论文部分内容阅读
自20世纪60年代以来,模式识别的理论与方法研究及在工程中的实际应用取得了很大的进展。本文先简要回顾模式识别领域的发展历史和主要方法的演变,然后围绕模式分类这个模式识别的核心问题,就概率密度估计、特征选择和变换、分类器设计几个方面介绍近年来理论和方法研究的主要进展,最后简要分析将来的发展趋势。本研究实现了利用模式识别技术在智能移动设备上的树叶自动识别功能。主要应用了图像分割,特征提取和支持向量机技术。最终实现了83%的分类准确率。
Since the 1960s, great progress has been made in the study of theory and method of pattern recognition and its practical application in engineering. This paper briefly reviews the history of development and the evolution of the main methods in the field of pattern recognition, and then focuses on the core issues of pattern recognition, such as pattern classification, probability density estimation, feature selection and transformation, and classifier design. The main progress, the final brief analysis of future trends. This study has realized the automatic recognition of leaves on smart mobile devices using pattern recognition technology. The main application of image segmentation, feature extraction and support vector machine technology. Finally achieved 83% of the classification accuracy.