航空学报 > 2025, Vol. 46 Issue (23): 631963-631963   doi: 10.7527/S1000-6893.2025.31963

干扰环境下无人机多源感知专栏

时-谱动态约束的光谱波段自适应成像探测

王宇宁, 梅少辉(), 马明阳, 任永政, 曹孟涛, 冯燕   

  1. 西北工业大学 电子信息学院,西安 710129
  • 收稿日期:2025-03-11 修回日期:2025-03-24 接受日期:2025-04-17 出版日期:2025-05-07 发布日期:2025-05-06
  • 通讯作者: 梅少辉 E-mail:meish@nwpu.edu.cn
  • 基金资助:
    国家自然科学基金(62171381);国家自然科学基金(62201445)

Spectral band adaptive imaging detection with temporal-spectral dynamic constraints

Yuning WANG, Shaohui MEI(), Mingyang MA, Yongzheng REN, Mengtao CAO, Yan FENG   

  1. School of Electronics and Information,Northwestern Polytechnical University,Xi’an 710129,China
  • Received:2025-03-11 Revised:2025-03-24 Accepted:2025-04-17 Online:2025-05-07 Published:2025-05-06
  • Contact: Shaohui MEI E-mail:meish@nwpu.edu.cn
  • Supported by:
    National Natural Science Foundation of China(62171381)

摘要:

光谱探测技术在无人机感知领域具有重要应用价值,其多维信息获取能力显著提升了目标识别精度与环境适应性。然而传统高光谱成像系统采用固定波段采集模式,存在硬件复杂、效率低、数据冗余等问题,难以满足无人机快速低成本目标探测需求。针对上述挑战,提出一种光谱波段自适应成像探测算法,支持动态调整光谱参数,设计结构简化的光学成像系统。首先,建立基于信息量与目标可分离度的光谱成像质量评价体系,量化分析窄带光谱数据的识别贡献;其次,提出基于最小切换时间间隔和最大切换谱段范围约束的光谱自适应模型,实现目标特征可辨识度与数据维度压缩的优化平衡;最后,结合阵列式光学模组给出一体化成像探测系统结构设计。实验验证了所提方法的有效性,为无人机光谱感知提供一条全新的高效解决方案。

关键词: 光谱自适应, 信息量评估, 时-谱约束, 成像探测, 高光谱成像

Abstract:

Spectral detection technology holds significant application value in the field of UAV perception, where its multi-dimensional information acquisition capability substantially enhances target recognition accuracy and environmental adaptability. However, conventional hyperspectral imaging systems, relying on fixed-band acquisition modes, suffer from complex hardware architectures, low efficiency, and high data redundancy, thereby failing to meet the demand for rapid and low-cost target detection in UAV platforms. To address these challenges, this paper proposes an adaptive spectral band imaging detection algorithm, enabling dynamic adjustment of spectral parameters and the design of structurally simplified optical imaging systems. First, a spectral imaging quality evaluation framework based on information entropy and target separability is established to quantitatively assess the recognition contributions of narrow-band spectral data. Second, a spectral adaptive model incorporating constraints of minimum switching time intervals and maximum spectral switching ranges is developed, achieving an optimal balance between target feature discriminability and data dimensionality reduction. Finally, an integrated imaging detection system structure is designed using array-based optical modules. Experimental results validate the effectiveness of the proposed method, offering a novel and efficient solution for UAV-based spectral perception.

Key words: spectral adaptation, information content evaluation, temporal-spectral constraints, imaging detection, hyperspectral imaging

中图分类号: