多特征融合的花卉图像深度学习分类算法

Deep Learning-based Flower Image Classification Algorithm with Multi-features Fusion

许多图像分类问题具有类内相似而类间差异的特点,然而花卉图像的分类往往存在着类间相似和类内差异的现象,因此,基于传统人工设计的图像特征进行花卉图像分类效果一般不够理想.针对这个问题,本文提出融合深度特征和人工特征的花卉图像特征提取方法并在此基础上实现花卉图像的分类.首先构建基于卷积神经网络CNN的特征提取框架,然后利用CNN模型从颜色、亮度多特征角度提取目标对象特征,并利用CNN低层级上的特征图设计了一种基于卷积神经网络的纹理特征,最后将上述多个特征与传统的人工设计图像特征经过融合得到一组花卉图像特征.分类实验结果表明,本文提取的融合特征不仅维度低于传统的人工设计特征,而且具有更好的分类准确性.

While many image classification problems are featured by intra-class similarity and inter-class dissimilarity,the phenomenon of inter-class similarity and intra-class dissimilarity often occurs in flower image classification,thus yielding unsatisfactory flower image classification accuracy if traditional manual features are used. To solve this problem,a method combining deep features and manual feature of flower images is proposed for flower image classification. Firstly,a unified depth feature extraction framework is constructed based on deep convolutional neural network. Then,color and intensity features are obtained via CNN model respectively. Next,we design a texture feature based on CNN through the feature map in the low layers. Finally,after the multi-feature fusion with above features and a manual feature,a more comprehensive description of flower image is acquired. The classification experiments show that the proposed features at lower dimensions than the traditional ones achieve better performance in flower image classification.