CAJ | 학술논문

为解决玉米种胚特征提取问题，提出了一种利用 HSI 模型结合 OSTU 方法进行玉米种胚检测的算法。使用研制的图像采集系统获取玉米种子的RGB图像，并利用公式转为HSI 图像进行分析，发现胚部与非胚部的S分量的差异较之 B分量更加明显。同时，采用屏蔽背景像素的 OSTU 方法，对 S 分量图像进行分割，获得玉米种胚区域。试验表明，此算法能检测到的玉米种胚区域与人眼观测到的基本一致。
위해결옥미충배특정제취문제，제출료일충이용 HSI 모형결합 OSTU 방법진행옥미충배검측적산법。사용연제적도상채집계통획취옥미충자적RGB도상，병이용공식전위HSI 도상진행분석，발현배부여비배부적S분량적차이교지 B분량경가명현。동시，채용병폐배경상소적 OSTU 방법，대 S 분량도상진행분할，획득옥미충배구역。시험표명，차산법능검측도적옥미충배구역여인안관측도적기본일치。
In order to solve the problem of feature extraction of maize germ , this paper proposes a method of using the HSI model combined OSTU algorithm to detect corn embryo .Using the research team developed image acquisition system for maize seed RGB images , and using the formula into HSI image , using shielded background pixel OSTU method to segment the S component image to obtain maize embryo area .The results showed that this algorithm is able to detect re-gions of the corn seed embryo with the human eye observed basically the same .