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
130
Static object detection algorithm using Haar-like Features
Considering a native RGB image, Haar-like features are generated features
that are a representation of pixel intensities up to a particular location in an
image, this implies that Haar-like features can be calculated fast and easily
since they are only mathematical sums based on the concept of integral
images [1] .
Dynamic object detection using Convolutional Neural Network
Deep Learning is algorithms can be used in any field where data of the
past events applies to generate predictive models with high levels of
accuracy. For implementation in this design, a deep learning model is trained
to detect hands and fingers in infrared images by feeding a large enough
datasets of hand images.
Results and Discussion
Hands were detected using Haar-Like Features, which was written in
Python programming language. The results include but not limited to the
successful detection of hands in different positions. A sample
of a real-time
image is as shown in Figure 2.
Figure 2:
Hand Detection Using (Haar Cascades) Haar -Like Features
Dynamic object detection using a convolutional neural network (CNN)
The results of utilizing a CNN (Deep Learning Model) were satisfactory
since the model was trained on a broad range of infrared pictures and the
dataset was split into test and train samples with a
total of 20,000 photos.
Dostları ilə paylaş: 
