Use less Expensive way to Forecast Weather and at the same time present good output. This is an implementation of Neural Networks and K-NN on Matlab.The model predicts Season in the given Image .
The repository includes:
-
Source code of Feature Extraction using HOG.
-
Source code of implementing Neural Networks (Backpropagation).
-
Source code of implementing K-NN algorithm.
-
Data set of Weather Images.
-
Testing .
-
feature_extraction.m : the directway to start .It show feature extraction from predefined dataset.
-
Knn_classify.m :shows how to train and classify using k-NN algorithm.
-
Neural_net.m: show the implementation of Neural Network Model for classifying the Season.
-
knn_accuracy.m: Code to Show the Accuracy of model using K-NN algorithm.
-
The Model achieved accuracy of 92% using Neural Networks.
-
The model also shows the Accuracy difference between KNN aand Neural Network Backpropagation.