肉品气味成分复杂、含量低、热稳定性差，如何实现气味的绿色安全、快速、精确检测是一个值得深入研究的科学问题。本项目提出基于天然色素传感器的可视化检测方法，从提高天然色素气敏传感器的检测特异性、稳定性及信号提取准确性三个方面，突破传统色素传感器理论和技术的瓶颈。从研究天然色素与肴肉特征气味成分的反应机理入手，通过构建色素与肴肉特征气味成分反应的分子动力学模型，筛选出选择性好、灵敏度高的天然色素，提高传感器的检测特异性。在传感器开发方面，保留天然色素分子结构中能与气体分子结合的反应基团，将不稳定基团与稳定结构复合，制备天然色素衍生物以提高天然色素的稳定性。信号提取方面，通过高光谱图像技术捕捉反应前后天然色素分子结构变化引起的可见-近红外光谱差异，将信号捕捉的精度和准确性提升到分子层面。项目的完成将得到一种肴肉新鲜度快速、非接触式检测的新方法。

The smell of meat and meat products has complex components, low content and poorer thermal stability. So it is deserves research deeply to achieve green, safe, fast, non-contact detection of gas. In the project a colorimetric gas sensor method based on natural pigment was proposed. This research will help to break through the theoretical and technical bottleneck of traditional pigment sensor by improving the selectivity and stability of the natural pigment sensor, also the extraction accuracy of the pigment signal. Starting with the reaction mechanism between natural pigments and the characteristic odors of decaying Yao-meat, the molecular dynamics models of the reaction would be constructed to screen the highly selective and sensitive natural pigments. For the development of the sensor, derivatives of natural pigments would be synthesized by transform the unstable group into a stable one so that better stability could be obtained, while retaining the sensitive part. For the signal extraction respect, hyperspectral imaging technique would be employed to obtain the change of the visible-near infrared information caused by the structure change of the pigments, which elevates the accuracy and precision for signal capturing to a molecule level. This research would acquire a novel fast, non-contact detection method for Yao-meat freshness, thus, provide a new tool for on-line monitoring of food safety.