智能家居系统全景界面可控点自动识别的研究  

Study on automatic recognition of controllable point on panorama interface of smart home system

作　　者：唐菲菲[1] 果连成 张珣[1] 

机构地区：[1]杭州电子科技大学电子信息学院,浙江杭州310018 [2]山海关铁路技师学院,河北秦皇岛066200

出　　处：《现代电子技术》2013年第2期32-35,共4页Modern Electronics Technique

摘　　要：为了满足人们对家居系统客户终端个性化智能化的需求,提出了360°全景漫游的交互界面,及对其中的家居可控点状态的自动识别。全景漫游技术根据surf算法在图像尺度空间提取特征点,通过图像拼接,生成场景在不同视线方向上的透视视图;以窗帘为例,利用灰度共生矩阵提取可控点纹理特征值,根据支持向量机进行分类。实验表明,surf算法具有良好地鲁棒性及高效的配准速度,同时基于纹理特征的模式识别,能较准确地区分图像中的可控点状态。In order to satisfy the requirements of people to intelligentized and personalized home systems,360°panoramic roaming interaction interface and the state automatic identification of controllable point were adopted.The feature points are extracted in the scale space of images by panorama roaming technology according to Surf algorithm.The perspective view of the scene in the different lines of sight directions is generated by image splicing technology.Taking a curtain as an example,the texture feature values of the controllable points are extracted with gray level co-occurrence matrix,and then classified based on SVM.The experimental results show that Surf algorithm has good robustness and high efficient matching capacity,and the pattern recognition based on the texture feature can accurately classify the state of controllable points in an image.

关 键 词：智能家居 全景漫游界面 模式识别 纹理特征 图像拼接 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]