Object Predictor

~ A step towards Digital Vision !

Inspiration:

-As recent advances in deep learning made tasks such as Image and speech recognition possible also the idea of Scanning and image recognition is very fascinating.
-I wanted to dive deeper into the idea like google lens.

Objective:

• I have built a deep neural network that can recognize images with an high accuracy of while explaining the techniques used throughout the process.
• Deep learning excels in recognizing objects in images as it’s implemented using 3 or more layers of artificial neural networks where each layer is responsible for extracting one or more feature of the image.

What it does:

1. Camera will continuously run in the background.
2. A GUI window is used to capture images of objects to store at a specified location.
3. It will then Create different classes for different objects.
4. It will predict the object shown in the camera.

Challenges I ran into:

Since I'm new to Deep learning and Machine learning technology so initially it was very difficult for me to research and study about these technologies. As I developed this project solo and there were no one to guide or support me wherever I got stuck so developing neural networks project became more challenging to me. But all we need is Determination and Consistency which can make everything possible.

Accomplishments that I am proud of:

• I was able to build an artificial convolutional neural network that can recognize images with a high accuracy
• I did so by pre-processing the images to make the model more generic, split the dataset into a number of batches and finally build and train the model.

What I learned:

• How a Python application uses camera input to train a Support Vector Machine (SVM) to respond to specific actions using neural networks to train a dataset.
• Also we learned an AI project that is a fully-functioning camera classifier by combining principles of computer vision with principles of neural networks.

What's next for Object Predictor:

• I wish to add Text to speech voice assistant for visually Impaired people.
• I would like to add Search results in accordance with the image for better exposure to the product.

Built With

• anaconda
• artificial-intelligence
• deep-learning
• machine-learning
• opencv
• python
• tkinter