基于Gibbs采样和残差连接的AlexNet蔬菜识别算法  

AlexNet vegetable recognition algorithm based on Gibbs sampling and residual connection

作　　者：刘嫚嫚 代琦[1] Liu Manman;Dai Qi(College of Computer Science and Technology,Zhejiang Sci-Tech University,Hangzhou,Zhejiang 310000,China)

机构地区：[1]浙江理工大学计算机科学与技术学院,浙江杭州310000

出　　处：《计算机时代》2023年第9期43-47,共5页Computer Era

摘　　要：为了进一步提高蔬菜识别的精度,提出了基于Gibbs采样和残差卷积神经网络的蔬菜识别算法,本文将其命名为GiRAlexNet算法。根据马尔科夫随机场与吉布斯随机场的等价性构建图像概率模型,用Gibbs采样获取最优样本点集合,随机取点切割图片。通过GoogleNet、ResNet和AlexNet模型实验显示,分类准确率分别提升了9.22%,3.34%和9.19%。大量实验表明,该GiRAlexNet算法对蔬菜识别的准确率达到98.14%。In order to further improve the accuracy of vegetable recognition,a vegetable recognition algorithm based on Gibbs sampling and residual convolution neural network,named GiRAlexNet algorithm,is proposed.The image probability model is constructed according to the equivalence of Markov random field and Gibbs random field.The Gibbs sampling is used to obtain the optimal sample points set,and the random points are taken to cut the image.The experiments of GoogleNet,ResNet and AlexNet models show that the classification accuracy is improved by 9.22%,3.34%and 9.19%,respectively.Extensive experiments show that this GiRAlexNet algorithm can achieve 98.14%accuracy for vegetable recognition.

关 键 词：蔬菜识别 MRF GIBBS采样 Alexnet 残差结构 切割图像 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]