**Author/s: **
Emrullah Acar
,
Mehmet Siraç Özerdem

Knowing the value of soil surface moisture in the agricultural areas are very important in many ways such as minimizing the harmful effects of drought cases, preventing salinity caused by over watering, protecting agricultural lands and using the irrigation system efficiently. The main purpose of this study is that determining a relationship between measurements of local soil moisture and images in agricultural Mardin region and prediction of soil moisture with the determined relationship. The images are derived from TARBIL (http://www.tarbil.org) database. The texture feature vectors are extracted from the images by using Histogram of Oriented Gradients (HOG) algorithm. The obtained feature vectors are then classified into three (much, middle and little) groups by using k-Nearest Neighbor (k-NN) and Multilayer Perceptron (MLP) classifiers.

**Author/s: **
Baris Baykant Alagoz

The depth extraction from visual information is one of underpinning research area of robotics and there is a growing trend in development of autonomous and intelligent systems for real-life applications. These unmanned systems need reliable depth estimations in order to move in three-dimensional space, autonomously. Inspiring from biological vision systems, stereo imaging systems promise a solution for the depth estimation from binocular image pairs provided by stereo cameras. One of the major problems in the depth estimation from the stereo image pair is low depth resolution. This paper discusses the depth resolution problem and presents a depth resolution analysis for stereo imaging systems.

**Author/s: **
Ali Karcı

Hierarchical Fibonacci Cubes HFC(n+2) can be obtained from the Hierarchic Cubic Network HCN(n,n) by removing certain nodes and edges. This problem is very simple when no faulty node exists in an HCN(n,n), however, it becomes very sophisticated if some faulty nodes appear in an HCN(n,n). In this paper, we tried to distinguish HFC(n+2) in faulty HCN(n,n), and it can also be considered as a fault-tolerant embedding in HCN(n,n). Then, we shall show how to dierctly embed a HFC(n+2) into a faulty HCN(n,n) and prove that if no more than two clusters which contain faulty nodes, then HFC(n+2) can be directly embedded. Another case is that if there are more than two clusters which contain faulty nodes, then the labels of faulty nodes must be (Ii,K) and (Ii,L) for each cluster.

**Author/s: **
Serdar Alasu
,
Muhammed Fatih Talu

This paper includes comparision of automatic and interactive segmentation methods. Both methods are used for color images segmentatiton and based Gaussian Mixture Model. In automatic segmentation is segmented image pixels without any prior knowledge provided by the user. Interactive segmentation needs prior knowledge provided by the user and segmentation process are based prior knowledge. Obtained results demonstrate that interactive segmentatiton is faster and more accure than automatic segmentation.

**Author/s: **
Baris Baykant Alagoz
,
M. Emin Tağluk

Understanding real physical meaning of fractional order derivation and integration is important. In this study fractional order derivative of sinusoidal functions and the parameters that affect on the derived function are investigated. First, the fractional order derivation process was demonstrated for single component sinusoidal waves and then for a square wave which is in the format of multi component linear time invariant (LTI) signals. The obtained results showed that the derivative operation affects the amplitude and phase of sinusoidal components in such a way that each of components is shifted in relation to order of operator and the amplitude is affected by frequency, derivative order and phase shift caused by derivation operation. While the integer order derivative shifted the sinusoidal function by 90 degree and generated a signal orthogonal to the original one, the fractional order derivative did not create an orthogonal signal, but created shifted version of the original signal. the study was also confirmed that for an LTI signal and/or system the fractional order derivation can be achieved by expanding the signal in terms of Fourier series, and applying fractional order derivative operation component vise, and then add up all of the derived components.

**Author/s: **
Emrah Dönmez

In this study; we aimed to propose design of a QoS aware resource management infrastructure for a GPGPU supported Grid computing system. This Grid system consists of hybrid (CPU + CPU) and heterogeneous (Nvidia + AMD Radeon) GPGPU computational nodes. It can manage both small scale unit (connections, threads, buffer pools etc.) and large scale unit (whole computing machines). As increasing of the network communication bandwidth and developing powerful computer hardware (CPU, GPU etc.), distributed computing systems acquire more and more attention day by day. Grid computing is as a major player in such kind of distributed system environments like cloud, volunteer, hybrid and etc. Since it supports large scale resource sharing between geographically distributed computer clusters and even single computers. Nowadays, there is another important technology pillar to implement high performance computing rather than CPU, it is known as GPU computing. The GPU systems are ideal especially to data intensive applications; such as image processing, data mining, financial computations etc. Therefore, GPU based grids give an undertaking higher computational performance. GPU processor consists of lots of controllable cores which can be used for high performance demanded applications. Ultimately, the major concerns in grid computing are particularly related to managing QoS requirements, granularity of resources, and heterogeneous resources (both CPU and GPU).