Eigenimage2Eigenimage (E2E):一个自我监督的深度学习网络,用于超光谱图像去除
概括
本研究介绍了一种自我监督的深度学习方法,用于超光谱图像 (HSI) 否定,克服了在遥感中缺乏清洁的训练数据的缺陷. Eigenimage2Eigenimage (E2E) 框架有效地消除了HSI中的噪音,而不需要配对数据.
科学领域:
- 遥感 遥感 遥感 遥感
- 图像处理 图像处理
- 深度学习 (Deep Learning) 是一种深度学习.
背景情况:
- 深度学习无效化器需要广泛的配对噪音清洁数据,这对于高光谱图像 (HSI) 很少.
- 现有的方法难以应对高质量信息的高维度和光谱冗余.
研究的目的:
- 开发一种自我监督的学习框架,用于超光谱图像消毒.
- 为了实现有效的HSI无声化,而不需要配对的噪音清洁训练数据.
主要方法:
- 提出了Eigenimage2Eigenimage (E2E) 框架,将HSI排斥转化为自身形象排斥.
- 开发了一种自我监督的学习策略,以从单个噪音高频传输器生成噪音-噪音配对的训练数据.
- 应用了E2E框架,在没有对频谱频段数量的限制的情况下,拒绝HSI.
主要成果:
- 在E2E框架中,只使用噪音数据成功训练了一个denoiser.
- 实验结果表明,与现有的基于深度学习的HSI解密方法相比,其性能优越.
- 该方法有效地在不同频谱频段计数中表示HSI.
结论:
- 自主监督学习提供了一个可行的解决方案,用于在没有清洁数据的情况下拒绝HSI.
- E2E框架提供了一种强大而灵活的方法,用于超光谱图像无色化.
- 拟议的方法推进了远程传感图像分析中的深度学习应用.
相关概念视频
Deconvolution
191
Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
191
Downsampling
188
When considering a sampled sequence with zero values between sampling instants, one can replace it by taking every N-th value of the sequence. At these integer multiples of N, the original and sampled sequences coincide. This process, known as decimation, involves extracting every N-th sample from a sequence, thereby creating a more efficient sequence.
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
188


