基于深度层次模型的图像分类算法

Image Classification Algorithm Based on Depth Layer Model

图像分类是大规模图像检索的基础,为了提高图像分类的准确率,提出了基于深度层次模型的图像分类算法。首先提取图像的颜色、纹理和边缘等特征,并进行归一化处理,然后采用深度层次模型对图像分类的训练样本集进行学习,建立图像分类器,最后在Matlab2014平台上采用图像数据集对算法的性能进行了测试。实验结果表明,算法能够获得理想的图像分类结果,图像分类正确率要远远高于对比图像分类算法,具有更高的实际应用价值。

Image classification is the basis of large-scale image retrieval, in order to improve the accuracy or- image classification, a novel image classification algorithm based on the depth hierarchy model is proposed. Firstly, the features such as color, texture and edge of the image are extracted and normalized, and then, a depth level model is used to learn training set of image classification, and the image classifier is built, finally, the performance is tested on the Matlab 2014 platform by using some image datasets. Experimental results show that the proposed algorithm can obtain ideal image classification results, and the image classification accuracy rate is much higher than the contrast image classification algorithms, so it has higher practical value.