The problem IOT spot || DL simplified SSOC contributions solves

DL Simplified

Machine learning is an application of artificial intelligence, which allows the computer to operate in a self-learning mode, without being explicitly programmed. It is a very interesting and complex topic, which could drive the future of technology.

There are copious applications of Machine learning, out of which Image Classification is one. To classify images, here we are using SVM. Scikit-learn is a free software machine learning library for the Python programming language and Support vector machine(SVM) is subsumed under Scikit-learn.

Shoe Classification

• A deep learning model that was trained to identify if a shoe was from adidas or nike
• SVM accuracy - 72%
• CNN Accuracy - 82%

Indian Food Classification

• A deep learning model that was trained to identify Indian foods out of a given 88 different types of foods.
• ANN - 1.00%
• CNN - 22%
• CNN with max pooling - 22%

Portuguess Food Classification

• A deep learning model that was trained to identify Portuguese foods out of a given 24 different types of foods.
• ANN - 8%
• CNN - 42%
• CNN with max pooling - 45%

IOT Spot

I mainly focused on ESP8266 and ESP32-related projects. Here are the projects I have contributed to

• Basic LED Blink using ESP8266
• HTTP Authentication using ESP8266
• Basic LED Blink using ESP32
• HTTP Authentication using ESP32
• Hall sensor on ESP32
• LED control on a webserver using ESP32
• Relay control using Web server on ESP32
• Servo_control_using_ESP32
• Temperature Humidity measure using ESP32
• Touch sensor on ESP32

Challenges I ran into

I had to refer through various documentation and many youtube videos to understand the various modules being implemented. I had to read about various pre-trained models to learn which model would be suitable in which situation. All the data then extracted could be implemented in various ML models for the prediction and classification of images.