Innovations in Science and Technologies Volume 2 Issue 1 (2025) · pp. 96-104
EFFICIENCY METHODS OF THE CNN DNN MODEL IN ENHANCING THE QUALITY OF MICROSCOPIC IMAGES AND THEIR VIZUALIZATION
Ismailov O.M, Ismailov O.M, Eshmurodov D.E, Eshmurodov D.E, Tulaganova F.K, Tulaganova F.K
Abstract
The application of deep training algorithms is of great importance when it comes to detecting micrascopic images and isolating them from objects. In particular, convolution neural networks (CNN) show high efficiency in object recognition and image analysis in medicine. This article is devoted to the application of deep training methods for segmentation and object recognition of microscopic objects and the comparative analysis of these methods with traditional and other algorithms. With CNN, the images are cut into small pieces and the main features are identified in each piece. In the process, the color values of the image (RGB) are transmitted to the network through the input layer and then analyzed using filters in the convolution layer. Cotton fibers and bacteria are separated by means of Sobel and Canny filters. This article explores the effectiveness of segmentation and classification algorithms.
Deep training (Deep Learning), convolutional neural network (CNN), micrascopic image segmentation, image analysis, object separation, classification algorithms.
Metadata source: the journal's OAI-PMH archive · Sindex does not store the full text; it links to the source.