Build Self Checkout Machine & Virtual Keyboard with OpenCV

Welcome to Building Self Checkout Machine & Virtual Keyboard with OpenCV course. This is a comprehensive project based course where you will learn step by step on how to build a fully automated self checkout system and interactive virtual keyboard using OpenCV, Keras, Convolutional Neural Networks, Media Pipe, and Tkinter. This course is a perfect combination between computer vision and object detection, making it an ideal opportunity to practice your programming skills while improving your technical knowledge in retail automation.

In the introduction session, you will learn the basic fundamentals of the self checkout system, such as getting to know its use cases, technologies that will be used, and some technical challenges. Then, in the next section, you will learn how the self checkout machine works. This section will cover data collection, preprocessing, model training, object detection, matching the product to the dataset, displaying product name and price.

Afterward, we will create training data which will consist of one folder containing product images and an excel file containing product information like product ID, product name, price, and discount. Once, everything is all set, we will start the first project, firstly, we will train the self checkout model using CNN and Keras, after that we will build simple user interface using Tkinter and we will also embed OpenCV webcam to the interface, once the camera has been activated, the user will be able to scan products and the system will automatically calculate the total price. In addition, we will also create a simple payment simulation where users can enter the payment amount and the system will check if the entered payment amount is more than the total price, if yes, then it will display the change but if the entered payment amount is less than total price, the system will ask the user to enter the right amount.

Meanwhile, in the second project section, we will build an interactive virtual keyboard using OpenCV and Media Pipe. This system will be able to recognise hand gestures and provide users with touchless typing experience. After building these two models, we will be conducting testing to make sure these models have been fully functioning and all logics have been implemented correctly.