data. However, if the size of the
anomaly is very small, limited information is available to distinguish
it from the surrounding environment. Spectral detection provides a
useful source of additional information and may help to detect
anomalies with a size of a few pixels or less. Unfortunately, spectral
cameras are expensive because of the difficulty of separating two
spatial in addition to one spectral dimension. We investigate the
possibility of modifying a simple spectral line detector for outdoor
detection. This may be especially useful if the area of interest forms a
line, such as the horizon. We use a monochrome CCD that also
enables detection into the near infrared. A simple camera is attached
to the setup to determine which part of the environment is spectrally
imaged. Our preliminary results indicate that sensitive detection of
very small targets is indeed possible. Spectra could be taken from the
various targets by averaging columns in the line image. By imaging a
set of lines of various widths we found narrow lines that could not be
seen in the color image but remained visible in the spectral line
image. A simultaneous analysis of the entire spectra can produce
better results than visual inspection of the line spectral image. We are
presently developing calibration targets for spatial and spectral
focusing and alignment with the spatial camera. This will present
improved results and more use in outdoor application.
Abstract: As a method of expanding a higher-order tensor data to tensor products of vectors we have proposed the Third-order Orthogonal Tensor Product Expansion (3OTPE) that did similar expansion as Higher-Order Singular Value Decomposition (HOSVD). In this paper we provide a computation algorithm to improve our previous method, in which SVD is applied to the matrix that constituted by the contraction of original tensor data and one of the expansion vector obtained. The residual of the improved method is smaller than the previous method, truncating the expanding tensor products to the same number of terms. Moreover, the residual is smaller than HOSVD when applying to color image data. It is able to be confirmed that the computing time of improved method is the same as the previous method and considerably better than HOSVD.
Abstract: In this paper, a method to detect multiple ellipses is presented. The technique is efficient and robust against incomplete ellipses due to partial occlusion, noise or missing edges and outliers. It is an iterative technique that finds and removes the best ellipse until no reasonable ellipse is found. At each run, the best ellipse is extracted from randomly selected edge patches, its fitness calculated and compared to a fitness threshold. RANSAC algorithm is applied as a sampling process together with the Direct Least Square fitting of ellipses (DLS) as the fitting algorithm. In our experiment, the method performs very well and is robust against noise and spurious edges on both synthetic and real-world image data.
Abstract: Clustering is the process of subdividing an input data set into a desired number of subgroups so that members of the same subgroup are similar and members of different subgroups have diverse properties. Many heuristic algorithms have been applied to the clustering problem, which is known to be NP Hard. Genetic algorithms have been used in a wide variety of fields to perform clustering, however, the technique normally has a long running time in terms of input set size. This paper proposes an efficient genetic algorithm for clustering on very large data sets, especially on image data sets. The genetic algorithm uses the most time efficient techniques along with preprocessing of the input data set. We test our algorithm on both artificial and real image data sets, both of which are of large size. The experimental results show that our algorithm outperforms the k-means algorithm in terms of running time as well as the quality of the clustering.
Abstract: This paper presents a generalized form of the
mechanistic deconvolution technique (GMD) to modeling image sensors applicable in various pan–tilt planes of view. The mechanistic deconvolution technique (UMD) is modified with the
given angles of a pan–tilt plane of view to formulate constraint parameters and characterize distortion effects, and thereby, determine
the corrected image data. This, as a result, does not require experimental setup or calibration. Due to the mechanistic nature of
the sensor model, the necessity for the sensor image plane to be
orthogonal to its z-axis is eliminated, and it reduces the dependency on image data. An experiment was constructed to evaluate the
accuracy of a model created by GMD and its insensitivity to changes in sensor properties and in pan and tilt angles. This was compared
with a pre-calibrated model and a model created by UMD using two sensors with different specifications. It achieved similar accuracy
with one-seventh the number of iterations and attained lower mean error by a factor of 2.4 when compared to the pre-calibrated and
UMD model respectively. The model has also shown itself to be robust and, in comparison to pre-calibrated and UMD model, improved the accuracy significantly.
Abstract: Image data holds a large amount of different context
information. However, as of today, these resources remain largely
untouched. It is thus the aim of this paper to present a basic technical
framework which allows for a quick and easy exploitation of context
information from image data especially by non-expert users.
Furthermore, the proposed framework is discussed in detail
concerning important social and ethical issues which demand special
requirements in system design. Finally, a first sensor prototype is
presented which meets the identified requirements. Additionally,
necessary implications for the software and hardware design of the
system are discussed, rendering a sensor system which could be
regarded as a good, acceptable and justifiable technical and thereby
enabling the extraction of context information from image data.
Abstract: This paper introduces and studies new indexing techniques for content-based queries in images databases. Indexing is the key to providing sophisticated, accurate and fast searches for queries in image data. This research describes a new indexing approach, which depends on linear modeling of signals, using bases for modeling. A basis is a set of chosen images, and modeling an image is a least-squares approximation of the image as a linear combination of the basis images. The coefficients of the basis images are taken together to serve as index for that image. The paper describes the implementation of the indexing scheme, and presents the findings of our extensive evaluation that was conducted to optimize (1) the choice of the basis matrix (B), and (2) the size of the index A (N). Furthermore, we compare the performance of our indexing scheme with other schemes. Our results show that our scheme has significantly higher performance.