HOME
Full Articles List
Search by Author
Title: AUTOMATION OF THE TOTAL CLINICAL BLOOD TEST BASED ON THE USE OF A HYBRID ALGORITHM
BibTex 
Open this article in the author's DB |
|
AUTOMATION OF THE TOTAL CLINICAL BLOOD TEST BASED ON THE USE OF A HYBRID ALGORITHM
|
|
|
|
|
|
|
|
||
|
|
|
|
|
|
1314-2704
|
|
|
|
||
|
|
English
|
|
|
|
18
|
|
|
|
2.1
|
|
|
|
|
|
|
|
||
|
|
Segmentation of badly structured images is a complex task for different models of artificial intelligence systems. The given work presents algorithmic solutions for intercellular relations measurement in peripheral blood. There has been proposed hybrid algorithm, allowing to distinguish three classes of formed elements: ?erythrocyte?, ?leukocyte? and ?not erythrocyte and not leukocyte?. The algorithm provides for two-level processing of the image. Analysis of geometric attributes of image segments and their colour characteristics is the basis of the algorithm. Different images, being got by use of morphologic operators ?erosion? and ?dilatation? to the initial image, are analysis in this article. The process of segmentation is carried out on black-and-white image by gradient method. At each stage of the algorithm decision is made on the basis of aggregation of two values obtained as a result of analysis of segment colour and geometric attributes. Final decision is made on the basis of decision comparing, accepted at the first and second stages of processing. The algorithm may be used in automated systems of hematologic analysis.
|
|
|
|
conference
|
|
|
|
||
|
|
||
|
|
18th International Multidisciplinary Scientific GeoConference SGEM 2018
|
|
|
|
18th International Multidisciplinary Scientific GeoConference SGEM 2018, 02-08 July, 2018
|
|
|
|
Proceedings Paper
|
|
|
|
STEF92 Technology
|
|
|
|
International Multidisciplinary Scientific GeoConference-SGEM
|
|
|
|
Bulgarian Acad Sci; Acad Sci Czech Republ; Latvian Acad Sci; Polish Acad Sci; Russian Acad Sci; Serbian Acad Sci & Arts; Slovak Acad Sci; Natl Acad Sci Ukraine; Natl Acad Sci Armenia; Sci Council Japan; World Acad Sci; European Acad Sci, Arts & Letters; Ac
|
|
|
|
189-194
|
|
|
|
02-08 July, 2018
|
|
|
|
website
|
|
|
|
cdrom
|
|
|
|
518
|
|
|
|
Segmentation of images; Neural networks; RGB-codes of pixels; Morphologic operators Erosion; Dilatation
|
|
Download PDF