THE 3 rd INTERNATIONAL SCIENTIFIC CONFERENCES OF STUDENTS AND YOUNG RESEARCHERS dedicated to the 99
th
anniversary of the National Leader of Azerbaijan Heydar Aliyev
177
is detected and the status changes to Tracking (2
nd
picture). Except this,
another valuable information as the number of people entered/exited and
how many there are inside and total number passed is counted and
displayed.
The previously mentioned issues can be avoided and fixed in further
versions of the program via including other related tools and more complex.
Needless to mention, that the code can be updated to detect other objects,
such as other means of transport, animals, household items and others
included in the imported library.
References [1] D. Cochard, "MobilenetSSD : A Machine Learning Model for Fast Object Detection," 2021.
[2] U. Michelucci, "Advanced Applied Deep Learning: Convolutional Neural Networks and
Object Detection," 2019.
[3] J. Howse, "OpenCV Computer Vision with Python: Learn to capture videos, manipulate
images, and track objects with Python using the OpenCV Library," 2013.
[4] "SciPy documentation".
RETINAL FUNDUS IMAGE BLOOD VESSEL SEGMENTATION FOR MEDICAL DIAGNOSIS Samra Huseynova, Fidan Mahmudova Baku Higher Oil School Baku, Azerbaijan samra.huseynova.std@bhos.edu.az; fidan.mahmudova.std@bhos.edu.az Supervisor: PhD, Associate Professor Leyla Muradkhanli Keywords: contrast limited histogram equalization, principal component analysis, DRIVE
The morphology of blood vessels in retinal fundus images can be an
important indicator of diseases which have direct impact on eyes, like
malaria, glaucoma, hypertension and diabetic retinopathy. In this case
quality of the segmentation plays an essential role to obtain accurate results.
Since different fields can benefit from analysis of retinal blood vessels,we
believe that it is worthful to develop the best method for obtaining precise
map of tiny capillaries with the help of some high level image processing
techniques and mathematical tools. So, our main objective is to fulfill the task
of retinal fundus image blood vessel segmentation through several
procedures to be further used for medical diagnosis.