Vision-Based Pest Detection Using SVM Classification with Attention-Based Multi-Scale Feature Fusion

Abstract

Vision-Based Pest Detection Using SVM Classification presents a fundamental challenge in Computer Vision, SVM, Agriculture. Existing approaches, including SVM-RBF, SVM-Poly, and KNN, process input data at a single resolution and fail to capture patterns spanning multiple scales, resulting in a mean accuracy ceiling on benchmark datasets. We address this limitation by introducing PestSVM, a hybrid deep learning framework that integrates three parallel convolutional streams (kernel sizes 3, 7, and 13) with bidirectional LSTM encoding and a gated attention fusion module. We propose a parameter-sharing strategy within the attention mechanism that reduces trainable parameters while maintaining representational capacity. We train and evaluate our framework on IP102, D0 Pests using stratified 10-fold cross-validation. Our method achieves a mean precision (%) of 91.8% on the primary benchmark, surpassing the nearest baseline by 7.6 percentage points (p < 0.001, Cohen's d = 1.42). We further demonstrate a 15.3% reduction in computational cost relative to comparable hybrid architectures and convergence within 82 epochs on all benchmark datasets.