Serkan Dereli

Since only pixel intensities are taken into account in the binarization of gray images during the thresholding stage, it brings with it a significant problem. Because, since the relationship between pixels in the image is neglected, it is seen that noises are sometimes defined as an object, sometimes plays a role in changing the detected object, especially in noisy images where illumination is not uniform. In this study, a locally adaptive thresholding algorithm called Haytham Thresholding is proposed in order to eliminate these limitations of global thresholding algorithms and to eliminate noise caused by lighting during the binarization of the image. Especially in the literature, it is seen that noise is high in methods performed by taking the standard deviation into account when the image has a gradient feature. To prevent this, pixel values were normalized by taking into account the weights of the pixels in the window region instead of their standard deviation. These normalized values were added to the matrix values obtained by the average filter and then subtracted from the original image matrix. In the experiments, the proposed method was compared with Otsu and three different local thresholding algorithms by using four different image types also used in the literature. The comparison of the methods was made both visually and with image quality metrics such as PSNR and SSIM. As a result, it has been observed that the proposed method produces successful results compared to both global thresholding and local thresholding algorithms frequently used in the literature.

Bu çalışmada görüntü işleme alanında nesne tespiti yapan yöntemlerden biri olarak literatürde kabul gören şablon eşlemeyi güçlendirmek amacıyla Komşuluk Havuzu Algoritması önerilmiştir. Bu yeni teknik ile şablon eşleme işleminin gerçekleştirildiği uygulamalarda en büyük eşleşme derecesinin yanı sıra sahte eşleşmelerin kolaylıkla tespit edilmesi sağlanmıştır. Klasik bir görüntü işleme tekniği olan şablon eşleme, küçük boyutlu olan şablon görüntüyü daha büyük bir kaynak görüntüde piksel bazlı arar ve piksellerin benzerlik derecesini matematiksel olarak hesaplar. Bu hesaplama işlemi neticesinde kaynak görüntüde elde edilen eşleşmeler gerek benzeyen/benzemeyen gerekse de zayıf/orta/güçlü benzerlik dereceleriyle ortaya konabilmektedir. Ancak, bu teknik görüntü tarama yöntemini kullandığından hem oldukça yavaş ve hem de sahte eşleşmelere neden olduğundan doğruluğu düşüktür. Önerilen komşuluk havuzu fikri sahte eşleşmelerin önüne geçmekle birlikte en yüksek benzerlik derecesine sahip eşleşmenin daha kolay elde edilmesini de sağladığından şablon eşleme yönteminin doğruluğunu da artırmıştır. Deneyler üç farklı görüntü üzerinde, görüntü boyutu, benzerlik eşik değeri, komşuluk büyüklüğü ve hesaplama süresi olmak üzere dört farklı parametre temelinde gerçekleştirilmiştir. Tüm deneylerde iki farklı benzerlik eşik değeri kullanılarak eşik değerinin doğruluğa ve hesaplama süresine etkilerinin incelenmesinin yanı sıra farklı boyutlarda görüntüler kullanılarak görüntü boyutunun hesaplama süresine olan etkileri de incelenmiştir. Sonuçta önerilen yöntemin sahte eşleşmelerin önüne geçme geçtiği deneylerle kanıtlanmıştır.

This study has been inspired by golf ball movements during the game to improve particle swarm optimization. Because, all movements from the first to the last move of the golf ball are the moves made by the player to win the game. Winning this game is also a result of successful implementation of the desired moves. Therefore, the movements of the golf ball are also an optimization, and this has a meaning in the scientific world. In this sense, the movements of the particles in the PSO algorithm have been associated with the movements of the golf ball in the game. Thus, the velocities of the particles have converted to parabolically descending structure as they approach the target. Based on this feature, this meta-heuristic technique is called RDV (random descending velocity) IW PSO. In this way, the result obtained is improved thousands of times with very small movements. For the application of the proposed new technique, the inverse kinematics calculation of the 7-joint robot arm has been performed and the obtained results have been compared with the traditional PSO, some IW techniques, artificial bee colony, firefly algorithm and quantum PSO.

It is the non-repetitive distribution that makes the random numbers important in artificial intelligence techniques, cryptology, transferring a real environment to the virtual world and many more applications. Since the source of true random numbers consists of data from the physical world, the same number chain is never produced. In this study, images taken from the outside world were used as the source of randomness. The resulting image was first converted into an 8-bit gray image, and then the histogram of this gray image was revealed. As is known, an image histogram shows the color distribution in that image. In this study, the color distribution resulting from the histogram has been converted into a random distribution between 0 and 1. As a result, it was observed that the resulting distribution of numbers overlapped with histogram. Since the distribution of the numbers depends on the ratio of the pixels and the ratio of the pixels on the image obtained, the result is a real random number sequence.

Digital systems consist of thousands of digital circuit blocks operating in the background, working in their simplest form such as addition, subtraction, multiplication, division. In exponential expressions like square roots and cube roots, just like these circuits, it is found in many digital systems and performs tasks. Although these processes seem to be used only in circuits carrying out mathematical operations, they actually take an active role in solving many engineering problems. In this study, a digital circuit design that computes both the integer and a floating point exponent of a 32-bit floatingpoint number has been realized. This digital circuit, which is coded with VHDL language, can be used from beginner to advanced level in FPGA based systems. This digital circuit, which is coded with VHDL language, can be used from beginner to advanced level in FPGA based systems. In addition, three floating IP cores-logarithm, multiplication and exponent-were used in this digital circuit, and results were obtained with a total of five finite state machines in sixty-six clock pulse time.