摘要:机器视觉是研究如何使计算机对图像数据产生智能化感知的一门科学。物体识别在机器视觉领域属于一项基础研究，对图像理解目标的实现起着至关重要的作用。

本课题主要研究基于机器视觉的物体识别。主要方法是对利用机器视觉技术，在计算机处理能力的强大基础上，对采集到的的图像进行分析处理，包括图象采集然后进行图像灰度化和滤波处理，为了提取到障碍物体的特征在预处理好的图像上进行分割处理和细化处理。这样就可以进行物体识别，构建出环境地图而达到识别障碍物体的效果。在论文中，详细分析了需要识别的物体的图像的处理算法，并且在Windows操作系统平台上，利用VC++6.0实现了对物体进行识别的几个基础模块。一系列实验表明，系统的处理结果都具有一定的代表性，有一定的应用价值。 

　　利用机器视觉，结合计算机的强大处理能力，对障碍物体进行识别，不仅更加智能化而且实用化，为进一步的应用和发展打下结实的基础。

关键词：机器视觉，物体识别，图像处理，图像分割，特征提取

Abstract:Machine vision is studying how to make computer generated image data intelligent perception of a science. Object recognition in machine vision field a basic research, belonging to the realization of the image understanding target plays a vital role.

  This subject is the main research object recognition based on machine vision. Main method is use of machine vision technology, in computer processing ability, based on the strong for the collected image analysis, including image acquisition and processing image gray is changed and filtering processing, in order to extract the features to obstacles in pretreatment good image segmentation treatment and refining processing. So can undertake object recognition, construct environment map and achieve the effect of identifying obstacles. In the paper, a detailed analysis on the need to identify the object of the image processing algorithms, and in Windows operating system platform, realized by vc + + 6.0 on several basic module object recognition. A series of experiments show that the system with the result of the representative, have a certain application value.

  Use in machine vision, combined with the powerful computer processing capability, identification of obstacles, not only more intelligent and practical for further applications, the development and lay a solid foundation.

Key words : Machine vision, Object recognition, Image processing, Image segmentation, Feature extraction