abstract: Detection of extruded features like rooftops and trees in aerial images automatically is a very active area of research. Elevated features identified from aerial imagery have potential applications in urban planning, identifying cover in military training or flight training. Detection of such features using commonly available geospatial data like orthographic aerial imagery is very challenging because rooftop and tree textures are often camouflaged by similar looking features like roads, ground and grass. So, additonal data such as LIDAR, multispectral imagery and multiple viewpoints are exploited for more accurate detection. However, such data is often not available, or may be improperly registered or inacurate. In this thesis, we discuss a novel framework that only uses orthographic images for detection and modeling of rooftops. A segmentation scheme that initializes by assigning either foreground (rooftop) or background labels to certain pixels in the image based on shadows is proposed. Then it employs grabcut to assign one of those two labels to the rest of the pixels based on initial labeling. Parametric model fitting is performed on the segmented results in order to create a 3D scene and to facilitate roof-shape and height estimation. The framework can also benefit from additional geospatial data such as streetmaps and LIDAR, if available. / Dissertation/Thesis / M.S. Computer Science 2013

A brain tumor is detected and classified by biopsy that is conducted after the brain surgery. Advancement in technology and machine learning techniques could help radiologists in the diagnosis of tumors without any invasive measures. We utilized a deep learning-based approach to detect and classify the tumor into Meningioma, Glioma, Pituitary tumors. We used registration and segmentation-based skull stripping mechanism to remove the skull from the MRI images and the grab cut method to verify whether the skull stripped MRI masks retained the features of the tumor for accurate classification. In this research, we proposed a transfer learning based approach in conjunction with discriminative learning rates to perform the classification of brain tumors. The data set used is a 3064 T MRI images dataset that contains T1 flair MRI images. We achieved a classification accuracy of 98.83%, 96.26%, and 95.18% for training, validation, and test sets and an F1 score of 0.96 on the T1 Flair MRI dataset.

This thesis describes design and implementation of system that allows batch generation of graphical presentations. The system also includes modules for image quality evaluation using no-reference blur metric and salient object detection. Selected methods for evaluation of image quality are described in detail and implemented in corresponding chapters, including proposed modifications and changes. Blur detection is based on wavelet transform, and salient object detection is achieved by investigating image contrast. Capabilities of these modules are evaluated on suitable image datasets.

This thesis aims to present a method for detection of faults (burn marks) on insulator using only image processing algorithms. It is accomplished by extracting the insulator from the background image and then detecting the burn marks on the segmented image. Apart from several other challenges encountered during the detection phase, the main challenge was to eliminate the connector marks which might be detected as burn-marks. The technique discussed in this thesis work is one of a kind and not much research has been done in areas of burn mark detection on the insulator surface. Several algorithms have been pondered upon before coming up with a set of algorithms applied in a particular manner. The first phase of the work emphasizes on detection of the insulator from the image. Apart from pre-processing and other segmentation techniques, Symmetry detection and adaptive GrabCut are the main algorithms used for this purpose. Efficient and powerful algorithms such as feature detection and matching were considered before arriving at this method, based on pros and cons. The second phase is the detection of burn marks on the extracted image while eliminating the connector marks. Algorithms such as Blob detection and Contour detection, adapted in a particular manner, have been used for this purpose based on references from medical image processing. The elimination of connector marks is obtained by applying a set of mathematical calculations. The entire project is implemented in Visual Studio using OpenCV libraries. Result obtained is cross-validated across an image data set.