图像颜色特征提取及其分类研究  

Study on Image Color Feature Extraction and Classification

作　　者：张锡英[1] 车鑫[1] 

机构地区：[1]东北林业大学信息与计算机工程学院,黑龙江哈尔滨150040

出　　处：《安徽农业科学》2014年第7期2196-2198,共3页Journal of Anhui Agricultural Sciences

摘　　要：在图像分类和检索中,如何对图像进行特征提取及提取图像的哪些特征信息,影响着图像分类的准确性。提出了一种基于颜色的特征提取算法,同时利用最新的多层的深度学习算法对图像进行分类。结果表明,该研究提出的图像分类方法比传统图像分类算法准确度有较大的提高。In image classification and retrieval, how to extract the feature of image and what kind of image feature information to extract influences the accuracy of image classification. An algorithm to extract color feature was proposed, the newest deep learning algorithm was adopted to classify the images. The results showed that the accuracy of the proposed image classification method is improved significantly compared with the traditional algorithm.

关 键 词：图像颜色特征 深度学习 图像分类 

分 类 号：S126[农业科学—农业基础科学]