Skip to content

A Python implementation of object recognition using a pre-trained convolutional neural network called ResNet50. The goal of the project is to recognize objects in images accurately. To achieve this, the code uses various libraries such as NumPy, Pandas, PIL, Matplotlib, and OpenCV.

Notifications You must be signed in to change notification settings

Nargesmohammadi/Object-Recognition-using-ResNet50

Repository files navigation

Object-Recognition-using-ResNet50

This repository contains code for object recognition using ResNet50, a pre-trained convolutional neural network. The code is written in Python and uses various libraries such as NumPy, Pandas, PIL, Matplotlib, and OpenCV.

Dataset

The dataset used in this project is taken from the CIFAR-10 dataset. It contains images of various objects such as airplanes, automobiles, birds, cats, deer, dogs, frogs, horses, ships, and trucks. The labels for these images are available in the trainLabels.csv file.

Code

The code is available in the object_recognition_using_resnet50.ipynb Jupyter Notebook file. It contains the following sections:

  1. Loading the dataset
  2. Displaying sample images
  3. Converting images to NumPy arrays
  4. Splitting the dataset into training and testing sets
  5. Scaling the data
  6. Building the neural network using ResNet50
  7. Compiling and training the neural network

Requirements

To run this code, you need to have the following libraries installed:

  • NumPy
  • Pandas
  • PIL
  • Matplotlib
  • OpenCV
  • TensorFlow

Credits

The dataset used in this project is taken from the CIFAR-10 dataset.

The ResNet50 model used in this project is available in the Keras Applications module.

About

A Python implementation of object recognition using a pre-trained convolutional neural network called ResNet50. The goal of the project is to recognize objects in images accurately. To achieve this, the code uses various libraries such as NumPy, Pandas, PIL, Matplotlib, and OpenCV.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published