中文摘要：

物体识别是家居服务机器人的主要问题之一，考虑到家居非结构化环境下物体识别的复杂性，将不依赖图像分割的局部特征作为关键特征。针对传统SURF算法运算量大的问题，模拟生物视觉功能，提出了一种基于显著性区域指导的局部特征算法。首先采用视觉选择性注意机制提取图像显著区域，然后提取显性物体区域SURF特征，最后完成与目标图像的特征点匹配，实现场景中目标物体的识别。实验证明，和传统SURF算法相比，改进算法速度得到有效提高，同时识别率提高了约10％。

英文摘要：

Object recognition is one of the main problems of the home service robot. In view of the complexity of the object recogni- tion in the indoor non-structural environment, local feature is selected as key features ,which doesn＇t dependent on image segmen- tation. To reduce the computation of the SURF algorithm, a local feature algorithm based on saliency region guidance is proposed. First of all, salient region is extracted from the image based on visual selective attention mechanism. Then, invariant SURF feature is extracted from the salient region. Finally, by matching the feature points, the recognition of target object in the scene is real- ized. Experimental results show that the speed is effectively improved compared with the traditional SURF algorithm, and the rec- ognition rate is improved about 10 percent.