Convolutional network based Classification

-
CNN based weather classifier

CNN(convolutional neural network) are very powerful Computation Weapon made by multiple mathematical processes. CNN can be used by robots to see environment through camera and understand the state of environment. A single layer of CNN works as a filter much like gaussian blur or edge detector filters and when they are used in multiple layers they can understand much  more than just color or shape, they can understand story of the image and today we are going to learn how to train a CNN to predict weather through image.

Weather-CNN


Before we start make sure you have tensorflow,keras,OpenCV, Matplotlib and pandas. we have to download dataset for wether images. Link to image dataset link(https://md-datasets-public-files-prod.s3.eu-west-1.amazonaws.com/64a54851-f95a-439d-aff7-3334a684e532)
dataset consist of 1025 images of weather, it consist of 4 labels 'cloudy','rain','shine','sunrise', image shape varies.




Import dataset, read image labels and save the file as lable.txt this is a part of preprocessing 






Create a new Training.py file and import libraries, all imports are given below.Now read images and labels.txt file and convert labels to hot vectors. Hot vectors works like n-arm bandit problem where we select 1 of all available actions. A hot vector looks like [0 1 0 0] for action  = 2, and are used in classification problems. Reshape images so that all images are of same shape in this case we have reshaped it to (100, 100).




Make sure to fix the random seed to get reproduceable results,make a sequential model with 4 convolutional 2d layers then flatten it and add 2 dense layers and 1 output layer. Full Network summery is given below.

we are using tanh activation function in 2d CNN layers as they produce better results than RELU and Sigmoid activation functions. Accuracy and loss funtion for all three are attached below with epoch and batch size fixed at epoch = 10 and batch size = 50. Plot of model loss and accuracy are given below.
 
Some of the results from prediction are given below, model can not predict right always but gives a fair result.
 
 
      






Comments

Post a Comment

Popular posts from this blog

Opencv python object detection Tutorial (part 1)

-
Image
OpenCV Python  OpenCV Python hand Tracking Series(Part-1) There are many reasons to detect object using camera which allows system to interact with the environment. Detecting obstacles by self driving car(to avoid them of course:) Can be used in gripping object with robotic arm Servaliance Purpose or may be your professor asked you to do it :( we will use OpenCV python to detect human arm with basic image processing techniques. For this method we require OpenCV and numpy library. In this tutorial we will use color as feature and detect objects. For now we will detect a ball but shape does not matter as we are usingcolor as feature and not shape. There are multiple ways to perform the Task such as Color Feature Background Subtraction Shape Features Sift Features SURF Features Deep learning we will cover all the methods but for now we focus on color features. To start we take a good photo of object you want to Track like the image given below Now let's Start Coding, First import nump
Read more

Background Subtraction OpenCv Python

-
Image
Background Subtraction Background Subtraction is a method in image processing in which we subtract the background to detect the object or more specifically changes in the environment. We will create a python code with simple mathematical operations to create background subtraction. The way background subtraction work is we directly subtract background image from input image and then we binarize the image and after some operations like gaussian blur and erosion we get the object. Take 2 images, one background image and one with the object. The idea here is we are working with stable non movable camera. This type of camera are generally used for security.  Read the images in python using cv2.imread command then resize the images to desired size(resizing is always very helpful).  background = cv2.imread( 'background.jpg' ) image = cv2.imread( 'testforground.jpg' ) shape = ( 200 ,  200 ) bg = cv2.resize(background, shape) Convert them to gray scale and then take difference
Read more