AI-Powered Emotion Recognition in Dog

Year : 2026 | Volume : 04 | Issue : 01 | Page : 20 32
    By

    Abhijeet Singh,

  • Faiz Ali,

  • Harsh Dev,

  1. Student, Department of Computer Science Engineering | Artificial Intelligence, Babu Banarasi Das University, Uttar Pradesh, India
  2. Student, Department of Computer Science Engineering | Artificial Intelligence, Babu Banarasi Das University, Uttar Pradesh, India
  3. Professor, Department of Computer Science Engineering | Artificial Intelligence, Babu Banarasi Das University, Uttar Pradesh, India

Abstract

Understanding animal emotions is important for improving veterinary care, human animal interaction, and overall pet well-being. Inspired by previous research that utilized a modified EfficientNetB5 model for emotion classification in cats and dogs, our study builds upon this foundation with a focus on real-time emotion recognition in dogs. While earlier approaches achieved high accuracy using Dense Residual and Squeeze-and-Excitation blocks, they often lacked real-time applicability and were not optimized for deployment. This paper presents a two-stage, real-time emotion detection pipeline using the YOLO architecture. A custom-trained model classifies the dog’s emotions into five categories using a dataset of 65,000 labeled images. The model applies transfer learning techniques to boost accuracy and generalization across varied breeds and facial expressions. Our approach ensures low-latency processing, making it suitable for smart pet monitoring systems and diagnostic tools in veterinary settings. By integrating detection and classification into a unified system, our work offers a scalable and practical solution for emotion recognition in dogs. This research not only builds upon prior deep learning methods but also advances the field toward real-world applications. In addition, the proposed system emphasizes computational efficiency and robustness, which are essential for real-time environments. The architecture is designed to operate effectively under varying lighting conditions, pose variations, and partial occlusions commonly observed in practical scenarios. Extensive experimentation demonstrates that the model maintains consistent performance while preserving fast inference speed. The framework is adaptable and can be extended to other animal emotion recognition tasks with minimal modification. Furthermore, this work highlights the potential of combining detection and classification within a unified pipeline to achieve reliable and scalable solutions.

Keywords: Yolo, Object Detection, Deep Learning, Dog Emotion Detection, Computer Vision

[This article belongs to International Journal of Machine Systems and Manufacturing Technology ]

How to cite this article:
Abhijeet Singh, Faiz Ali, Harsh Dev. AI-Powered Emotion Recognition in Dog. International Journal of Machine Systems and Manufacturing Technology. 2026; 04(01):20-32.
How to cite this URL:
Abhijeet Singh, Faiz Ali, Harsh Dev. AI-Powered Emotion Recognition in Dog. International Journal of Machine Systems and Manufacturing Technology. 2026; 04(01):20-32. Available from: https://journals.stmjournals.com/ijmsmt/article=2026/view=239510


References

  1. Russell S, Norvig P. Artificial Intelligence: A Modern Approach. 3rd ed. Upper Saddle River: Prentice Hall; 2010.
  2. LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015;521(7553):436–44.
  3. Redmon J, Farhadi A. YOLOv3: An Incremental Improvement. arXiv preprint. 2018; arXiv:1804.02767.
  4. “Gradient-based learning applied to document recognition,” Proc IEEE, 86, no. 11, p. 2278–2324, 1998.
  5. Brownlee J. Recurrent Neural Networks Tutorial. Machine Learning Mastery; 2020.
  6. “A survey on image data augmentation for deep learning,” J Big Data, 6, no. 1, p. 60, 2019.
  7. Hyperparameter Tuning Guide. Ultralytics Documentation; 2024.
  8. “Microsoft COCO: Common Objects in Context,” 2014.
  9. SmartOne AI. Bounding boxes explained: Enhancing object detection. SmartOne AI; 2024.
  10. Confusion matrix glossary. Ultralytics; 2024.
  11. Sharma A. Confusion matrix, accuracy, precision, recall, F1 score. Analytics Vidhya; 2020.
  12. Rosebrock A. Intersection over Union (IoU) for object detection. PyImageSearch; 2016.
  13. Evidently AI. Explaining the ROC curve. Evidently AI; 2024.
  14. K. SY Kim, “A study on dog emotion recognition using deep learning.,” AIP Conference Proceedings, vol. 2790, no. 1, p. 020002, 2023.
Regular Issue Subscription Original Research
Volume 04
Issue 01
Received 10/09/2025
Accepted 09/02/2026
Published 11/03/2026
Publication Time 182 Days
32

Login


My IP

PlumX Metrics