Abstract

The problem of detection of objects on their spectral patterns by hyperspectral mapping is considered. The main issues hindering an effective and fast solution of the problem are described. The method of on-line object–background discrimination is validated, which utilizes multi-band tunable spectral filters. The principles of operation and technical implementation of these devices are formulated. The assumptions forming the basis of this method and its features are discussed. An original method based on a modulated acousto-optic filter is proposed. We have made conclusions on a conceptual possibility of the considered methods based on a novel technique of multi-band acousto-optic filtering, on its relevance, a short-time response, operational efficiency, and adjustability.

中文翻译：

---

使用多频带声光滤波器的光谱模式识别

摘要

考虑了通过高光谱映射在物体的光谱图案上检测物体的问题。描述了阻碍有效，快速解决问题的主要问题。通过使用多波段可调谐光谱滤波器，对在线物体-背景识别方法进行了验证。阐述了这些设备的操作原理和技术实现。讨论了构成该方法基础及其功能的假设。提出了一种基于调制声光滤波器的原始方法。我们已经基于一种基于多频带声光滤波的新技术，在考虑的方法的概念可能性，相关性，短时响应，操作效率和可调整性方面得出了结论。