Нечітке моделювання в задачі оцінки якості зображень лазерних плям
Laser beam spot images have been analyzed according to a number of features which were specific to graphical objects of this kind. Statistical and geometrical informa-tional features of laser spot images were extracted; in particular, the Hemming distance between fuzzy hash function values for the a...
Збережено в:
Дата: | 2014 |
---|---|
Автори: | , |
Формат: | Стаття |
Мова: | Ukrainian |
Опубліковано: |
The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute"
2014
|
Онлайн доступ: | http://journal.iasa.kpi.ua/article/view/32104 |
Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
Назва журналу: | System research and information technologies |
Репозитарії
System research and information technologiesРезюме: | Laser beam spot images have been analyzed according to a number of features which were specific to graphical objects of this kind. Statistical and geometrical informa-tional features of laser spot images were extracted; in particular, the Hemming distance between fuzzy hash function values for the analyzed and the template images is proposed as one of the features. Fuzzy hash functions allow identifying an image even in a case of small modifications. A slight change in an argument of such a function does not result in the change of its value. A fuzzy classifier model has been developed, which assigns input images to classes according to the quality of the laser spot image in them. Classification results may be interpreted as a laser beam quality indication; a measure of how well the laser beam can be focused. The system does not only identify the object on an image, a laser spot in the current context, but also provides a qualitative estimation of its correspondence to the template image. |
---|