[Raspberry Pi] IoT Deep Learning Computer Vision Practical Project

Internet of Things and Deep Learning in the field of Computer Vision
Learn how to use it and see examples.

machine learning, deep learning
Now it's real!

Computer Vision, a hot field in artificial intelligence, was born as a practical project with the Internet of Things (IoT) and Raspberry Pi .

Develop your ability to leverage it!

While it's important to build a solid theoretical foundation while learning and teaching artificial intelligence, machine learning, and deep learning, I also believe it's important to develop practical skills.

So, we prepared a project that utilizes Internet of Things (IoT) devices in the field of Computer Vision, a representative field that uses artificial intelligence, machine learning, and deep learning.

The course is structured so that you can learn the theory step by step while completing interesting tasks. Upon completion, you'll be inspired to pursue a variety of computer vision deep learning projects and businesses. It's also been a great help in preparing for my current project.

While creating this course, I developed various contents that were not included in the course, such as 'counting the number of entrants', 'counting the number and speed of vehicles', 'identifying age and gender by looking at faces', and 'receipt and business card recognition' , and dreamed of follow-up courses such as 'Machine Learning, Deep Learning Computer Vision Comprehensive Course', 'Mobile Deep Learning Computer Vision Practical Project', and 'Robot IoT Deep Learning Practical Project' .

Project Introduction

Let's implement a function to recognize directly written numbers using deep learning technology using a Raspberry Pi, a web camera, and OpenCV .

Deep learning can be used to detect various objects (object recognition). YOLO and its friends count the number of parked cars in an image. Count the numbers and save them to a cloud server in real time.

The latest Computer Vision technology recognizes letters and numbers in images and videos (Text Recognition). Let's try a fun project that uses a camera to recognize vehicle license plates.

Now, let's identify faces and eyes (face and eye detection) in images and videos. Then, we'll use deep learning to detect movement. We'll check for drowsiness through real-time video and, if so, wake you up with an alarm.

You can create an access control system that not only recognizes faces but also confirms movements.
When a registered person enters, we'll notify you via the server, Dropbox, or email. If an unregistered person enters, we'll sound an alarm.

Special lecture

1. 'Model accuracy over 99%'  We also offer a special lecture titled 'Raising the bar'. This lecture is titled ' [ Raspberry Pi ] IoT Deep Learning Computer Vision Practical'.  The project began with a question from students in the MNIST handwriting model: "Why can't the MNIST handwriting model say '7' is '7'?" While the model's accuracy is a factor, as are the program's exception handling and the raw MNIST data, the existing Nueral Network model was too simple for training purposes, so I reconfigured it to increase its accuracy to 99.38%.

2. This special lecture will teach you how to create your desired model using TensorFlow and Keras. Using the same images used for YOLO training, you'll train a model with Keras and use it to identify objects. You'll also learn how to train YOLO and Keras, and compare them.

What tools do you use?

What tools will be covered in this course? This course is based on OpenCV, a leading computer vision software library, Python, and TensorFlow.

We'll also use the Raspberry Pi, a leading IoT platform. In addition, we'll install several useful software programs, which we'll explain one by one in this lecture.

Materials needed for the lecture

1. Raspberry Pi board (B+ recommended), PiCamera

Related lectures

[OpenCV]
Python Deep Learning Image Processing Project - Find Son Heung-min!
Create fun and useful projects using the various image and video processing techniques and modules you learned in this course.

bonus!!
Provide additional information on this!

As a bonus, in the lecture 'Python Raspberry Pi IoT Project - Remote Monitoring Car', you will learn about Raspberry Pi and
In the lecture 'Complete Mastery of Angular Firebase - PetStore Shopping Mall Project', lectures on Firebase are provided in a special lecture format .

🙋🏻‍♂️ I'm curious!

Q. What are the features of this course?
A. I thought about how to apply deep learning and machine learning in practice.

This course covers not only the theoretical foundations of Computer Vision, a representative field, but also teaches deep learning through practical projects. Specifically, it uses the Raspberry Pi to help students create practical projects that can be applied in the field, enabling them to leverage their skills in the future.

Q. Can non-majors also take the course?
A. Deep learning and data science aren't necessarily fields exclusive to those with a computer science degree. If you have the passion, you can learn and apply them.