Dog Breed Classification using Deep Learning
The current research involves determining the breed of a dog in a given image, which is a fine-grained image recognition challenge including multi-class categorization. The proposed system makes use of cutting-edge deep learning techniques. This proposed approach is for the identification of dog breeds. Under a deep Convolutional Neural Network, classifying dogs breed is a difficult task. A collection of photographs of a dog’s breed is used to classify and learn the breed’s characteristics. This paper is about a study that uses Xception architecture to classify different dog breeds. Xception architecture proves to be an effective solution for image classification. For improved classification accuracy, the Xception architecture requires a large number of photos as training data and a significant amount of time to train the data. Given the large number of breeds in this fine-grained categorization challenge, we prove our results to be a success. Given the considerable diversity both between and within the 133 different breeds in the dataset, we are able to effectively predict the proper breed over time in one guess, a result that few humans could equal. The dataset was obtained through Kaggle. Our system achieved 91.34 percent training accuracy and 89.45 percent validation accuracy.
PROJECT OUTPUT VIDEO:
ALGORITHM / MODEL USED:
- System : Pentium i3 Processor.
- Hard Disk : 500 GB.
- Monitor : 15’’ LED
- Input Devices : Keyboard, Mouse
- Ram : 4 GB
- Operating system : Windows 10.
- Coding Language : Python 3.8
- Web Framework : Flask