论文部分内容阅读
应用机器视觉技术对大豆的外观品质进行检测成为近年来的研究热点,大豆的外观特征提取是检测的重要内容之一。为提高大豆样本的识别率,减少噪声对特征提取造成的污染,提出了一种基于小波矩的大豆外观品质特征提取方法。该方法对大豆样本图像进行基于小波变换的不变矩特征提取,有效地解决了由于大豆本身存在的大小不同、移动等造成的特征不明的问题。试验证明:此方法不仅能够精确地描述大豆外观品质特征而且对噪声不敏感,此方法识别精度高,正确识别率达到99%。
The application of machine vision technology to detect the appearance quality of soybean has become a research hotspot in recent years. The appearance feature extraction of soybean is one of the important contents of detection. In order to improve the recognition rate of soybean samples and reduce the pollution caused by noise on feature extraction, a method of extracting appearance quality features of soybean based on wavelet moment is proposed. The method extracts the invariant moments based on the wavelet transform of the soybean sample image, and effectively solves the problem of unknown characteristics caused by different sizes and movements of the soybean itself. The experiment shows that this method can not only accurately describe the appearance quality characteristics of soybean but also is insensitive to noise. This method has high recognition accuracy and a correct recognition rate of 99%.