Visualization and determination of the geometrical parameters of frieze yarns

Carpet yarns have been commonly used in the carpet industry. Producers have been required to generate a new type of yarn by the competitive conditions in the carpet industry. To achieve this, carpet yarns have been passed by a particular process. With this process, a new type of yarn, called frieze/crimped carpet yarn, has been obtained. This crimp/frieze structure have different geometric shape and affect the opalescent effect on the carpet. The crimp count and geometric shape become important as it affects the carpet's appearance. In the carpet industry, crimp structure has been evaluated by human eye. This subjective evaluation causes confusion when determining the crimp count. These results cause the uncertainty about how much crimp count is available on the yarn. This main aim of this thesis is to develop an image processing algorithm to evaluate the frieze carpet yarn's crimp count objectively. As a part of this thesis, the algorithms have been developed to evaluate the crimp count and similarity to each other. With the developed algorithm, a basic vision system that consists of a digital camera, a computer and a platform with white ground have been used. With the aid of the developed algorithm and with the image processing algorithm, crimp count and difference of crimp structure the yarns samples were easily determined and the results were presented. Key Words: Frieze carpet yarn, Crimp count, Crimp geometry, Image processing, The developed algorithm