Spelling suggestions: "subject:"attern arecognition"" "subject:"attern 2recognition""
521 |
A comparative study of segmentation algorithms applied to 2- and 3- dimensional medical imagesManousakas, Ioannis January 1995 (has links)
A method that enables discriminating between CSF-grey matter edges and grey-white matter edges separately has been suggested. It was obvious that edges from this method are more complete that those resolved by the original method and have fewer artifacts. Some edges that were undetected before, are now detected because they do not have any influence from stronger nearby edges. Texture noise is also suppressed and this allows us to work at higher space scales. These 3D edge detection methods proved to be superior to the equivalent 2D methods because they can calculate the gradient more accurately and the edges detected have better continuity which is uniformly preserved along all three directions. The split and merge technique was the second method that has been examined. The existing algorithms need data structures that have dimensions that are powers of two (quadtrees). Such a 3D method would not a practical for volume analysis because of memory limitations. For example, a 256x256x256 array of bytes is about 17Mbytes and since the method requires about 14 bytes per voxel, memory sizes that computers usually have are exceeded. In order to solve this problem an algorithm that applies a split and merge technique on non cubic datasets has been developed. Along the x,y axes, the data must have dimensions that are powers of 2 but along the z axis it is possible have any dimension that may meet the current memory limits. The method consist of three main steps a) splitting of an initial cutset, b) merging and c) grouping of the resulting nodes. An extra boundary elimination step is necessary to reduce the number of the resulting regions. The original method is controlled mainly by a parameter ε that is kept constant during the process. Currently, improvements that could be achieved introducing a level of optimisation during the grouping step are being examined. Here, the grouping is done in a way that stimulates the formation of a crystal during anealing by a progressive increase (relaxing) of the parameter ε. Such method has given different results from a method that consist of a split and merge step with ε = ε<sub>1</sub> and a step of grouping with constant ε = ε<sub>1</sub> and a step of grouping with constant ε = ε<sub>2</sub> > ε<sub>1</sub>. At the moment, it has been difficult to establish quantitative ways of measuring any level of improving since there is no objective segmentation to compare with. So far, the method has processed adequately up to a block of 32 56c256 sized slice and can produce 3D objects representing regions like the ventricles, the white or the grey matter.
|
522 |
Computer engineering techniques for collecting and processing bio-medical signals associated with human eye movementMuheilan, Mustafa M. January 1991 (has links)
This thesis is concerned with the investigation of two types of eye movements; smooth pursuit, and saccadic eye movement, each of which was analysed under normal condition, and then after the administration of alcohol. Parameters of interest in a selected range were measured using the novel approaches developed in this thesis and the results of a series of different tests compared. Much of the early work done in this area was based on minicomputers. Obviously, a microcomputer based system would be welcome because of costs, and the fact that they are readily available in many hospitals and health centres. The work reported here was carried out using the BBC microcomputer system, since it is inexpensive and commonly used in UK health institutes. The experimental facilities constructed for the work of this thesis were driven by the intention of producing a system free from many of the weaknesses in existing procedures, and to develop an essentially new approach to the problem. The starting point of the research described depends on the fact that whenever the eyes are moved a signal appears at the two poles of the eyes. This signal is known as the resting potential, or standing potential, with the cornea several mV positive with respect to the back of the globe. This potential is generated by the retinal pigment epithelium. By influencing the eyes to move in certain directions, and at certain velocities and frequencies, information can be gathered by further analysis of the signal captured. The signal captured is found to be very small (in the μV range), therefore an amplification of the signal is required, the amplifier needed must have specific features to meet the requirement of high input impedance so the signal is not distorted. This was achieved using a specially designed instrumentation amplifier. Noise which is always present in the signal, was rejected using filtering in analogue and digital forms. The analogue filter was a Butterworth filter with a frequency passband in the range between 0.1-30Hz. The digital filter chosen was the Hanning Window type. To ensure the safety of the person taking the tests care must be taken to isolate all equipment, consequently the signal collecting electronics was powered by batteries. The collected signal was interfaced to the computer using the 1MHz BUS of the BBC microcomputer. A second computer was used so that one of them can process the captured signal while the other generates a moving spot on the screen of a monitor as a stimulus for eye movement The collected signals are then processed in both the time and the frequency domain. The use of frequency domain techniques is a particularly useful form of analysis in the treatment of eye movement potentials, and is shown to extend the information that can be extracted from such signals.
|
523 |
Parametric protein shape recognitionRitchie, David W. January 1998 (has links)
This thesis investigates the use of parametric representations of protein surface shape as a computational method of recognising similar and complementary protein surfaces. Both two-dimensional and three-dimensional shape parametrisations are described. The 2D parametrisation is constructed as an expansion of real spherical harmonic basis functions. These may be augmented by special shape-scaled radial functions, using the Laguerre polynomials, to give a novel 3D parametric "surface skin" representation. Using the special rotational properties of the spherical harmonic functions, the 2D parametric surfaces of a pair of proteins may be rotated into superposition rapidly and accurately. In a similar manner, complementary arrangements of a pair of proteins are generated by finding orientations which maximise the degree of overlap between opposing pairs of 3D skins, thus "docking" the two surfaces. This novel approach to the "protein docking problem" has the form of six-dimensional Fourier correlation. Protein docking using parametric skins is at least as accurate as comparable three-dimensional grid-based Fourier docking correlations, but is around two orders of magnitude faster.
|
524 |
Finding image features using deformable templates and detailed prior statistical knowledgeBennett, Alan January 1992 (has links)
Much work in image processing is essentially a bottom-up approach. An image is analysed by first applying simple image filters. The output from these filters is then grouped to recognise features and, subsequently, recognise objects. The work described in this report details essentially top-down methods. The algorithms also use simple image filters but search for known objects by using geometric models of the object outlines. The algorithms use statistical knowledge about the variation of the objects being searched for in order to guide the search to only feasible regions. The main techniques used are various deformable template algorithms where optimisations are achieved by random sampling and simulated annealing to avoid non-global extrema. The particular application here is for locating facial features including head outlines, where the results give key locations on the face and allow approximate geometric representations of the features to be reconstructed.
|
525 |
Maximum a posteriori estimation algorithms for image segmentation and restorationHu, Ying January 1992 (has links)
No description available.
|
526 |
Fuzzy techniques in digital image processing for artificial colour matchingPienkowski, A. January 1988 (has links)
No description available.
|
527 |
Objective picture quality measurement for MPEG-2 coded videoTan, Kwee Teck January 2000 (has links)
No description available.
|
528 |
Efficient and robust wavelet based image/video coding techniquesKhan, Ekram January 2003 (has links)
No description available.
|
529 |
Towards a computational theory of cognitive mapsYeap, W. K. January 1984 (has links)
No description available.
|
530 |
Spatio-temporal filtering in television picture codingCrawford, D. I. January 1983 (has links)
No description available.
|
Page generated in 0.0974 seconds