LogoAlan Craig Data Science Portfolio Architecture

Deep Learning: Image Classifier

python logo

In this project, I build an Artificial Neural Network that recognizes objects held in a webcam. I start by installing a virutal Python environment using anaconda. Sometimes installed Python packages interfere, or you need specific versions that mutually exclude each other, or you want to test your program with different Python versions. In all these cases virtual environments come to the rescue. House

I install Keras package which requires the TensorFlow package before installing TensorFlow, and I install the OpenCV package.

I capture images from my webcam by collecting a data-set of images. I took pictures of different household objects. For each category I needed at least > 100 images. Also I included a default category empty that contains images of your room without any objects. I then Google Drive to share the images with each other. The steps in the project are:

What is Deep Learning?

The terms Artificial Neural Network (ANN) and Deep Learning usually refer to the same thing. The first neural networks were experimented with in the 1950s. During their history they have resurfaced several times, only to become uninteresting again. Over the last decade, ANNs have started a triumphant rise to the most powerful type of machine learning we know today

How Artificial Neurons work?

Single Neurons

In a single neuron, inputs and bias are multiplied with weights and then transformed by an activation function:

The maths behind a neuron can be described by a single formula:

y ^ = a c t ( X w + w 0 )

What do you get if you use a linear or sigmoid function for the activation?


Lets get in touch and talk about your next project.