Wheat Seeds Classification using Multi-Layer Perceptron Artificial Neural Network
Wheat Seeds Classification using Multi-Layer Perceptron Artificial Neural Network
复制标题
使用多层感知器人工神经网络进行小麦种子分类
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
Dr. Anas M. Quteishat
中科院分区：
文献类型：
--
作者：
N. A. Abdullah;Dr. Anas M. Quteishat
--小麦种子分级是一项重要的农业工序。提出了一种基于人工神经网络的小麦分类系统。该系统旨在将三种不同的小麦种子归入相应的类别。该系统由两个阶段组成。在第一阶段，对得到的图像进行图像处理，提取重要的几何特征。在第二阶段，提取的特征被反馈给使用反向传播学习算法训练的多层感知器(MLP)神经网络。进行了三个实验：第一个实验使用所有数据，第二个实验使用噪声数据，最后一个实验使用部分训练数据。实验结果表明，该分类系统能够对小麦种子进行分类，分类正确率在95%左右。
– Wheat seeds classification is an important agriculture process. In this paper a wheat classification system based on Artificial Neural Network (ANN) is presented. The proposed system aims to classify three different wheat seeds into their corresponding classes. The system consisted of two stages. In the first stage image processing is applied on the obtain images and important geometrical features are extracted. In the second stage the extracted features are fed in to a Multi-layer Perceptron (MLP) neural network trained using back propagation learning algorithm. Three experiments were conducted; the first experiment using all the data, the second experiment using noisy data, and the final experiment using part of the training data. The empirical results show that the proposed classification system was able to classify the wheat seeds with a testing accuracy of around 95%.