Enhancing Dog Breed Identification Accuracy Using Hybrid Inception V3 and Xception Model

Identification of dog breeds is useful in many domains, including competitive dog activities, pet services, and veterinary care. However, because of the wide variety of dog breeds and their outward similarities, it is still a difficult process. The efficiency of Convolutional Neural Networks (CNNs) in image-based applications, such as breed classification, has been well established. In order to improve the accuracy of dog breed identification from photos, this study presents a novel hybrid model that combines the advantages of the Xception and Inception V3 architectures. This method seeks to enhance the model's capacity to differentiate between breeds that seem same by utilizing the potent feature extraction capabilities of both networks. Experimental results demonstrate that the hybrid model outperforms individual models, achieving enhanced accuracy in identifying dog breeds.