Search

We have assessed the feasibility of delineating and identifying mineral ores from images of mine excavation faces, taking samples from a gallery mine in Sn ¿ W pegmatitic vein deposits for a case of study. This study consisted on: 1. Analysis of laboratory hyper-spectral images (450 ¿ 1780 nm) of the different minerals of interest from samples collected in the mine to evaluate spectral separability and machine-learning classification under close-to-optimal conditions. 2. Analysis and spectral classification of a hyper-spectral image (950 ¿ 1780 nm) of a panel of samples in the vertical position, acquired under room conditions from a distance of 2.5 m, as a previous test of (i) the deployment of the hyper-spectral equipment in the mine and (ii) the analysis of images acquired in sub-optimal conditions, closer to those expected to be present in gallery mines. 3. Automatic segmentation of conventional digital photos of the mine excavation front to delineate patches of mineral ores and estimate ore grade. The analysis of laboratory hyper-spectral images indicated that all spectral signatures are statistically separable although care has to be taken with quartz forms including numerous impurities. Classification (Linear Discriminant Analysis, Singular Vector Machines and Random Forest) of laboratory spectra had a very high accuracy rate (up to 98%). In the case of the panel, classification accuracy was highest using Random Forests (93.2%). Segmentation of a conventional image of the mine excavation face successfully delimited ore patches and provided area measurements, thus facilitating area-based ore grade estimates. Our results show the interest and feasibility of the combined use of conventional and hyper-spectral imagery to assess and map mineral ores in mine excavation faces objectively.

The paper describes a distributed spectral-screening PCT algorithm for fusing hyper-spectral images in remote sensing applications. The algorithm provides intrusion tolerance from information warfare attacks using the notion of computational resiliency. This concept uses replication to achieve fault tolerance, but goes further to dynamically regenerate replication in response to an attack or failure. The concepts of resiliency are incorporated through library technology that is application independent. This library hides the details of communication protocols required to achieve dynamic replication and reconfiguration in distributed applications. The paper provides a status report on our progress in developing the concept and applying it to image fusion. In particular we examine the performance of the PCT algorithm and compare the results with and without resiliency to assess the associated overheads

The paper describes a distributed spectral-screening PCT algorithm for fusing hyper-spectral images in remote sensing applications. The algorithm provides intrusion tolerance from information warfare attacks using the notion of computational resiliency. This concept uses replication to achieve fault tolerance, but goes further to dynamically regenerate replication in response to an attack or failure. The concepts of resiliency are incorporated through library technology that is application independent. This library hides the details of communication protocols required to achieve dynamic replication and reconfiguration in distributed applications. The paper provides a status report on our progress in developing the concept and applying it to image fusion. In particular we examine the performance of the PCT algorithm and compare the results with and without resiliency to assess the associated overheads

The paper describes a distributed spectral-screening PCT algorithm for fusing hyper-spectral images in remote sensing applications. The algorithm provides intrusion tolerance from information warfare attacks using the notion of computational resiliency. This concept uses replication to achieve fault tolerance, but goes further to dynamically regenerate replication in response to an attack or failure. The concepts of resiliency are incorporated through library technology that is application independent. This library hides the details of communication protocols required to achieve dynamic replication and reconfiguration in distributed applications. The paper provides a status report on our progress in developing the concept and applying it to image fusion. In particular we examine the performance of the PCT algorithm and compare the results with and without resiliency to assess the associated overheads

[EN] Hyperspectral imaging systems are starting to be used as a scientific tool for food quality assessment. A typical hyperspectral image is composed of a set of a relatively wide range of monochromatic images corresponding to continuous wavelengths that normally contain redundant information or may exhibit a high degree of correlation. In addition, computation of the classifiers used to deal with the data obtained from the images can become excessively complex and time-consuming for such high-dimensional datasets, and this makes it difficult to incorporate such systems into an industry that demands standard protocols or high-speed processes. Therefore, recent works have focused on the development of new systems based on this technology that are capable of analysing quality features that cannot be inspected using visible imaging. Many of those studies have also centred on finding new statistical techniques to reduce the hyperspectral images to multispectral ones, which are easier to implement in automatic, non-destructive systems. This article reviews recent works that use hyperspectral imaging for the inspection of fruit and vegetables. It explains the different technologies available to acquire the images and their use for the non-destructive inspection of the internal and external features of these products. Particular attention is paid to the works aimed at reducing the dimensionality of the images, with details of the statistical techniques most commonly used for this task. © 2011 Springer Science+Business Media, LLC.

[EN] Hyperspectral imaging systems are starting to be used as a scientific tool for food quality assessment. A typical hyperspectral image is composed of a set of a relatively wide range of monochromatic images corresponding to continuous wavelengths that normally contain redundant information or may exhibit a high degree of correlation. In addition, computation of the classifiers used to deal with the data obtained from the images can become excessively complex and time-consuming for such high-dimensional datasets, and this makes it difficult to incorporate such systems into an industry that demands standard protocols or high-speed processes. Therefore, recent works have focused on the development of new systems based on this technology that are capable of analysing quality features that cannot be inspected using visible imaging. Many of those studies have also centred on finding new statistical techniques to reduce the hyperspectral images to multispectral ones, which are easier to implement in automatic, non-destructive systems. This article reviews recent works that use hyperspectral imaging for the inspection of fruit and vegetables. It explains the different technologies available to acquire the images and their use for the non-destructive inspection of the internal and external features of these products. Particular attention is paid to the works aimed at reducing the dimensionality of the images, with details of the statistical techniques most commonly used for this task. © 2011 Springer Science+Business Media, LLC.

In the study, multi-spectral data with 30m resolution and hyperspectral images with 100m resolution are used to classify the land cover into six types: water body, public green space, agriculture land, built-up areas and non-use land, clouds, while thermal images with 300m resolution are used to evaluate the urban heat island effect. The field work at Sep 17th, 2009 is used to evaluate the accuracy of classification results. In order to quantify the degree along the rural-urban gradient, Moran's I index and semi-variance are used to assess the spatial autocorrection and describe the scale and pattern of spatial variability. The results show that the land cover map resulted from multi-spectral image has satisfactory accuracy. From the results of Moran' I index and semi-variance, it indicats that spatial pattern of homogeneous patches exist on small scales smaller 36km, meso scales between 36-81km and large scales bigger than 81km. The relationship between land cover types and UHI patterns is also studied., QC 20120217