No abstract

No abstract

The foreign substances, arising during the production and shaping of wool and cotton raw materials that are used in textile and cotton gin factories or coming from the outside, decrease considerably the quality of the obtained fabric or yarn. Nowadays, a different methods are used to separate foreign substances in the textile sector, most of these methods are not efficient in terms of speed and quality. Computerized vision systems play a vital role in the field of textiles as in other fields. In this study, Intuitionistic Fuzzy Algorithm is used to define the foreign substances in the images that obtained from a camera. CPU (Central Processing Unit) based applications have speed problems due to the structure of the algorithm. For this reason, GPU (Graphics Processing Unit) technology was used to overcome the speed problem. The otsu algorithm generates a dynamic threshold from the numerical values of the image obtained using the Intuitionistic fuzzy algorithm. By this means, the threshold value of each frame obtained from the camera was calculated on real time and implemented on the image timely. These algorithms were accelerated maximum 262 times using NVIDIA GTX 480 GPU supported display card.

Özetçe-Sezgisel bulanık kenar çıkartım (SBKÇ) algoritması, görüntüleri anlamlandırma veya tanımlama alanında kullanılmaktadır. SBKÇ algoritması, uzman kişilerce tasarlanan ve bu uzman kişilerin hatalarını en aza indirmeyi amaçlayan algoritmadır. SBKÇ algoritmasının paralel olarak uygulanabilir olması, algoritmanın grafik kartlarında gerçekleştirilerek hızlandırılmasının önünü açmaktadır. Bu çalışmada, SBKÇ algoritması farklı boyutlardaki resimleri, NVIDIA tarafından üretilmiş Compute Unified Device Architecture (CUDA) programlama ortamı ile farklı hesaplama kapasitelerine sahip grafik kartlarına aktarılarak test edilmiştir. Algoritmanın CUDA platformuna uyarlanmış paralel modelinin, işlemciler üzerinde çalışan seri uygulamasına kıyasla en az 67, en çok 641 kat çalışma sürelerini kısalttığı görülmüştür. Anahtar Kelimeler -CUDA, GPU, Sezgisel Bulanık Kenar Çıkartım.Abstract-Intuitionistic fuzzy edge detection (IFED) algorithm has been used in the signification or characterization of images. IFED algorithm has been designed by the experts and the algorithm provides to aim to minimize errors of them. To be applicable in parallel of IFED is pave the way for accelerating of algorithm by performing in the graphics card. In this study, IFED algorithm was tested by transferring different size images to graphics cards which has different computing capacity via Compute Unified Device Architecture (CUDA) programming environment which is manufactured by NVIDIA. Parallel model of the algorithm adapted to CUDA platform, compared to serial application running on processor, and has seen that shortened runtime at least 67 times, most 641 times.