The use of the genetic algorithm for shape recognition has been investigated in relation to features along a shape boundary contour. Various methods for encoding chromosomes were investigated, the most successful of which led to the development of a new technique to input normalised 'perceptually important point' features from the contour into a genetic algorithm. Chromosomes evolve with genes defining various ways of 'observing' different parts of the contour. The normalisation process provides the capability for multi-scale spatial frequency filtering and fine/coarse resolution of the contour features. A standard genetic algorithm was chosen for this investigation because its performance can be analysed by applying schema analysis to the genes. A new method for measurement of gene diversity has been developed. It is shown that this diversity measure can be used to direct the genetic algorithm parameters to evolve a number of 'good' chromosomes. In this way a variety of sections along the contour can be observed. A new and effective recognition technique has been developed which makes use of these 'good' chromosomes and the same fitness calculation as used in the genetic algorithm. Correct recognition can be achieved by selecting chromosomes and adjusting two thresholds, the values of which are found not to be critical. Difficulties associated with the calculation of a shape's fitness were analysed and the structure of the genes in the chromosome investigated using schema and epistatic analysis. It was shown that the behaviour of the genetic algorithm is compatible with the schema theorem of J. H. Holland. Reasons are given to explain the minimum value for the mutation probability that is required for the evolution of a number of' good' chromosomes. Suggestions for future research are made and, in particular, it is recommended that the convergence properties of the standard genetic algorithm be investigated.

我们提出了两个新的方法来解决低级别计算机视觉任务，即图像共同分割和降噪。 / 在共同分割模型上，我们发现对象对应可以为前景统计估计提供有用的信息。我们的方法可以处理极具挑战性的场景，如变形，角度的变化和显着不同的视角和尺度。此外，我们研究了一种新的能量最小化模型，可以同时处理多个图像。真实和基准数据的定性和定量实验证明该方法的有效性。 / 另一方面，噪音始终和高频图像结构是紧耦合的，从而使得减少噪音非常很难。在我们的降噪模型中，我们建议稍微使图像光学离焦，以减少图像和噪声的耦合。这使得我们能更有效地降低噪音，随后恢复失焦。我们的分析显示，这是可能的，并且用许多例子证明我们的技术，其中包括低光图像。 / We present two novel methods to tackle low level computer vision tasks,i.e., image cosegmentation and denoise . / In our cosegmentationmodel, we discover object correspondence canprovide useful information for foreground statistical estimation. Ourmethod can handle extremely challenging scenarios such as deformation, perspective changes and dramatically different viewpoints/scales. In addition, we develop a novel energy minimization model that can handlemultiple images. Experiments on real and benchmark data qualitatively and quantitatively demonstrate the effectiveness of the approach. / One the other hand, noise is always tightly coupled with high-frequencyimage structure, making noise reduction generally very difficult. In ourdenoise model, we propose slightly optically defocusing the image in orderto loosen this noise-image structure coupling. This allows us to more effectively reduce noise and subsequently restore the small defocus. Weanalytically show how this is possible, and demonstrate our technique on a number of examples that include low-light images. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Qin, Zenglu. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 64-71). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts also in Chinese. / Abstract --- p.i / Acknowledgement --- p.ii / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Motivation and Objectives --- p.1 / Chapter 1.1.1 --- Cosegmentation --- p.1 / Chapter 1.1.2 --- Image Denoise --- p.4 / Chapter 1.2 --- Thesis Outline --- p.7 / Chapter 2 --- Background --- p.8 / Chapter 2.1 --- Cosegmentation --- p.8 / Chapter 2.2 --- Image Denoise --- p.10 / Chapter 3 --- Cosegmentation of Multiple Deformable Objects --- p.12 / Chapter 3.1 --- Related Work --- p.12 / Chapter 3.2 --- Object Corresponding Cosegmentation --- p.13 / Chapter 3.3 --- Importance Map with Object Correspondence --- p.15 / Chapter 3.3.1 --- Feature Importance Map --- p.16 / Chapter 3.3.2 --- Importance Energy E[subscript i](xp) --- p.20 / Chapter 3.4 --- Experimental Result --- p.20 / Chapter 3.4.1 --- Two-Image Cosegmentation --- p.21 / Chapter 3.4.2 --- ETHZ Toys Dataset --- p.22 / Chapter 3.4.3 --- More Results --- p.24 / Chapter 3.5 --- Summary --- p.27 / Chapter 4 --- Using Optical Defocus to Denoise --- p.28 / Chapter 4.1 --- Related Work --- p.29 / Chapter 4.2 --- Noise Analysis --- p.30 / Chapter 4.3 --- Noise Estimation with Focal Blur --- p.33 / Chapter 4.3.1 --- Noise Estimation with a Convolution Model --- p.33 / Chapter 4.3.2 --- Determining λ --- p.41 / Chapter 4.4 --- Final Deconvolution and Error Analysis --- p.43 / Chapter 4.5 --- Implementation --- p.45 / Chapter 4.6 --- Quantitative Evaluation --- p.47 / Chapter 4.7 --- More Experimental Results --- p.53 / Chapter 4.8 --- Summary --- p.56 / Chapter 5 --- Conclusion --- p.62 / Bibliography --- p.64

Wen, Chengfeng. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2013. / Includes bibliographical references (leaves 77-83). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts also in Chinese.

The data used by a computer to represent and display a picture conventionally requires very large amounts of dedicated storage. As the trend continues towards high resolution graphics employing a wide range of colours the associated volume of data often expands to a point where sophisticated techniques become essential to maintain a practical interactive system. Considerable research has been conducted into coding schemes which compress the volume of picture data. These methods are clearly of great value in reducing the delay which results from transmitting picture data between sites. Codes which also serve directly as a source for real-time display are better still. The quadtree method of picture encoding, which has previously been successfully employed in the field of picture processing, is shown in this thesis to provide the basis for an integrated approach to data compression in the field of computer graphics. The display of colour pictures from quadtree code is performed by customised hardware which operates in conjunction with a general purpose processor. A prototype system is described which provides efficient downloading of pictures from a host and rapid display of locally stored pictures. The amenability of quadtree code to manipulation is demonstrated by incorporating a limited capability to pan and zoom on a picture. Application of the method to picture archive retrieval is examined.

Typescript (photocopy). / Digitized by Kansas Correctional Industries

Frequently, testing of Single Photon Emission Computed Tomography (SPECT) motion correction algorithms is done either by using simplistic deformations that do not accurately simulate true patient motion or by applying the algorithms directly to data acquired from a real patient, where the true internal motion is unknown. In this work, we describe a way to combine these two approaches by using imaging data acquired from real volunteers to simulate the data that the motion correction algorithms would normally observe. The goal is to provide an assessment framework which can both: simulate realistic SPECT acquisitions that incorporate realistic body deformations and provide a ground truth volume to compare against. Every part of the motion correction algorithm needs to be exercised: from parameter estimation of the motion model, to the final reconstruction results. In order to build the ground truth anthropomorphic numerical phantoms, we acquire high resolution MRI scans and motion observation data of a volunteer in multiple different configurations. We then extract the organ boundaries using thresholding, active contours, and morphology. Phantoms of radioactivity uptake and density inside the body can be generated from these boundaries to be used to simulate SPECT acquisitions. We present results on extraction of the ribs, lungs, heart, spine, and the rest of the soft tissue in the thorax using our segmentation approach. In general, extracting the lungs, heart, and ribs in images that do not contain the spine works well, but the spine could be better extracted using other methods that we discuss. We also go in depth into the software development component of this work, describing the C++ coding framework we used and the High Level Interactive GUI Language (HLING). HLING solved a lot of problems but introduced a fair bit of its own. We include a set of requirements to provide a foundation for the next attempt at developing a declarative and minimally restrictive methodology for writing interactive image processing applications in C++ based on lessons learned during the development of HLING.

Thesis (B.S.)--Massachusetts Institute of Technology, Dept. of Mathematics, 1980. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND SCIENCE. / Bibliography: leaves 50-57. / by Paul S. Heckbert. / B.S.

We introduce the use of describable visual attributes for face images. Describable visual attributes are labels that can be given to an image to describe its appearance. This thesis focuses mostly on images of faces and the attributes used to describe them, although the concepts also apply to other domains. Examples of face attributes include gender, age, jaw shape, nose size, etc. The advantages of an attribute-based representation for vision tasks are manifold: they can be composed to create descriptions at various levels of specificity; they are generalizable, as they can be learned once and then applied to recognize new objects or categories without any further training; and they are efficient, possibly requiring exponentially fewer attributes (and training data) than explicitly naming each category. We show how one can create and label large datasets of real-world images to train classifiers which measure the presence, absence, or degree to which an attribute is expressed in images. These classifiers can then automatically label new images. We demonstrate the current effectiveness and explore the future potential of using attributes for image search, automatic face replacement in images, and face verification, via both human and computational experiments. To aid other researchers in studying these problems, we introduce two new large face datasets, named FaceTracer and PubFig, with labeled attributes and identities, respectively. Finally, we also show the effectiveness of visual attributes in a completely different domain: plant species identification. To this end, we have developed and publicly released the Leafsnap system, which has been downloaded by almost half a million users. The mobile phone application is a flexible electronic field guide with high-quality images of the tree species in the Northeast US. It also gives users instant access to our automatic recognition system, greatly simplifying the identification process.

Yu Ying-Kin. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2004. / Includes bibliographical references (leaves 108-116). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgements --- p.iv / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Motivation and Objectives --- p.1 / Chapter 1.2 --- Problem Definition --- p.3 / Chapter 1.3 --- Contributions --- p.6 / Chapter 1.4 --- Related Publications --- p.8 / Chapter 1.5 --- Organization of the Paper --- p.9 / Chapter 2 --- Background --- p.11 / Chapter 2.1 --- Introduction --- p.11 / Chapter 2.2 --- Pose Estimation --- p.12 / Chapter 2.2.1 --- Overview --- p.12 / Chapter 2.2.2 --- Lowe's Method --- p.14 / Chapter 2.2.3 --- The Genetic Algorithm by Hati and Sen- gupta --- p.15 / Chapter 2.3 --- Structure and Motion --- p.17 / Chapter 2.3.1 --- Overview --- p.17 / Chapter 2.3.2 --- The Extended Lowe's Method --- p.20 / Chapter 2.3.3 --- The Extended Kalman Filter by Azarbaye- jani and Pentland --- p.23 / Chapter 3 --- Model-based Pose Tracking Using Genetic Algo- rithms --- p.27 / Chapter 3.1 --- Introduction --- p.27 / Chapter 3.2 --- Overview of the Algorithm --- p.28 / Chapter 3.3 --- Chromosome Encoding --- p.29 / Chapter 3.4 --- The Genetic Operators --- p.30 / Chapter 3.4.1 --- Mutation --- p.30 / Chapter 3.4.2 --- Crossover --- p.31 / Chapter 3.5 --- Fitness Evaluation --- p.31 / Chapter 3.6 --- The Roulette Wheel Proportionate Selection Scheme --- p.32 / Chapter 3.7 --- The Genetic Algorithm Parameters --- p.33 / Chapter 3.8 --- Experiments and Results --- p.34 / Chapter 3.8.1 --- Synthetic Data Experiments --- p.34 / Chapter 3.8.2 --- Real Scene Experiments --- p.38 / Chapter 4 --- Recursive 3D Structure Acquisition Based on Kalman Filtering --- p.42 / Chapter 4.1 --- Introduction --- p.42 / Chapter 4.2 --- Overview of the Algorithm --- p.43 / Chapter 4.2.1 --- Feature Extraction and Tracking --- p.44 / Chapter 4.2.2 --- Model Initialization --- p.44 / Chapter 4.2.3 --- Structure and Pose Updating --- p.45 / Chapter 4.3 --- Structure Updating --- p.46 / Chapter 4.4 --- Pose Estimation --- p.49 / Chapter 4.5 --- Handling of the Changeable Set of Feature Points --- p.52 / Chapter 4.6 --- Analytical Comparisons with Other Algorithms --- p.54 / Chapter 4.6.1 --- Comparisons with the Interleaved Bundle Adjustment Method --- p.54 / Chapter 4.6.2 --- Comparisons with the EKF by Azarbaye- jani and Pentland --- p.56 / Chapter 4.7 --- Experiments and Results --- p.57 / Chapter 4.7.1 --- Synthetic Data Experiments --- p.57 / Chapter 4.7.2 --- Real Scene Experiments --- p.58 / Chapter 5 --- Simultaneous Pose Tracking and Structure Acqui- sition Using the Interacting Multiple Model --- p.63 / Chapter 5.1 --- Introduction --- p.63 / Chapter 5.2 --- Overview of the Algorithm --- p.65 / Chapter 5.2.1 --- Feature Extraction and Tracking --- p.65 / Chapter 5.2.2 --- Model Initialization --- p.66 / Chapter 5.2.3 --- Structure and Pose Updating --- p.66 / Chapter 5.3 --- Pose Estimation --- p.67 / Chapter 5.3.1 --- The Interacting Multiple Model Algorithm --- p.67 / Chapter 5.3.2 --- Design of the Individual EKFs --- p.71 / Chapter 5.4 --- Structure Updating --- p.74 / Chapter 5.5 --- Handling of the Changeable Set of Feature Points --- p.76 / Chapter 5.6 --- Analytical Comparisons with Other EKF-Based Algorithms --- p.77 / Chapter 5.6.1 --- Computation Speed --- p.77 / Chapter 5.6.2 --- Accuracy of the Recovered Pose Sequences --- p.79 / Chapter 5.7 --- Experiments and Results --- p.80 / Chapter 5.7.1 --- Synthetic Data Experiments --- p.80 / Chapter 5.7.2 --- Real Scene Experiments --- p.80 / Chapter 6 --- Empirical Comparisons of the Structure and Mo- tion Algorithms --- p.87 / Chapter 6.1 --- Introduction --- p.87 / Chapter 6.2 --- Comparisons Using Synthetic Data --- p.88 / Chapter 6.2.1 --- Image Residual Errors --- p.88 / Chapter 6.2.2 --- Computation Efficiency --- p.89 / Chapter 6.2.3 --- Accuracy of Recovered Pose Sequences . . --- p.91 / Chapter 6.3 --- Comparisons Using Real Images --- p.92 / Chapter 6.4 --- Summary --- p.97 / Chapter 7 --- Future Work --- p.99 / Chapter 8 --- Conclusion --- p.101 / Chapter A --- Kalman Filtering --- p.103 / Bibliography --- p.107

Our method can be naturally applied in the application of manga stylization rendering. We propose a novel colorization technique that propagates color over manga regions exhibiting pattern-continuity as well as intensity-continuity. The proposed method works effectively on colorizing black-and-white manga which contains intensive amount of strokes, hatching, halftoning and screening. Once the user scribbles on the drawing, a local, statistical based pattern feature obtained with Gabor wavelet filters is applied to measure the pattern-continuity. The boundary is then propagated by the level set method that monitors the pattern-continuity. Regions with open boundaries or multiple disjointed regions with similar patterns can be sensibly segmented by a single scribble. With the segmented regions, various colorization techniques can be applied to replace colors, colorize with stroke preservation, or even convert pattern to shading. Based on the observation of the manga features, we then propose a framework to generate manga-style backgrounds from real photographs. It frees manga artists from the tedious and time-consuming background production. To mimic how manga artists produce the tidy background, it composes of two major steps, the line drawing and the screen laying. A line importance model is proposed to simplify and embolden lines in a stylish way. During the screen laying, texture analysis is utilized to automatically match the regions in photographs with screens in the database. Two types of screening mechanisms are proposed for matching shading & texture as well as the high-level structures. / Segmentation has always been a crucial and challenging topic in various computer vision and graphics applications. Information from image features can be incorporated into image segmentation techniques to make the process more accurate and robust. This dissertation focuses on encoding these feature information into the image processing procedures, including image segmentation and manga stylization rendering. / The fundamental part of this dissertation includes the discussion on the construction of the speed function, which is important in applying the curve-evolution based image segmentation. We firstly investigate the curvature term in the speed function, and then show how to transform the image segmentation problem into an interface propagating problem. We propose two formulations to enhance the speed function in level set methods, in order to tackle the segmentation problem of tagged MR images. First, a relaxation factor is introduced, aiming at relaxing the boundary condition when the boundary is unclear or blurry. Second, in order to incorporate human visual sensitive information from the image, a simple and general model is introduced to incorporate shape, texture and color features. By further extending this model, we present a unified approach for segmenting and tracking of the high-resolution color anatomical Chinese Visible Human (CVH) data. The underlying relationship of these two applications relies on the proposed variational framework for the speed function. Our proposed method can be used to segment the first slice of the volume data. Then based on the extracted boundary on the first slice, our method can also be adapted to track the boundary of the homogeneous organs among the subsequent serial images. In addition to the promising segmentation results, the tracking procedure requires only a small amount of user intervention. / Qu Yingge. / "July 2007." / Advisers: Heng Pheng Ann; Wong Tien Tsin. / Source: Dissertation Abstracts International, Volume: 69-01, Section: B, page: 0433. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (p. 107-120). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract in English and Chinese. / School code: 1307.