Useful information extraction from satellite images for the use of other higher level applications such as road network extraction and update, city planning etc. is a very important and active research area. It is seen that pixel-based techniques becomes insufficient for this task with increasing spatial resolution of satellite imaging sensors day by day. Therefore, the use of object-based techniques becomes indispensable and the segmentation method selection is very crucial for object-based techniques. In this thesis, various segmentation algorithms applied in remote sensing literature are presented and a segmentation process that is based on watersheds and multi-scale segmentation is proposed to use as the segmentation step of an object-based classifier. For every step of the proposed segmentation process, qualitative and quantitative comparisons with alternative approaches are done. The ones which provide best performance are incorporated into the proposed algorithm. Also, an unsupervised segmentation accuracy metric to determine all parameters of the algorithm is proposed. By this way, the proposed segmentation algorithm has become a fully automatic approach. Experiments that are done on a database formed with images taken from Google Earth&reg / software provide promising results.

At present, the quality of image taken from infrared cameras is low compared to the other cameras because of manufacturing technology. So, resolution enhancement processes are becoming more important for these cameras. Super resolution is a good approach to solve this resolution problem. In general, the systems that infrared cameras used require video processing to perform in real time. So, a suitable approach should be selected and implemented to work in real time. The computational load and processing time are big issues in this case. FPGAs are proven to be suitable hardware devices for these types of works. Super resolution involves two parts as global motion estimation and high resolution image reconstruction. In this study, one suitable algorithm, namely as PM, for global motion estimation in the literature is selected to be implemented in real time. On the other hand, for high resolution image reconstruction part, FPGA structures of some well known algorithms in the literature, namely as POCS, MLE, MAP and LMS are proposed and their performance, resource requirements and timing considerations are discussed. Most efficient one is selected and implemented in FPGA.

Road Networks play an important role in various applications such as urban and rural planning, infrastructure planning, transportation management, vehicle navigation. Extraction of Roads from Remote Sensed satellite images for updating road database in geographical information systems (GIS) is generally done manually by a human operator. However, manual extraction of roads is time consuming and labor intensive process. In the existing literature, there are a great number of researches published for the purpose of automating the road extraction process. However, automated processes still yield some erroneous and incomplete results and human intervention is still required. The aim of this research is to propose a framework for road network extraction from high spatial resolution multi-spectral imagery (MSI) to improve the accuracy of road extraction systems. The proposed framework begins with a spectral classification using One-class Support Vector Machines (SVM) and Gaussian Mixture Models (GMM) classifiers. Spectral Classification exploits the spectral signature of road surfaces to classify road pixels. Then, an iterative template matching filter is proposed to refine spectral classification results. K-medians clustering algorithm is employed to detect candidate road centerline points. Final road network formation is achieved by Markov Random Fields. The extracted road network is evaluated against a reference dataset using a set of quality metrics.

The share of peer-to-peer (P2P) protocol in the total network traffic grows dayby- day in the Turkish Academic Network (UlakNet) similar to the other networks in the world. This growth is mostly because of the popularity of the shared content and the great enhancement in the P2P protocol since it first came out with Napster. The shared files are generally both large and copyrighted. Motivated by the problems of UlakNet with the P2P traffic, we propose a novel method for P2P traffic detection in the network backbone in this thesis. Observing the similarity between detecting traffic that belongs to a specific protocol and detecting an intrusion in a computer system, we adopt an Intrusion Detection System (IDS) technique to detect P2P traffic. Our method is a passive detection procedure that uses traffic flows gathered from border routers. Hence, it is scalable and does not have the problems of other approaches that rely on packet payload data or transport layer ports.

This study proposes an industrialization model, Software Factory Automation, for establishing software product lines. Major contributions of this thesis are the conceptualization of Domain Specific Kits (DSKs) and a domain design model for software product lines based on DSKs. The concept of DSK has been inspired by the way other industries have been successfully realizing factory automation for decades. DSKs, as fundamental building blocks, have been deeply elaborated with their characteristic properties and with several examples. The constructed domain design model has two major activities: first, building the product line reference architecture using DSK abstraction / and second, constructing reusable asset model again based on DSK concept. Both activities depend on outputs of feature-oriented analysis of product line domain. The outcome of these coupled modeling activities is the reference architecture and asset model of the product line. The approach has been validated by constructing software product lines for two product families. The reusability of DSKs and software assets has also been discussed with examples. Finally, the constructed model has been evaluated in terms of quality improvements, and it has been compared with other software product line engineering approaches.

Motion estimation and compensation plays an important role in video processing applications. Two-dimensional block-based and mesh-based models are widely used in this area. A 2-D mesh-based model provides a better representation of complex real world motion than a block-based model. Mesh-based motion estimation algorithms are employed in both frame-based and object-based video compression and coding. A hierarchical mesh-based algorithm is applied to improve the motion field generated by a single-layer algorithm. 2-D mesh-based models also enable the manipulation of video objects which is included in the MPEG-4 standard. A video object in a video clip can be replaced by another object by the use of a dynamic mesh structure. In this thesis, a comparative analysis of 2-D block-based and mesh-based motion estimation algorithms in both frame-based and object-based video representations is performed. The experimental results indicate that a mesh-based algorithm produces better motion compensation results than a block-based algorithm. Moreover, a two-layer mesh-based algorithm shows improvement over a one-layer mesh-based algorithm. The application of mesh-based motion estimation and compensation to video object replacement and animation is also performed.

As the networks become faster and faster, the emerging requirement is to improve the performance of the Intrusion Detection and Prevention Systems (IDPS) to keep up with the increased network throughput. In high speed networks, it is very difficult for the IDPS to process all the packets. Since the throughput of IDPS is not improved as fast as the throughput of the switches and routers, it is necessary to develop new detection techniques other than traditional techniques. In this thesis we propose a rule-based IDPS technique to detect Layer 2-4 attacks by just examining the flow data without inspecting packet payload. Our approach is designed to work as an additional component to existing IDPS as we acknowledge that the attacks at Layer 5 and above require payload inspection. The rule set is constructed and tested on a real network to evaluate the performance of the system.

In this thesis, we address the problem of panoramic scene construction in which a single image covering the entire visible area of the scene is constructed from an aerial image video. In the literature, there are several algorithms developed for construction of panoramic scene of a video sequence. These algorithms can be categorized as feature based and featureless algorithms. In this thesis, we concentrate on the feature based algorithms and comparison of these algorithms is performed for aerial videos. The comparison is performed on video sequences captured by non-stationary cameras, whose optical axis does not have to be the same. In addition, the matching and tracking performances of the algorithms are separately analyzed, their advantages-disadvantages are presented and several modifications are proposed.

A feature of minimizing images of a class of binary Markov random field energies is introduced and proved. Using this, the collection of minimizing images of levels of higher order, levelable MRF energies is shown to be a monotone collection. This implies that these images can be combined to give minimizing images of the MRF energy itself. Due to the recent developments, second and third order binary MRF energies of the mentioned class are known to be exactly minimized by maximum flow/minimum cut computations on appropriately constructed graphs. With the aid of these developments an exact and efficient algorithm to minimize levelable second and third order MRF energies, which is composed of a series of maximum flow/minimum cut computations, is proposed and applications of the proposed algorithm to image restoration are given.

In this thesis, a pipeline for structure-from-motion with mixed camera types is described and methods for the steps of this pipeline to make it effective and automatic are proposed. These steps can be summarized as calibration, feature point matching, epipolar geometry and pose estimation, triangulation and bundle adjustment. We worked with catadioptric omnidirectional and perspective cameras and employed the sphere camera model, which encompasses single-viewpoint catadioptric systems as well as perspective cameras. For calibration of the sphere camera model, a new technique that has the advantage of linear and automatic parameter initialization is proposed. The projection of 3D points on a catadioptric image is represented linearly with a 6x10 projection matrix using lifted coordinates. This projection matrix is computed with an adequate number of 3D-2D correspondences and decomposed to obtain intrinsic and extrinsic parameters. Then, a non-linear optimization is performed to refine the parameters. For feature point matching between hybrid camera images, scale invariant feature transform (SIFT) is employed and a method is proposed to improve the SIFT matching output. With the proposed approach, omnidirectional-perspective matching performance significantly increases to enable automatic point matching. In addition, the use of virtual camera plane (VCP) images is evaluated, which are perspective images produced by unwarping the corresponding region in the omnidirectional image. The hybrid epipolar geometry is estimated using random sample consensus (RANSAC) and alternatives of pose estimation methods are evaluated. A weighting strategy for iterative linear triangulation which improves the structure estimation accuracy is proposed. Finally, multi-view structure-from-motion (SfM) is performed by employing the approach of adding views to the structure one by one. To refine the structure estimated with multiple views, sparse bundle adjustment method is employed with a modification to use the sphere camera model. Experiments on simulated and real images for the proposed approaches are conducted. Also, the results of hybrid multi-view SfM with real images are demonstrated, emphasizing the cases where it is advantageous to use omnidirectional cameras with perspective cameras.