Linear and Non-Linear Models for Remotely-Sensed Hyperspectral Image Visualization

Abstract

The visualization of hyperspectral images still constitutes an open question and may have an important impact on the consequent analysis tasks. The existing techniques fall mainly in the following categories: band selection, PCA-based approaches, linear approaches, approaches based on digital image processing techniques and machine/deep learning methods. In this article, we propose the usage of a linear model for color formation, to emulate the image acquisition process by a digital color camera. We show how the choice of spectral sensitivity curves has an impact on the visualization of hyperspectral images as RGB color images. In addition, we propose a non-linear model based on an artificial neural network. We objectively assess the impact and the intrinsic quality of the hyperspectral image visualization from the point of view of the amount of information and complexity: (i) in order to objectively quantify the amount of information present in the image, we use the color entropy as a metric; (ii) for the evaluation of the complexity of the scene we employ the color fractal dimension, as an indication of detail and texture characteristics of the image. For comparison, we use several state-of-the-art visualization techniques. We present experimental results on visualization using both the linear and non-linear color formation models, in comparison with four other methods and report on the superiority of the proposed non-linear model.

Multisource Remote Sensing Data Visualization Using Machine Learning

“…While the classical method based on band selection uses a very small part of the available information, disregarding the information available in the other unused bands, better results of visualization with AI models have already been reported in recent research papers [23]. Such methods, like the proposed one, allow for a more realistic rendering of colors and, in the same time, producing more appealing images (vivid colors, good contrast, large dynamic range, etc.)…”

“…Starting from the ideas in [23], a fully connected neural network was constructed and trained on a set of MS images, as mentioned in [24], in order to visualize spectral images acquired by different sensors. In the following section, the architecture of the network and the hyper-parameters used for training are described in detail.…”

“…Although research in the area of visualization is still limited, as already described in [24] and [23], a neural network trained on a set of MS images can be used successfully to visualize spectral images acquired by different sensors. Challenges in this field include limited labeled training data and spectral variability between different sensors, as they capture information at different spectral wavelengths.…”

wEscore: quality assessment method of multichannel image visualization with regard to angular resolution

“…The common approach is based on the calculation of various statistics for the resulting images, such as entropy [21 -23], standard deviation, an average of the absolute value of gradient [23], etc. [22]. In works that employ such an approach, it is assumed that the greater the value of the computed statistic, the more information is stored in the resulting image.…”

See as a Bee: UV Sensor for Aerial Strawberry Crop Monitoring

“…Fig.1: Spectral Sensitivity of Western Honey Bee compared to Human vision range. Bee Uv, B, and G are represented by the dotted black, blue and green lines respectively.Human sensitivity for Red, Green, and Blue correspond to the solid line colors[25],[26].…”