No description available.

This thesis studies the assessment of picture quality using the artificial neural network approach. In the first part, two main ways to evaluate the picture quality are described. It is the subjective assessment of picture quality, where a group of people watches the picture and evaluates its quality, and objective assessment which is based on mathematical relations. Calculation of structural similarity index (SSIM) is analyzed in detail. In the second part, the basis of neural networks is described. A neural network was created in Matlab, designed to simulate subjective assessment scores based on the SSIM index.

This thesis deals with quality of digital images and with methods for measurement their numerical characteristics. Our attention is drawn to measurement of sharpness, as a main factor of image quality, in both space and spectral domain. For this reason, the Fourier and the discrete Fourier transform is described in one and two dimensions. Methods, presented in this work and implemented on attached CD, can be used for automatic image quality classification and autofocus of optical systems.

Research conducted in the aviation industry includes two major areas, increased safety and a reduction of the environmental footprint. This thesis investigates the possibilities of increased situational awareness with computer vision in avionics systems. Image fusion methods are evaluated with appropriate pre-processing of three image sensors, one in the visual spectrum and two in the infra-red spectrum. The sensor setup is chosen to cope with the different weather and operational conditions of an aircraft, with a focus on the final approach and landing phases. Extensive image quality assessment metrics derived from a systematic review is applied to provide a precise evaluation of the image quality of the fusion methods. A total of four image fusion methods are evaluated, where two are convolutional network-based, using the networks for feature extraction in the detailed layers. Other approaches with visual saliency maps and sparse representation are also evaluated. With methods implemented in MATLAB, results show that a conventional method implementing a rolling guidance filter for layer separation and visual saliency map provides the best results. The results are further confirmed with a subjective ranking test, where the image quality of the fusion methods is evaluated further.

This research focuses on developing algorithms to automatically classify, detect, simulate and improve the quality of defective printed images since the human visual system is unreliable. With the development of deep learning algorithms, state-of-the-art accuracy could be achieved for many computer vision tasks. This research applies the deep learning method to printed image quality assessment. Because most deep learning approaches require a large amount of data even after data augmentation, we propose to use Generative Adversarial Networks for simulation images generation. The simulated images with artifacts could be used for training classifier, detector and corrector networks for printed image quality. Another essential preprocessing step for printed image quality assessment is image registration, which can detect the defect and difference between two input images. This research proposes to use the deep learning framework for global image registration by parallel computation acceleration. For deformable local registration, we implement the U-Net VoxelMorph-based method for printed image registration. Then we further propose the recurrent network-based method, R-RegNet. The experimental results show that the proposed R-RegNet method outperforms the U-Net VoxelMorph-based method in all three datasets that we considered. Finally, we propose a photorealistic image dataset simulation method for training deep neural networks. A new dataset with simulated images, named Extra FAT, is introduced for object detection and 6D pose estimation.

abstract: The depth richness of a scene translates into a spatially variable defocus blur in the acquired image. Blurring can mislead computational image understanding; therefore, blur detection can be used for selective image enhancement of blurred regions and the application of image understanding algorithms to sharp regions. This work focuses on blur detection and its application to image enhancement. This work proposes a spatially-varying defocus blur detection based on the quotient of spectral bands; additionally, to avoid the use of computationally intensive algorithms for the segmentation of foreground and background regions, a global threshold defined using weak textured regions on the input image is proposed. Quantitative results expressed in the precision-recall space as well as qualitative results overperform current state-of-the-art algorithms while keeping the computational requirements at competitive levels. Imperfections in the curvature of lenses can lead to image radial distortion (IRD). Computer vision applications can be drastically affected by IRD. This work proposes a novel robust radial distortion correction algorithm based on alternate optimization using two cost functions tailored for the estimation of the center of distortion and radial distortion coefficients. Qualitative and quantitative results show the competitiveness of the proposed algorithm. Blur is one of the causes of visual discomfort in stereopsis. Sharpening applying traditional algorithms can produce an interdifference which causes eyestrain and visual fatigue for the viewer. A sharpness enhancement method for stereo images that incorporates binocular vision cues and depth information is presented. Perceptual evaluation and quantitative results based on the metric of interdifference deviation are reported; results of the proposed algorithm are competitive with state-of-the-art stereo algorithms. Digital images and videos are produced every day in astonishing amounts. Consequently, the market-driven demand for higher quality content is constantly increasing which leads to the need of image quality assessment (IQA) methods. A training-free, no-reference image sharpness assessment method based on the singular value decomposition of perceptually-weighted normalized-gradients of relevant pixels in the input image is proposed. Results over six subject-rated publicly available databases show competitive performance when compared with state-of-the-art algorithms. / Dissertation/Thesis / Doctoral Dissertation Electrical Engineering 2019

Dark current is an unwanted source of noise in images produced by digital imagers, the de facto standard of imaging. The two most common types of digital imager architectures, Charged-Coupled Devices (CCDs) and Complementary Metal-Oxide-Semiconductor (CMOS), are both prone to this noise source. To accurately reflect the information from light signals this noise must be removed. This practice is especially vital for scientific purposes such as in astronomical observations. Presented in this dissertation are characterizations of dark current sources that present complications to the traditional methods of correction. In particular, it is observed that pixels in both CCDs and CMOS image sensors produce dark current that is affected by the presence of pre-illuminating the sensor and that these same pixels produce a nonlinear dark current with respect to exposure time. These two characteristics are not conventionally accounted for as it is assumed that the dark current produced will be unaffected by charge accumulated from either illumination or the dark current itself. Additionally, a model reproducing these dark current characteristics is presented. The model incorporates a moving edge of the depletion region, where charge is accumulated, as well as fixed recombination-generation locations. Recombination-generation sites in the form of heavy metal impurities, or lattice defects, are commonly the source of dark current especially in the highest producing pixels, commonly called "hot pixels." The model predicts that pixels with recombination-generation sites near the edge of an empty depletion region will produce less dark current after accumulation of charge, accurately modeling the behavior observed from empirical sources. Finally, it is shown that activation energy calculations will produce inconsistent results for pixels with the presence of recombination-generation sites near the edge of a moving depletion region. Activation energies, an energy associated with the temperature dependence of dark current, are often calculated to characterize aspects of the dark current including types of impurities and sources of dark current. The model is shown to generate data, including changing activation energy values, that correspond with changing activation energy calculations in those pixels observed to be affected by pre-illumination and that produce inconsistent dark current over long exposure times. Rather than only being a complication to dark current correction, the presence of such pixels, and the model explaining their behavior, presents an opportunity to obtain information, such as the depth of these recombination-generation sites, which will aid in refining manufacturing processes for digital imagers.

No description available.

No description available.

No description available.