Object Detection using TensorFlow

Abstract

Objects in the home that are often used tend to follow specific patterns in terms of time and location. Analyzing these trends can help us keep track of our belongings and increase efficiency by reducing the amount of time wasted forgetting or looking for them. Tensor Flow, a relatively new framework from Google, was utilised to model our neural network in our project. Multiple objects in real-time video streams are detected using the Tensor Flow Object Detection API. The system then detects trends and alerts the user if an abnormality is discovered. Finding REMO—detecting relative mobility patterns in geographic lifelines is a study reported by Laube et al. A neural network model is constructed and trained with the goal of being able to accurately identify digits from handwritten photographs. For this, the Tensor Flow syntax was employed, using Keras as the front end. The trained model can take an image of a handwritten digit as input and predict the digit's class, that is, it can predict the digit or the input picture's class. Machine vision improvements, in combination with a camera and artificial intelligence programming, may now be used by PCs to recognize images.