摘要
温室黄瓜的识别是黄瓜采摘机器人研究的重要环节,要解决此问题首先必须对采集的黄瓜图像进行分割。提出了一种结合颜色特征和纹理特征进行分割的方法。根据CIE-XYZ颜色模型及其色度图,在RGB模型中先通过颜色滤去不相关物体,然后利用灰度共生矩阵提取纹理特征,得到可区分果实和背景的两大纹理特征,即熵和能量,有效地解决了果实和背景颜色相似的识别问题。
For the cucumber harvesting robot,the recognition is the important steps.First of all,segmentation of cucumber image is required.A new segmentation method combined with color features and texture features is presented.Based on CIE-XYZ photo of chroma,the irrelevant object is filtered,entropy and energy which can distinguish the fruit and the background are got by using of the gray level co-occurrence matrix.Integrating the color and texture features,the recognition problem with little difference between the...
出处
《光学技术》
CAS
CSCD
北大核心
2009年第4期529-531,共3页
Optical Technique
基金
国家自然科学基金资助项目(30670529)
兰州理工大学电气与控制学科团队基金资助项目
关键词
颜色
纹理
黄瓜
图像分割
color
texture
cucumber
image segmentation