Hybrid CNN-SVM Framework for Automated Cattle Disease Identification using Image-Based Analysis
- Version
- Download 12
- File Size 372.07 KB
- File Count 1
- Create Date 20 March 2026
- Last Updated 20 March 2026
Hybrid CNN-SVM Framework for Automated Cattle Disease Identification using Image-Based Analysis
Gandla Subhash , Shashidhar, Pranay Reddy, Anumalla Sujal
UG Student Dept. of CSE (AI & ML) JNTUH, Karimnagar, India
Subhashgandla12@gmail.com ,shashidhar07@gmail.com
Dr. R. Jegadeesan
Professor & Project Guide ,Dept. of CSE JNTUH, Karimnagar, India UG Student
Dept. of CSE (AI & ML) JNTUH, Karimnagar, India
Pranayreddyy19@gmail.com ,Sujalanumalla24@gmail.com
Abstract—Timely identification of bovine ailments is critical for lowering livestock mortality and mitigating financial setbacks in the dairy and agricultural industries. This paper introduces a hybrid deep learning approach that merges Convolutional Neural Networks (CNN) with Support Vector Machines (SVM) for precise image-based classification of cattle diseases. The CNN module autonomously extracts visual features from bovine images, recognizing complex patterns such as skin lesions, discoloration, and swelling. The resulting deep feature representations are forwarded to an SVM classifier to refine decision boundaries and strengthen generalization. Experimental outcomes confirm that the proposed CNN-SVM hybrid consistently surpasses standalone CNN architectures in both accuracy and F1-scores. The system delivers a scalable, computationally efficient diagnostic tool well-suited for deployment in veterinary facilities and agricultural environments.
Keywords: Veterinary Diagnostics, Deep Learning, CNN, SVM, Cattle Disease Detection, Image Classification
Download