欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
面对图像迅速增长的局面和用户检索图像的要求,依靠先进的技术提取图像蕴含的情感语义并对其分类正是当前各行业急需解决的问题.为此,以自然风景图像为例,提出了一种基于Adaboost-BP神经网络的图像情感语义分类方法,通过OCC情感模型描述图像的情感,使用Adaboost算法组合15个BP神经网络弱分类器的输出,构建强分类器,旨在提高图像情感语义分类的效率.使用百度图片频道上下载的600张自然风景图像进行训练和测试,实验通过与BP神经网络方法测试结果相比较,取得了良好的分类效果,可为更多类型的图像情感自动分类打好基础,具有一定的实用价值.
英文摘要：
The rapid growth of images and the request of user’s retrieval have resulted in the rapid growth of digital images,and it has become an urgent problem to rely on advanced technology to extract sentiment of images and classify them automatically.A sentiment classification method of image based on Adaboost-BP neural network is proposed using natural scenery images as an example.The method describes image emotional level by OCC sentiment model,construct a stronger classifier by integrating 15 outputs of BP neural network using Adaboost algorithm.It aims at improving the efficiency of image emotional classification Using 600 natural scenery images downloaded by Baidu photo channel to train and test,experiments achieved good effect compared with BP neural network method.The method can lay a good foundation for more types of image sentiment automatic classification and has some practice value.
同期刊论文项目
同项目期刊论文
Analysis method based on the fMRI data of thinkingresearch on the modeling method of feature binding
Machine learning classifier using abnormal brain network topological metrics in major depressive dis
A cloud scheduling strategy based on the combination of Fuzzy Quotient Space Cluster and an Improved
Lung CT image segmentation based on combined multi-scales watershed method and region growing method
期刊信息
位置：