Programmable Spectrometry: Per-pixel Material Classification using Learned Spectral Filters
Programmable Spectrometry: Per-pixel Material Classification using Learned Spectral Filters
复制标题
DOI：
10.1109/iccp48838.2020.9105281
复制
发表时间：
2020-04
期刊：
影响因子：
--
通讯作者：
Vishwanath Saragadam;Aswin C. Sankaranarayanan
中科院分区：
文献类型：
--
作者：
Vishwanath Saragadam;Aswin C. Sankaranarayanan
许多材料具有不同的光谱轮廓，这有助于通过处理场景的高光谱图像(HSI)来估计场景的材料组成。然而，这一过程本质上是浪费的，因为获取高维HSI是昂贵的，并且只有HSI的一组线性投影有助于分类任务。本文提出了可编程光谱学的每像素级材料分类的概念，不是感知场景的HSI，然后对其进行处理，而是光学计算光谱滤波后的图像。这是使用具有可编程光谱响应的计算相机来实现的。我们的方法在采集速度(因为只采集相关测量数据)和信噪比(因为我们总是避免使用光效低的窄带滤光器)方面都有优势。在给定充足的训练数据的情况下，我们使用学习技术来识别有助于材料分类的光谱简档库。我们在模拟中验证了该方法，并使用相机的实验室原型验证了我们的发现。
Many materials have distinct spectral profiles, which facilitates estimation of the material composition of a scene by processing its hyperspectral image (HSI). However, this process is inherently wasteful since high-dimensional HSIs are expensive to acquire and only a set of linear projections of the HSI contribute to the classification task. This paper proposes the concept of programmable spectrometry for per-pixel material classification, where instead of sensing the HSI of the scene and then processing it, we optically compute the spectrally-filtered images. This is achieved using a computational camera with a programmable spectral response. Our approach provides gains both in terms of acquisition speed - since only the relevant measurements are acquired - and in signal-to-noise ratio - since we invariably avoid narrowband filters that are light inefficient. Given ample training data, we use learning techniques to identify the bank of spectral profiles that facilitate material classification. We verify the method in simulations, as well as validate our findings using a lab prototype of the camera.