#Picture text recognition software install#
#Picture text recognition software download#

You can see here first we have samples from 0 to 9, then we have another different sample of 0 to 9 (of different handwriting). If we read more than 10 images for instance 15, the result will be: Plt.imshow(image, cmap=plt.cm.gray_r, interpolation='nearest') So for loop iterates through the handwritten images and through the target labels as well: > for i, (image, label) in enumerate(images): As we have stored our images and target data into a list named images, we will use the enumerate method so that the handwritten images go into the image variable in for loop and the target labels go into the label variable in for loop. Use the for loop to iterate through the first 10 images and plot them. To plot the images, define the size of the plot screen: > plt.figure(figsize=(4,4)) The list() method creates a list of the concatenated images and labels. The zip() function joins together the handwritten images and the target labels. Fetch the target labels and the handwritten images and store them as below: > images = list(zip(digits_data.images, digits_data.target)) The handwritten images are stored in the image attribute of the dataset and the target labels or the original numbers are stored in the target attribute of the dataset. In the next step, we will implement the machine learning algorithm on first 10 images of the dataset.Ĭonsider the following steps: Visualize the images with matplotlib: The pixel values range from 0 to 255 where 0 stands for black and 255 represents a white pixel as shown below: > digits_data.images Each element of the array represents a pixel of the array. To convert this image into gray image use: > plt.gray() > plt.matshow(digits_data.images) > plt.show()įor machine learning, all the images will be grayscale images represented as an array. Now display this matrix using show() method of matplotlib: > plt.show() So here we have selected the 1 st image from our dataset whose index is 0. Use the matshow() method which will display the image array as a matrix. Import the matplotlib to plot the images as: > import matplotlib.pyplot as plt Let’s check if we actually got them as image or not: From this we can derive that all 1797 values are the different forms of range from 0 to 9 and we just have different samples of numbers from 0 to 9. 1797 samples corresponds to 1797 target values but the target values have 10 unique values from 0 to 9. This means that we have 1797 image samples with 64 different features. If you want to check what is inside digits_data, type the following command: > digits_ The load_digits() method will read the digits into the digits_data variable. Now read the dataset and store it in a variable: > digits_data = load_digits() Here we say, load the digits from the datasets provided by sklearn module! First import the module: > from sklearn.datasets import load_digits The first step that is required to do is to load the dataset. Okay, so we have everything to get started.

#Picture text recognition software install#

To install scikit-learn, run the following pip command: pip install sklearn The scikit-learn or sklearn library comes with standard datasets for example digits that we will be using. Python provides us an efficient library for machine learning named as scikit-learn. From this tutorial, we will start from recognizing the handwriting.

The matplotlib is used to plot the array of numbers (images). The image is actually a matrix which will be converted into array of numbers. We are using numpy to convert our images in to arrays as our machines understand arrays and numbers or to flatten our images using reshape. The numpy module is used for arrays, numbers, mathematics etc. These are the very basic modules that we need for images. Similarly, import the other two modules: import matplotlib If the module is not installed, you will get an error. To check, if the required modules are installed, import the modules in python shell using the import keyword as follows: import numpy This is called the python shell where the python commands can be executed.

Hit the enter key and you will have the following window opened: Open python shell from start menu and search python IDLE. Let’s check if the modules that we installed are actually there or not.

Now to install Matplotlib, you will write: pip install matplotlibĪs I have already installed the module so it says requirement is satisfied. Run the following pip command in command prompt to check if we have pip installed or not: pip -version The modules Matplotlib, numpy, and sklearn can be easily installed using the Python package Manager.

#Picture text recognition software download#

When python is installed, pip is also installed and you can download any modules/ libraries using pip. Remember to add Python to environment variable. You can install Python from Download Python. The tools that we are going to use in this tutorial are: