摘要
新型的气体检测装置是根据传感器敏感单元遇到不同气味所呈现的颜色差异,通过将气味的特征信息转换为图像信息识别气体,具有检测精度高和受环境湿度影响较小的优点。文章分析了传感器的工作原理和响应图像,给出了消除杂散光对图像干扰的处理方法和图像识别算法,介绍了基于ARM/Linux的嵌入式系统作为电子鼻实现平台的思路。利用该系统准确识别了环己胺、二甲基亚砜和乙腈三种有机化合物,验证了所构造的系统的有效性。
Smell-seeing electronic nose utilizes color changes induced in array of sensors when exposed to organic substances to detect odors. It is a new odor detection device with the advantages of high detecting precision and immunity to environment vapor. Principle of electronic nose and response of sensor are analyzed; eliminating technique aimed to heterogeneous light and recognizing image method are designed; an idea that using embedded system based on ARM/Linux as hardware/software platform of electronic nose is provided. The system can detect hexylamine, dimethyl sulphoxide and acetonitrile properly and is verified that it is valid.
出处
《传感技术学报》
EI
CAS
CSCD
北大核心
2006年第3期562-565,568,共5页
Chinese Journal of Sensors and Actuators
基金
西北工业大学研究生种子基金
英才计划联合资助
关键词
可视嗅觉
图像识别
电子鼻
smell-seeing
image recognition
electronic nose
