基于深度学习卷积神经网络的桑果成熟度检测研究

Research on Mulberry Maturity Detection Based on Deep Learning and Convolution Neural Network

桑果成熟度检测目的是根据输入的桑果图像实现对桑果成熟度的自动判定,方便果农对果园作物成熟度进行了解。为此,介绍了基于深度学习卷积神经网络的目标检测算法,并基于Faster R-CNN搭建了桑果检测模型,通过MatLab对模型参数进行训练和优化,实现了对桑果的成熟度检测。实验结果表明:基于图像处理和卷积神经网络的桑果成熟度检测系统对桑果成熟度检测准确率较高,具有一定的实用价值。

The purpose of mulberry maturity detection is to automatically determine the maturity of mulberry fruit according to the input mulberry image,so as to facilitate farmers to understand the maturity of orchard crops.It first introduces the target detection algorithm based on deep learning convolution neural network,and then builds a mulberry detection model based on Faster R-CNN.Through the training and optimization of model parameters in Matlab,it realized the maturity detection of mulberry.The experimental results show it has a high accuracy for the mulberry maturity detection system based on image processing and convolution neural network of mulberry maturity detection,which has certain practical significance.