id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-024346-shauvo3j	Kruglov, Vasiliy N.	Using Open Source Libraries in the Development of Control Systems Based on Machine Vision	2020-05-05		.txt	text/plain	1767	116	58	The possibility of the boundaries detection in the images of crushed ore particles using a convolutional neural network is analyzed. To build a neural network and apply machine learning methods, a sample of images of crushed ore stones in gray scale was formed. It is this type of neural network that will be used in constructing a model for recognizing boundary points of fragments of stone images. These modifications of the base convolutional neural network did not lead to an improvement in its performance -all models had the worst quality on the test sample (in the region of 88-90% accuracy). In this work, a convolutional neural network was developed and tested to recognize boundaries on images of crushed ore stones. Based on the drawn borders on the test images, it can be concluded that the convolutional neural network is able to correctly identify the boundary points with a high probability.	./cache/cord-024346-shauvo3j.txt	./txt/cord-024346-shauvo3j.txt