irml.tar.gz

image-rec

If you want to dive into the technology behind computer vision and self driving cars, this is where to start.

In this course you'll learn how to process data from image files and create convolutional neural networks (CNNs) to classify different types of images. After completing this course, you will have a solid understanding of why CNNs work and which CNN architectures to use for different tasks.

The code for this course is built around the TensorFlow framework, one of the premier frameworks for industry machine learning, and the Python pandas library for data analysis. Knowledge of Python and TensorFlow are prerequisites. 

This course was created by AdaptiLab, a company specializing in evaluating, sourcing, and upskilling enterprise machine learning talent. It is built in collaboration with industry machine learning experts from Google, Microsoft, Amazon, and Apple.

Image Recognition with Machine Learning

<p>In this section you will learn about image processing with <b>TensorFlow</b> and <b>PIL</b>. Before doing any image-related tasks, we need to convert the images into a usable form for deep learning models. TensorFlow and PIL provide us utilities that make image processing very straightforward.</p>



<h4>A. Image recognition</h4>

<div id="section1_l0" class="collapse show">
<p>Computer vision, the field concerning machines being able to understand images and videos, is one of the hottest topics in the tech industry. Robotics, self-driving cars, and facial recognition all rely on computer vision to work. At the core of computer vision is <i>image recognition</i>, the task of recognizing what an image represents.</p>
<p>Before performing any task related to images, it is almost always necessary to first process the images to make them more suitable as input data. This section of the course focuses on image processing, specifically how we can convert images from JPEG or PNG files to usable data for our neural networks.
</div>


<h4>B. Core libraries</h4>

<div id="section2_l0" class="collapse show">
<p>The two main libraries that will be used in this Lab for image processing are TensorFlow and PIL. TensorFlow provides a variety of utility functions to obtain image data from files, resize the images, and even transform a large set of images all at once. The original PIL (Python Imaging Library) is an image manipulation library built for Python. Although its development stopped in 2011, an offshoot called <a href="https://python-pillow.org/" target="_blank">Pillow</a> was created and continues to be developed to provide better image manipulation tools for Python users. We'll use the Pillow version of PIL for more fine-grain image processing.</p>
</div>


# A. Image recognition

<div id="section1_l0" class="collapse show">
<p>Computer vision, the field concerning machines being able to understand images and videos, is one of the hottest topics in the tech industry. Robotics, self-driving cars, and facial recognition all rely on computer vision to work. At the core of computer vision is <i>image recognition</i>, the task of recognizing what an image represents.</p>
<p>Before performing any task related to images, it is almost always necessary to first process the images to make them more suitable as input data. This section of the course focuses on image processing, specifically how we can convert images from JPEG or PNG files to usable data for our neural networks.
</div>


# B. Core libraries

<div id="section2_l0" class="collapse show">
<p>The two main libraries that will be used in this Lab for image processing are TensorFlow and PIL. TensorFlow provides a variety of utility functions to obtain image data from files, resize the images, and even transform a large set of images all at once. The original PIL (Python Imaging Library) is an image manipulation library built for Python. Although its development stopped in 2011, an offshoot called <a href="https://python-pillow.org/" target="_blank">Pillow</a> was created and continues to be developed to provide better image manipulation tools for Python users. We'll use the Pillow version of PIL for more fine-grain image processing.</p>
</div>


# A. Image recognition

<div id="section1_l0">
<p>Computer vision, the field concerning machines being able to understand images and videos, is one of the hottest topics in the tech industry. Robotics, self-driving cars, and facial recognition all rely on computer vision to work. At the core of computer vision is <i>image recognition</i>, the task of recognizing what an image represents.</p>
<p>Before performing any task related to images, it is almost always necessary to first process the images to make them more suitable as input data. This section of the course focuses on image processing, specifically how we can convert images from JPEG or PNG files to usable data for our neural networks.
</div>


# B. Core libraries

<div id="section2_l0">
<p>The two main libraries that will be used in this Lab for image processing are TensorFlow and PIL. TensorFlow provides a variety of utility functions to obtain image data from files, resize the images, and even transform a large set of images all at once. The original PIL (Python Imaging Library) is an image manipulation library built for Python. Although its development stopped in 2011, an offshoot called <a href="https://python-pillow.org/" target="_blank">Pillow</a> was created and continues to be developed to provide better image manipulation tools for Python users. We'll use the Pillow version of PIL for more fine-grain image processing.</p>
</div>


What you'll learn from this course

Image Processing

CNN

SqueezeNet

ResNet

Introduction

A. Image recognition

B. Core libraries