Тип публикации: доклад, тезисы доклада, статья из сборника материалов конференций
Конференция: Computational Methods in Systems and Software, CoMeSySo 2021
Год издания: 2021
Идентификатор DOI: 10.1007/978-3-030-90321-3_28
Ключевые слова: automated microscopy, computer-aided diagnosis, laplacian, neuro-fuzzy networks
Аннотация: Automated medical imaging is at the heart of modern medical diagnostic systems. The authors propose a model for computer analysis of light microscopy images. This model includes filtering and segmentation modules, a feature extraction module, and a classification module. The formation of the classifying function is based on a neuroПоказать полностью-fuzzy model, and segmentation was carried out using the Laplacian-Gaussian. The authors propose a modified version of the Laplacian of the Gaussian by adding a non-sharp masking procedure. The authors tested the quality of filtering and segmentation using the Normalised Colour Difference (NCD) metric. The feature extraction module is based on contour and color descriptors. The authors use Fourier descriptors as contour descriptors and the color values of the RGB and HSV model parameters of the object and the area around the object as color descriptors. The resulting model is implemented as a prototype of an automated system and tested on a test set of images. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.
Журнал: Lecture Notes in Networks and Systems
Выпуск журнала: Vol. 231 LNNS
Номера страниц: 351-355
ISSN журнала: 23673370
Издатель: Springer Science and Business Media Deutschland GmbH