首页分享基于CNN和SVM分类优化的大蒜鳞芽朝向识别研究

基于CNN和SVM分类优化的大蒜鳞芽朝向识别研究

来源：花匠小妙招时间：2025-01-13 20:15

摘要： 针对大蒜鳞芽朝上、直立栽种的特殊种植需求，研究实用性好、准确率高、抗干扰强的鳞芽朝向自动识别算法，具有重要的工程应用意义。提出基于卷积神经网络(CNN)和支持向量机(SVM)分类优化的改进算法(CNN-SVM)，实现大蒜鳞芽朝向的自动识别与修正；提出SVM分类优化方案与随机参数择优、损失函数检测方法，以解决感受野小、分类效果差、过拟合等问题。研究结果表明：CNN-SVM算法的识别准确率为99.8%，单张图片识别时间为0.024 s。与经典CNN、SVM算法相比，本文所提算法对于感受野小、干扰强的识别效果更好；同时具有识别准确率高、计算规模小、对局部特征敏感等优点。不仅为大蒜自动智能播种设备的研发提供算法储备，而且可以推广应用于其他小物体识别。

关键词: 大蒜鳞芽, 朝向识别, 卷积神经网络, 支持向量机, 分类优化, 深度学习

Abstract: Aiming at the special planting requirements of garlic scale buds facing up and upright sowing, an automatic recognition algorithm of scale bud orientation with good practicability, high accuracy, and strong antiinterference was developed. This paper proposed an improved algorithm (CNN-SVM) based on convolutional neural network (CNN) and support vector machine (SVM) classification optimization to realize automatic identification and correction of garlic scale bud orientation. Additionally, SVM classification optimization scheme and random parameter selection was investigated, as well as a loss function detection method to solve the problems of small perception, poor classification effect, and overfitting. The research results showed that the recognition accuracy of the CNN-SVM algorithm was 99.8%, and the recognition time of a single image was 0.024 s. Compared to the classic CNN and SVM algorithms, the proposed algorithm had a better effect on the recognition of small fields and strong interference, while simultaneously having advantages of high recognition accuracy, small calculation scale, and sensitivity to local features. This research not only provides algorithm reserves for the research and development of garlic automatic intelligent seeding equipment, but can also be promoted to other small object recognition.

Key words: garlic scale buds, orientation recognition, convolutional neural network, support vector machine, classification optimization, deep learning

中图分类号:

S223.2