Showing items 1 - 18 of 18

Your selections:

Characterization of a small helicopter UAV’s main rotor blade through Image Processing

- Shipman, W.J., Roodt, Y., Du Plessis, F.

TB detection using modified Local Binary Pattern features

- Leibstein, Joshua, Nel, Andre

  • Authors: Leibstein, Joshua , Nel, Andre
  • Date: 2017
  • Subjects: Computer vision , Image processing , Biomedical imaging , Tuberculosis - Diagnosis
  • Language: English
  • Type: Conference proceedings
  • Identifier: http://hdl.handle.net/10210/218346 , uj:21760 , Citation: Leibstein, J. & Nel, A. 2017. TB detection using modified Local Binary Pattern features.
  • Description: Abstract: This paper explores a computer-aided detection scheme to aid radiologists in making a higher percentage of correct diagnoses when analysing chest radiographs. The approach undertaken in the detection process is to use several proprietary image processing algorithms to adjust, segment and classify a radiograph. Firstly, a Difference of Gaussian (DoG) energy normalisation method is applied to the image. By doing this, the effect of differing equipment and calibrations is normalised. Thereafter, the lung area is detected using Active Shape Models (ASMs). Once identified, the lungs are analysed using Local Binary Patterns (LBPs). This technique is combined with a probability measure that makes use of the the locations of known abnormalities in the training dataset. The results of the segmentation when compared to ground truth masks achieves an overlap segmentation accuracy of 87,598±3,986%. The challenges faced during classification are also discussed.
  • Full Text:

Bandwidth efficient virtual classroom

- Van der Schyff, Marco

  • Authors: Van der Schyff, Marco
  • Date: 2009-02-27T05:38:56Z
  • Subjects: Image processing , Human face recognition (Computer science) , Distance education , Computer-assisted instruction
  • Type: Thesis
  • Identifier: uj:8182 , http://hdl.handle.net/10210/2186
  • Description: M.Ing. , Virtual classrooms and online-learning are growing in popularity, but there are still some factors limiting the potential. Limited bandwidth for audio and video, the resultant transmission quality and limited feedback during virtual classroom sessions are some of the problems that need to be addressed. This thesis presents information on the design and implementation of various components of a virtual classroom system for researching methods of student feedback with a focus on bandwidth conservation. A facial feature technique is implemented and used within the system to determine the viability of using facial feature extraction to provide and prioritise feedback from students to teacher while conserving bandwidth. This allows a teacher to estimate the comprehension level of the class and individual students based on student images. A server determines which student terminal transmits its images to the teacher using data obtained from the facial feature extraction process. Feedback is improved as teachers adapt to class circumstances using experience gained in traditional classrooms. Feedback is also less reliant on intentional student participation. New page-turner, page suggestion and class activity components are presented as possible methods for improving student feedback. In particular, the effect of virtual classroom system parameters on feedback delays and bandwidth usage is investigated. In general, delays are increased as bandwidth requirements decrease. The system shows promise for future use in research on facial feature extraction, student feedback and bandwidth conservation in virtual classrooms.
  • Full Text:

The contour tracking of a rugby ball : an application of particle filtering

- Janse van Rensburg, Tersia

  • Authors: Janse van Rensburg, Tersia
  • Date: 2012-02-06
  • Subjects: Electric filters , Automatic tracking , Image processing , Monte Carlo method
  • Type: Thesis
  • Identifier: uj:1996 , http://hdl.handle.net/10210/4350
  • Description: M.Ing. , Object tracking in image sequences, in its general form, is very challenging. Due to the prohibitive complexity thereof, research has lead to the idea of tracking a template exposed to low-dimensional deformation such as translation, rotation and scaling. The inherent non-Gaussianity of the data acquired from general tracking problems renders the trusted Kalman filtering methodology futile. For this reason the idea of particle filtering was developed recently. Particle filters are sequential Monte Carlo methods based on multiple point mass (or "particle") representations of probability densities, which can be applied to any dynamical model and which generalize the traditional Kalman filtering methods. To date particle filtering has already been proved to be successful filtering method in different fields of science such as econometrics, signal processing, fluid mechanics, agriculture and aviation. In this dissertation, we discuss the problem of tracking a rugby ball in an image sequence as the ball is being passed to and fro. First, the problem of non-linear Bayesian tracking is focused upon, followed by a particular instance of particle filtering known as the condensation algorithm. Next, the problem of fitting an elliptical contour to the travelling rugby ball is dealt with in detail, after which the problem of tracking this evolving ellipse (representing the rugby ball's edge) over time along the image sequence by means of the condensation algorithm follows. Experimental results are presented and discussed and concluding remarks follow at the end.
  • Full Text:

Design considerations for a marker-free visual-based interfacing device for telco operation

- Visser, W., Roodt, Y., Clarke, W.A.

Aircraft recognition using generalised variable-kernel similarity metric learning

- Naudé, Johannes Jochemus

  • Authors: Naudé, Johannes Jochemus
  • Date: 2014-12-01
  • Subjects: Image processing , Computer vision , Pattern recognition systems , Optical pattern recognition , Airplanes - Recognition
  • Type: Thesis
  • Identifier: http://ujcontent.uj.ac.za8080/10210/387397 , uj:13139 , http://hdl.handle.net/10210/13113
  • Description: M.Ing. , Nearest neighbour classifiers are well suited for use in practical pattern recognition applications for a number of reasons, including ease of implementation, rapid training, justifiable decisions and low computational load. However their generalisation performance is perceived to be inferior to that of more complex methods such as neural networks or support vector machines. Closer inspection shows however that the generalisation performance actually varies widely depending on the dataset used. On certain problems they outperform all other known classifiers while on others they fail dismally. In this thesis we allege that their sensitivity to the metric used is the reason for their mercurial performance. We also discuss some of the remedies for this problem that have been suggested in the past, most notably the variable-kernel similarity metric learning technique, and introduce our own extension to this technique. Finally these metric learning techniques are evaluated on an aircraft recognition task and critically compared.
  • Full Text:

Distributed image processing in an intranet environment

- Van den Berg, Pierre

  • Authors: Van den Berg, Pierre
  • Date: 2012-08-28
  • Subjects: Image processing , Computer vision , Computer networks , Intranets (Computer networks)
  • Type: Thesis
  • Identifier: uj:3342 , http://hdl.handle.net/10210/6742
  • Description: M.Sc. , Image Processing/Computer Vision and Computer Networks (in particular Intranets) may seem to have very little in common if one only looks at these fields of study superficially. In this dissertation we will look at some fundamentals and characteristics of Image Processing and examine them to see where the problem areas lie, focusing on the problem of computational requirements. We will also examine the fundamental characteristics of Computer Networks and Distributed Processing, looking for areas where we can potentially find a synergy with computational problems inherent in Computer Vision/Image Processing. To accomplish the goals stated above, the dissertation is divided into three parts. The first part examines Computer Vision and Image Processing and is followed by a section examining Distributed Computing models and Computer Networks. The final part is dedicated to suggesting a model to solve the problem of computational load associated with Image Processing. The aim of the model is to take advantage of and use the latent processing power available in an Intranet environment by distributing the processing among the machines on the network. The model is also intended to be flexible and to minimize the network load incurred by distributing the processing. In order to do so, the model is split into units that deliver specialized functionality in order to keep the components small and also to incur the minimum load on a specific machine.
  • Full Text:

Recognition of unconstrained handwritten digits with neural networks

- De Jaeger, André

  • Authors: De Jaeger, André
  • Date: 2014-11-19
  • Subjects: Optical character recognition devices , Image processing , Optical storage devices , Computers - Optical equipment , Neural networks (Computer science)
  • Type: Thesis
  • Identifier: uj:12910 , http://hdl.handle.net/10210/12799
  • Description: D.Ing. (Electrical and Electronic ) , This thesis describes a neural network based system for the classification of handwritten digits as found on real-life mail pieces. The proposed neural network uses a modular architecture which lends itself to parallel implementation. This modular architecture is shown to produce adequate performance levels while significantly reducing the required training time. The aim of the system is not only to achieve a high recognition performance, but also to gain more insight into the functioning of the neural networks. This is achieved by using separate feature extraction and classification stages. The output of the feature extraction stage gives a good indication of the final performance level of the classifier, even before training. The need for an optimal feature set is expressed to elevate the performance levels even further.
  • Full Text:

A study of machine enabled communication for the severely disabled patient, with reference to image processing techniques

- Nel, André Leon

  • Authors: Nel, André Leon
  • Date: 2014-02-18
  • Subjects: Speech disorders , Communication devices for people with disabilities , Image processing
  • Type: Thesis
  • Identifier: uj:4139 , http://hdl.handle.net/10210/9486
  • Description: M.Ing. (Mechanical Engineering) , This research aims at identifying technologies which could be used to produce a communication aid for the speechless severely disabled person. A number of techniques are investigated which would enable the disabled to communicate by means of decoding eye movements. The tracking of the disabled's eye positions by means of two dimensional image processing and pattern recognition forms the basis of the input mechanism. A complete comparison between a number of candidate edge detectors is done to ensure that the system would function in as closeto real-time as is possible. Ancillary techniques such as a reduced grammar and a Markov model for letter posterior probabilities are developed and shown to improve the communication channel bandwidth. Simulated resultsshowwhat communication rate gains could be achieved. It appears that a possible communication aid could be built to perform at rates exceeding present day communication aids for the severely disabled at prices that could make such a device an economic possibility.
  • Full Text:

Multimodal verification of identity for a realistic access control application

- Denys, Nele

  • Authors: Denys, Nele
  • Date: 2008-11-18T09:08:58Z
  • Subjects: Pattern recognition systems , Human face recognition (Computer science) , Optical character recognition devices , Image processing , Automatic control
  • Type: Thesis
  • Identifier: uj:14730 , http://hdl.handle.net/10210/1734
  • Description: D. Ing. , This thesis describes a real world application in the field of pattern recognition. License plate recognition and face recognition algorithms are combined to implement automated access control at the gates of RAU campus. One image of the license plate and three images of the driver’s face are enough to check if the person driving a particular car into campus is the same as the person driving this car out. The license plate recognition module is based on learning vector quantization and performs well enough to be used in a realistic environment. The face recognition module is based on the Bayes rule and while performing satisfactory, extensive research is still necessary before this system can be implemented in real life. The main reasons for failure of the system were identified as the variable lighting and insufficient landmarks for effective warping.
  • Full Text:

Segmentation of C-spine MRI images using the watershed transform

- Botha, Jacobus Johannes

  • Authors: Botha, Jacobus Johannes
  • Date: 2012-08-15
  • Subjects: Cervical vertebrae - Magnetic resonance imaging , Image processing
  • Type: Thesis
  • Identifier: uj:9406 , http://hdl.handle.net/10210/5841
  • Description: M.Ing. , Automatic classification of images has always been an important part of pattern recognition. The segmentation and classification of MRI images has always been a challenge. A segmented image is often a very important input to the classification process. Many classification techniques use segmented images as input to the classification process. Certain segments or areas of an image serve as important features that will be used for classification. Important information can be derived from the features that are present in the segmented image. Sometimes there might be a need to extract a certain object from an image to do classification on the object. In the case of MRI images, certain structures of the human body like organs and tissue can be isolated by the segmentation process. These objects of interest (001) can give vital information for the identification of medical abnormalities (anomalies) and diseases. Segmented objects can play an important role to assist medical practitioners in the diagnosis and treatment of medical problems. I would like to test the performance of the watershed segmentation algorithm on MRI images of the cervical (C) spine. Much work has been done on the segmentation and classification of MRI images. Various techniques have been generated and tested over the past decades. Segmentation techniques like thresholding, convolution, pyramid segmentation and morphological segmentation have been utilised. All these techniques have their advantages and disadvantages. The pre-processing of an image plays a very important role in the success of the segmentation process. Histogram manipulation, filtering, thresholding and edge detection are important pre-processing techniques to yield good segmentation results. Many segmentation and classification techniques have been implemented on MRI images. The latest techniques include support vector machines (SVMs), neural networks (NNs), statistical methods, threshold techniques and normalised cuts. Segmentation of bony structures plays an important role in image guided surgery of the spine [1]. Physicians have commonly relied on computed tomography (CT) images to support their decisions in the diagnosis, treatment, and surgery of different pathologies of the spine due to the high resolution and good visualization of bone offered by this medical imaging modality. CT relies on the use of ionizing radiation, and does not depict soft tissue pathology, unlike magnetic resonance imaging (MRI) [1]. While the segmentation of vertebral bodies from CT images Segmentation Of C-Spine MRI Images Using The Watershed Transform Page 6 University of Johannesburg of the spine has commonly been accomplished with seed growing segmentation techniques [1], this task is more difficult in MRI, with variations in soft tissue contrast, and with the RF inhomogeneities, which increase the level of complexity. The primary goal of this project is to develop segmentation techniques for C-spine MRI images. This method will also be compared against other methods like pyramid segmentation and morphological segmentation. The watershed segmentation will be implemented and tested as the final step of the segmentation process. This project will try to use a combination of techniques, rather than to implement and evaluate one single method. It has been learned from literature and also from experience that the pre-processing of the raw data plays a crucial role in the quality of the segmentation process. Therefore, some attention will be given to the pre-processing of the images as part of the segmentation process.
  • Full Text:

Image processing techniques for hazardous weather detection

- Hardy, Caroline Hazel

  • Authors: Hardy, Caroline Hazel
  • Date: 2012-06-05
  • Subjects: Imaging systems in meteorology , Hazardous weather , Remote sensing , Satellite imagery , Image processing , Multispectral image analysis
  • Type: Thesis
  • Identifier: uj:2390 , http://hdl.handle.net/10210/4844
  • Description: M.Ing. , Globally, hazardous weather phenomena such as violent storms, oods, cyclones, tornadoes, snow and hail contribute to signi cant annual xed property damages, loss of movable property and loss of life. The majority of global natural disasters are related to hydro-meteorological events. Hazardous storms are destructive and pose a threat to life and property. Forecasting, monitoring and detecting hazardous storms are complex and demanding tasks, that are however essential. In this study automatic hazardous weather detection utilizing remotely sensed meteorological data has been investigated. Image processing techniques have been analyzed and applied to multispectral meteorological satellite image data obtained from the Spinning Enhanced Visible and Infrared Imager (SEVIRI) instruments on-board the Meteosat Second Generation (MSG) geostationary meteorological satellites Meteosat-8 and Meteosat-9. The primary focus of this study is the detection of potentially hazardous hydrometeorological phenomena in South Africa. A methodology for detecting potentially hazardous storms over South Africa using meteorological satellite imagery from MSG/SEVIRI is presented. An index indicative of the hazardous potential of a storm is de ned to aid in the identi cation of a ected geographical areas and to quantify the destructive potential of the detected storm. The Hazardous Potential Index (HPI) is generated through the use of image processing techniques such as cloud masking, cloud tracking and an image-based analysis of the constituent elements of a severe convective storm. A retrospective review was performed with respect to 20 case studies of documented storms which had adversely a ected areas of South Africa. A red-green-blue (RGB) composite image analysis technique, that may be utilized in the identi cation of severe convective storms using SEVIRI image data, was also applied to these case studies.
  • Full Text:

Comparison of pixel-based and object-oriented classification approaches for detection of camouflaged objects

- Lubbe, Minette

  • Authors: Lubbe, Minette
  • Date: 2012-02-28
  • Subjects: Camouflage (Military science) detection , Image analysis , Remote-sensing images , Image processing
  • Type: Thesis
  • Identifier: uj:2111 , http://hdl.handle.net/10210/4455
  • Description: M.A. , The dissertation topic is the comparison of pixel-based and object-oriented image analysis approaches for camouflaged object detection research. A camouflage field trial experiment was conducted during 2004. For the experiment, 11 military vehicles were deployed along a tree line and in an open field. A subset of the vehicles was deployed with a variety of experimental camouflage nets and a final subset was left uncovered. The reason for deploying the camouflaged objects in the open without the use of camouflage principals was to create a baseline for future measurements. During the next experimental deployment, the camouflaged targets will be deployed according to camouflage principals. It must be emphasised that this is an experimental deployment and not an operational deployment. Unobstructed entity panels were also deployed and served as calibration entities. During the trial, both airborne (colour aerial photography) and space borne (multi-spectral QuickBird) imagery were acquired over the trial sites, and extensive calibration and ground truthing activities were conducted in support of these acquisitions. This study further describes the processing that was done after acquisition of the datasets. The goal is to determine which classification techniques are the most effective in the detection of camouflaged objects. This will also show how well or poor the SANDF camouflage nets and paint potentially perform against air and space based sensors on the one hand and classification techniques on the other. Using this information, DPSS can identify the nets and paints that need to be investigated for future enhancements (e.g. colour selection, colour combinations, base material, camouflage patterns, entity shapes, entity textures, etc.). The classification techniques to be used against SANDF camouflaged objects will also give an indication of their performance against camouflaged advesarial forces in the future.
  • Full Text:

Optiese tegnologie

- Schnaar-Campbell, Annelie

An evaluation of local two-frame dense stereo matching algorithms

- Van der Merwe, Juliaan Werner

  • Authors: Van der Merwe, Juliaan Werner
  • Date: 2012-06-06
  • Subjects: Computer vision , Stereo vision , Depth perception , Image processing , Computer algorithms
  • Type: Thesis
  • Identifier: uj:2512 , http://hdl.handle.net/10210/4966
  • Description: M. Ing. , The process of extracting depth information from multiple two-dimensional images taken of the same scene is known as stereo vision. It is of central importance to the field of machine vision as it is a low level task required for many higher level applications. The past few decades has witnessed the development of hundreds of different stereo vision algorithms. This has made it difficult to classify and compare the various approaches to the problem. In this research we provide an overview of the types of approaches that exist to solve the problem of stereo vision. We focus on a specific subset of algorithms, known as local stereo algorithms. Our goal is to critically analyse and compare a representative sample of local stereo algorithm in terms of both speed and accuracy. We also divide the algorithms into discrete interchangeable components and experiment to determine the effect that each of the alternative components has on an algorithm’s speed and accuracy. We investigate even further to quantify and analyse the effect of various design choices within specific algorithm components. Finally we assemble all of the knowledge gained through the experimentation to compose and optimise a novel algorithm. The experimentation highlighted the fact that by far the most important component of a local stereo algorithm is the manner in which it aggregates matching costs. All of the top performing local stereo algorithms dynamically define the shape of the windows over which the matching costs are aggregated. This is done in a manner that aims to only include pixels in a window that is likely to be at the same depth as the depth of the centre pixel of the window. Since the depth is unknown, the cost aggregation techniques use colour and proximity information to best guess whether pixels are at the same depth when defining the shape of the aggregation windows. Local stereo algorithms are usually less accurate than global methods but they are supposed to be faster and more parallelisable. These cost aggregation techniques result in very accurate depth estimates but unfortunately they are also very expensive computationally. We believe the focus of local stereo algorithm development should be speed. Using the experimental results we developed an algorithm that achieves accuracies in the same order of magnitude as the state-of-the-art algorithms while reducing the computation time by over 50%.
  • Full Text:

Subjective analysis of image coding errors

- El-Hesnawi, Mohamed Rahoma

  • Authors: El-Hesnawi, Mohamed Rahoma
  • Date: 2009-02-26T12:18:47Z
  • Subjects: Visual pathways , Image processing , Image analysis , Algorithms
  • Type: Thesis
  • Identifier: uj:8156 , http://hdl.handle.net/10210/2162
  • Description: D.Ing. , The rapid use of digital images and the necessity to compress them, has created the need for the development of image quality metrics. Subjective evaluation is the most accurate of the image quality evaluation methods, but it is time consuming, tedious and expensive. In the mean time widely used objective evaluations such as the mean squared error measure has proven that they do not assess the image quality the way a human observer does. Since the human observer is the final receiver of most visual information, taking the way humans perceive visual information will be greatly beneficial for the development of an objective image quality metric that will reflect the subjective evaluation of distorted images. Many attempts have been carried out in the past, which tried to develop distortion metrics that model the processes of the human visual system, and many promising results have been achieved. However most of these metrics were developed with the use of simple visual stimuli, and most of these models were based on the visibility threshold measures, which are not representative of the distortion introduced in complex natural compressed images. In this thesis, a new image quality metric based on the human visual system properties as related to image perception is proposed. This metric provides an objective image quality measure for the subjective quality of coded natural images with suprathreshold degradation. This proposed model specifically takes into account the structure of the natural images, by analyzing the images into their different components, namely: the edges, texture and background (smooth) components, as these components influence the formation of perception in the HVS differently. Hence the HVS sensitivity to errors in images depends on weather these errors lie in more active areas of the image, such as strong edges or texture, or in the less active areas such as the smooth areas. These components are then summed to obtain the combined image which represents the way the HVS is postulated to perceive the image. Extensive subjective evaluation was carried out for the different image components and the combined image, obtained for the coded images at different qualities. The objective (RMSE) for these images was also calculated. A transformation between the subjective and the objective quality measures was performed, from which the objective metric that can predict the human perception of image quality was developed. The metric was shown to provide an accurate prediction of image quality, which agrees well with the prediction provided by the expensive and lengthy process of subjective evaluation. Furthermore it has the desired properties of the RMSE of being easier and cheaper to implement. Therefore, this metric will be useful for evaluating error mechanisms present in proposed coding schemes.
  • Full Text:

Polar rectification of stereo images implemented on a GPU

- Nel, A. L.

The effects of evaluation and rotation on descriptors and similarity measures for a single class of image objects

- Loots, Conrad

  • Authors: Loots, Conrad
  • Date: 2008-06-06T10:32:17Z
  • Subjects: Multimedia systems , Information storage and retrieval systems , JPEG (image coding standard) , MPEG (video coding standard) , Image processing
  • Type: Thesis
  • Identifier: uj:9199 , http://hdl.handle.net/10210/564
  • Description: “A picture is worth a thousand words”. If this proverb were taken literally we all know that every person interprets images or photos differently in terms of its content. This is due to the semantics contained in these images. Content-based image retrieval has become a vast area of research in order to successfully describe and retrieve images according to the content. In military applications, intelligence images such as those obtained by the defence intelligence group are taken (mostly on film), developed and then manually annotated thereafter. These photos are then stored in a filing system according to certain attributes such as the location, content etc. To retrieve these images at a later stage might take days or even weeks to locate. Thus, the need for a digital annotation system has arisen. The images of the military contain various military vehicles and buildings that need to be detected, described and stored in a database. For our research we want to look at the effects that the rotation and elevation angle of an object in an image has on the retrieval performance. We chose model cars in order to be able to control the environment the photos were taken in such as the background, lighting, distance between the objects, and the camera etc. There are also a wide variety of shapes and colours of these models to obtain and work with. We look at the MPEG-7 descriptor schemes that are recommended by the MPEG group for video and image retrieval as well as implement three of them. For the military it could be required that when the defence intelligence group is in the field, that the images be directly transmitted via satellite to the headquarters. We have therefore included the JPEG2000 standard which gives a compression performance increase of 20% over the original JPEG standard. It is also capable to transmit images wirelessly as well as securely. Including the MPEG-7 descriptors that we have implemented, we have also implemented the fuzzy histogram and colour correlogram descriptors. For our experimentation we implemented a series of experiments in order to determine the effects that rotation and elevation has on our model vehicle images. Observations are made when each vehicle is considered separately and when the vehicles are described and combined into a single database. After the experiments are done we look at the descriptors and determine which adjustments could be made in order to improve the retrieval performance thereof. , Dr. W.A. Clarke
  • Full Text:
  • 1
  • Disclaimer
  • Privacy
  • Copyright
  • Contact