GitHub - Joelvino/Human-Detection: In this work, we will use radar pulse waves emitted by a radar in a specific area to detect and count the number of people present in an area at a particular moment .Thus, we will study the reflected signals coming from this area. To satisfy our goal, we are confronted with many problems, namely the difficulty of separating the human signal from the noises and the cluters, the under-detection problem where certain instance of human being are not detected. In order to solve these problems and to reach our goal, we proceed to the analysis of the reflected signal by applying the wavelet transformation in order to be able to decrease the size of the signal and to eliminate the noises,then we proceed to the creation of the features based on the statistical and entropic properties of the signal those which will allow us to apply on our created features four algorithm of classification has to say ANN, RandomForest, Logistic Regression and Adaboost in order to predict the number of person present in this specific area. The result is that the RandomForest classifier with the daubechies 4 (db4) wavelet gives us a better accuracy equal to 97%,compared to the coiflet 4 (coif4), biorthogonal 3.5 (bior3.5) and symlet 4 (sym4) wavelets. The different tools used in this work are the pywavelets , sklearn and TensorFlow libraries of python.

Joelvino / Human-Detection Public

Notifications You must be signed in to change notification settings
Fork 1
Star 3

In this work, we will use radar pulse waves emitted by a radar in a specific area to detect and count the number of people present in an area at a particular moment .Thus, we will study the reflected signals coming from this area. To satisfy our goal, we are confronted with many problems, namely the difficulty of separating the human signal from…

3 stars 1 fork Branches Tags Activity

Star

Notifications

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 1 Commit
Code.ipynb		Code.ipynb
Essays.ipynb		Essays.ipynb
Logistic_Regression.ipynb		Logistic_Regression.ipynb
Random_forest.ipynb		Random_forest.ipynb
coif4-1.csv		coif4-1.csv
db4-1.csv		db4-1.csv
db4-2.csv		db4-2.csv