Scholarlink

This paper proposes a hybrid deep learning model for efficient and accurate crop image classification. First, the original image is processed using a Bidirectional Long Short-Term Memory (BiLSTM) to extract contextual and temporal features. Simultaneously, the image is decomposed into its Red, Green, and Blue (RGB) channels and passed through a 3D Convolutional Neural Network (3D-CNN) to capture volumetric and spatial color texture features. The features obtained from both networks are combined to form a high-dimensional feature vector. Later, Grey Wolf Optimizer (GWO) is utilized to select the most discriminative and relevant features. These selected features given to fully connected dense layers to perform classification. Experimental results demonstrated that the proposed hybrid model improved classification performance and achieved an accuracy of 92.25%.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Hybrid Deep Learning Framework Utilizing BiLSTM and 3D-CNN with Grey Wolf Optimizer for Crop Image Classification

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Dinesh Patel,
  • Kamalesh Vasanthkumari Nagaraja,
  • G. Madhukar

摘要

This paper proposes a hybrid deep learning model for efficient and accurate crop image classification. First, the original image is processed using a Bidirectional Long Short-Term Memory (BiLSTM) to extract contextual and temporal features. Simultaneously, the image is decomposed into its Red, Green, and Blue (RGB) channels and passed through a 3D Convolutional Neural Network (3D-CNN) to capture volumetric and spatial color texture features. The features obtained from both networks are combined to form a high-dimensional feature vector. Later, Grey Wolf Optimizer (GWO) is utilized to select the most discriminative and relevant features. These selected features given to fully connected dense layers to perform classification. Experimental results demonstrated that the proposed hybrid model improved classification performance and achieved an accuracy of 92.25%.