Abstract
We present a dataset of close range hyperspectral images of materials that span the visible and near infrared spectrums: HyTexiLa (Hyperspectral Texture images acquired in Laboratory). The data is intended to provide high spectral and spatial resolution reflectance images of 112 materials to study spatial and spectral textures. In this paper we discuss the calibration of the data and the method for addressing the distortions during image acquisition. We provide a spectral analysis based on non-negative matrix factorization to quantify the spectral complexity of the samples and extend local binary pattern operators to the hyperspectral texture analysis. The results demonstrate that although the spectral complexity of each of the textures is generally low, increasing the number of bands permits better texture classification, with the opponent band local binary pattern feature giving the best performance.
Highlights
When electromagnetic waves are incident on a material’s surface, some part of it is absorbed and some of it is reflected back
We present a dataset of close range hyperspectral images of materials that span the visible and near infrared spectrums: HyTexiLa (Hyperspectral Texture images acquired in Laboratory)
We provide a spectral analysis based on non-negative matrix factorization to quantify the spectral complexity of the samples and extend local binary pattern operators to the hyperspectral texture analysis
Summary
When electromagnetic waves are incident on a material’s surface, some part of it is absorbed and some of it is reflected back. In the push broom imaging technique, line scanning of the scene is performed and the spectrogram of a particular scan is recorded on a charged couple device sensor Another method for the acquisition of hyperspectral images is through the use of a liquid crystal tunable filter (LCTF) [2] and MEMS. To evaluate this compromise and assess its performance, high quality data with high spatial and high spectral resolutions are required Creating such a dataset takes time and effort, but it is necessary to the research community. The idea behind creating this dataset is to provide a platform for the benchmark analysis of various applications and processing The areas where this data could be used are in the fields of image processing, computational imaging and computer vision, such as surface identification, spatio-spectral analysis of textured surfaces, image sensor simulation, color reproduction, image relighting and so on.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
