scholarly journals High-Resolution Aerial Detection of Marine Plastic Litter by Hyperspectral Sensing

2021 ◽  
Vol 13 (8) ◽  
pp. 1557
Author(s):  
Marco Balsi ◽  
Monica Moroni ◽  
Valter Chiarabini ◽  
Giovanni Tanda
Keyword(s):  
Sea Water ◽  
Visible Spectrum ◽  
Hyperspectral Data ◽  
Real Time Processing ◽  
Linear Discriminant ◽  
Embedded Computer ◽  
Custom Made ◽  
Spectral Device ◽  
Short Video ◽  
Hyperspectral Sensing

An automatic custom-made procedure is developed to identify macroplastic debris loads in coastal and marine environment, through hyperspectral imaging from unmanned aerial vehicles (UAVs). Results obtained during a remote-sensing field campaign carried out in the seashore of Sassari (Sardinia, Italy) are presented. A push-broom-sensor-based spectral device, carried onboard a DJI Matrice 600 drone, was employed for the acquisition of spectral data in the range 900−1700 nm. The hyperspectral platform was realized by assembling commercial devices, whereas algorithms for mosaicking, post-flight georeferencing, and orthorectification of the acquired images were developed in-house. Generation of the hyperspectral cube was based on mosaicking visible-spectrum images acquired synchronously with the hyperspectral lines, by performing correlation-based registration and applying the same translations, rotations, and scale changes to the hyperspectral data. Plastics detection was based on statistically relevant feature selection and Linear Discriminant Analysis, trained on a manually labeled sample. The results obtained from the inspection of either the beach site or the sea water facing the beach clearly show the successful separate identification of polyethylene (PE) and polyethylene terephthalate (PET) objects through the post-processing data treatment based on the developed classifier algorithm. As a further implementation of the procedure described, direct real-time processing, by an embedded computer carried onboard the drone, permitted the immediate plastics identification (and visual inspection in synchronized images) during the UAV survey, as documented by short video sequences provided in this research paper.

scholarly journals Pavement Crack Detection from Hyperspectral Images Using a Novel Asphalt Crack Index

2020 ◽  
Vol 12 (18) ◽  
pp. 3084 ◽  
Author(s):  
Mohamed Abdellatif ◽  
Harriet Peel ◽  
Anthony G. Cohn ◽  
Raul Fuentes
Keyword(s):  
Crack Detection ◽  
Early Stage ◽  
Visible Spectrum ◽  
Hyperspectral Data ◽  
Hyperspectral Images ◽  
Surface Material ◽  
Pavement Condition ◽  
The Road ◽  
Road Pavement ◽  
Crack Index

Detection of road pavement cracks is important and needed at an early stage to repair the road and extend its lifetime for maintaining city roads. Cracks are hard to detect from images taken with visible spectrum cameras due to noise and ambiguity with background textures besides the lack of distinct features in cracks. Hyperspectral images are sensitive to surface material changes and their potential for road crack detection is explored here. The key observation is that road cracks reveal the interior material that is different from the worn surface material. A novel asphalt crack index is introduced here as an additional clue that is sensitive to the spectra in the range 450–550 nm. The crack index is computed and found to be strongly correlated with the appearance of fresh asphalt cracks. The new index is then used to differentiate cracks from road surfaces. Several experiments have been made, which confirmed that the proposed index is effective for crack detection. The recall-precision analysis showed an increase in the associated F1-score by an average of 21.37% compared to the VIS2 metric in the literature (a metric used to classify pavement condition from hyperspectral data).

The roles of filters in the photophores of oceanic animals and their relation to vision in the oceanic environment

1985 ◽  
Vol 225 (1238) ◽  
pp. 63-97 ◽  
Keyword(s):  
Sea Water ◽  
External Surface ◽  
Red Light ◽  
Visible Spectrum ◽  
Transmission Band ◽  
Spectral Absorption ◽  
Good Match ◽  
Mesopelagic Zone ◽  
Light Filters ◽  
Colour Match

In many of the photophores found in deep-sea fishes and invertebrates, light filters containing pigments lie between the tissues that generate light and the sea. The loss of light within such filters has been measured throughout the visible spectrum for a variety of animals. These filters differ greatly in their spectral absorption characteristics and do not all contain the same pigments. All those from ventral photophores have a transmission band in the blue corresponding to the daylight that penetrates best into oceanic waters. For two fishes it is shown that the light generated inside their photophores is a relatively poor spectral match for the ambient submarine daylight while the light emitted into the sea, after passing through the filters, is a good match. For a third fish a similar improvement in ‘colour match’ is brought about not by passing the light through a filter containing pigments but by reflecting the light into the sea by a blue mirror. All these observations support the hypothesis that the ventral photophores are used for camouflage. Malacosteus niger Ayres 1848 is an oceanic fish which emits red light from a large suborbital photophore. The red light generated inside the photophore is largely absorbed by a coloured filter over its external surface which transmits only a band of light of wavelengths around 700 nm. This is a waveband which is heavily absorbed by oceanic sea water. It is shown, however, that animals that can emit and are sensitive to such far-red light will have very great advantages in being able to see without being seen. The ranges over which such red light can be useful for vision are, however, relatively small. The nature of the pigments found in these various photophores is discussed. It is also calculated that the intensities of penetrating daylight are such that visual acuity could be fairly good down to considerable depths in the mesopelagic zone.

Hyperspectral Images Terrain Classification Based on Power Transformation

2012 ◽  
Vol 220-223 ◽  
pp. 2886-2890
Author(s):  
Jing Liu ◽  
Yi Liu
Keyword(s):  
Infrared Imaging ◽  
Hyperspectral Image ◽  
Recognition Rate ◽  
Bayesian Classifier ◽  
Hyperspectral Data ◽  
Hyperspectral Images ◽  
Power Transformation ◽  
Terrain Classification ◽  
Linear Discriminant ◽  
Class Distribution

In order to make the class distribution be close to Gaussian distribution, a hyperspectral images terrain classification method is presented base on power transformation (PT). Firstly PT is performed to original hyperspectral data to improve the class distribution Gaussianity, and then direct linear discriminant analysis (dLDA) is used to extract features, finally Bayesian classifier is designed in the achieved feature subspace for recognition. Experimental results based on airborne visible/infrared imaging spectrometer (AVIRIS) hyperspectral image show that, with data power transformed, the recognition rate of Bayesian classifier is dramatically improved, and the feature extraction effect of dLDA is enhanced to a certain degree.

Hyperspectral Data Analysis for Detecting Lead Pollution in Rice

2013 ◽  
Vol 433-435 ◽  
pp. 456-459
Author(s):  
Wei Hong Zhu ◽  
Cheng Zhe Xu
Keyword(s):  
Wavelet Transform ◽  
Data Analysis ◽  
Discriminant Analysis ◽  
Linear Discriminant Analysis ◽  
Hyperspectral Data ◽  
Experimental Result ◽  
Discrete Wavelet ◽  
Lead Pollution ◽  
Linear Discriminant ◽  
Reduced Data

This paper presents a new method for detecting lead pollution in rice by analyzing hyperspectral data. First, preprocessing method is used to remove the outliers which deviate so much from other hyperspectral data. Then, dimensionality-reduced data are made by using discrete wavelet transform. Finally, linear discriminant analysis is utilized to extract the feature which characterizes polluted and unpolluted rice. The experimental result based on the proposed method shows the good performance in detecting lead pollution in rice.

scholarly journals Early Identification of Root Rot Disease by Using Hyperspectral Reflectance: The Case of Pathosystem Grapevine/Armillaria

2021 ◽  
Author(s):  
Federico Calamita ◽  
Hafiz Ali Imran ◽  
Loris Vescovo ◽  
Mohamed Lamine Mekhalfi ◽  
Nicola La Porta
Keyword(s):  
Root Rot ◽  
Woody Plants ◽  
Near Infrared ◽  
Plant Defence ◽  
Disease Diagnosis ◽  
Hyperspectral Data ◽  
Learning Context ◽  
Linear Discriminant ◽  
Root Rot Disease ◽  
Rot Disease

The Armillaria genus represents one of the most common causes of chronic root rot disease in woody plants. The disease damage prompt assessment is crucial for pest management. However, the disease detection current methods are limited at the field scale. Therefore, an alternative approach that can enhance or supplement traditional techniques is needed. In this study, we investigated the potential of hyperspectral methods to identify the changes between fungi-infected and uninfected plants of Vitis vinifera in early detecting the Armillaria disease. The hyperspectral imaging sensor Specim-IQ was used to acquire images of leaves of the Teroldego Rotaliano grapevine cultivar. We analysed three groups of plants: healthy, asymptomatic, and diseased. Highly significant differences were found in the Near infrared (NIR) spectral region with a decreasing pattern from healthy to diseased plants attributable to internal leaf structure changes. Asymptomatic plants emerged from the other groups due to a smaller reflectance in the red-edge spectrum (around 705nm). Hypothetically associated with the presence of secondary metabolites involved in plant defence strategy. Furthermore, significant differences were observed in the wavelengths close to 550 nm in diseased plants versus asymptomatic. We used linear discriminant analysis from a machine learning context to classify the leaves based on the most significant variables (vegetation indices and single bands), with resulting overall accuracies of 85% and 84% respectively in healthy vs. diseased and healthy vs. asymptomatic. To our knowledge, this study represents the first report on the possibility of using hyperspectral data for root rot disease diagnosis on woody plants. Although further validation studies are required, it appears that the spectral reflectance technique, possibly implemented on unmanned aerial vehicles (UAV), could be a promising tool for a cost-effective, non-destructive method of Armillaria disease early diagnosis and mapping in the field, contributing to a significant step forward in precision viticulture.

scholarly journals AIRBORNE LASER BATHYMETRY FOR DOCUMENTATION OF SUBMERGED ARCHAEOLOGICAL SITES IN SHALLOW WATER

2015 ◽  
Vol XL-5/W5 ◽  
pp. 99-107 ◽  
Author(s):  
M. Doneus ◽  
I. Miholjek ◽  
G. Mandlburger ◽  
N. Doneus ◽  
G. Verhoeven ◽  
...  
Keyword(s):  
Shallow Water ◽  
Large Scale ◽  
Sea Water ◽  
Laser Scanner ◽  
Visible Spectrum ◽  
Laser Pulses ◽  
Green Laser ◽  
Archaeological Sites ◽  
Underwater Topography ◽  
Airborne Laser

Knowledge of underwater topography is essential to the understanding of the organisation and distribution of archaeological sites along and in water bodies. Special attention has to be paid to intertidal and inshore zones where, due to sea-level rise, coastlines have changed and many former coastal sites are now submerged in shallow water. Mapping the detailed inshore topography is therefore important to reconstruct former coastlines, identify sunken archaeological structures and locate potential former harbour sites. However, until recently archaeology has lacked suitable methods to provide the required topographical data of shallow underwater bodies. Our research shows that airborne topo-bathymetric laser scanner systems are able to measure surfaces above and below the water table over large areas in high detail using very short and narrow green laser pulses, even revealing sunken archaeological structures in shallow water. Using an airborne laser scanner operating at a wavelength in the green visible spectrum (532 nm) two case study areas in different environmental settings (Kolone, Croatia, with clear sea water; Lake Keutschach, Austria, with turbid water) were scanned. In both cases, a digital model of the underwater topography with a planimetric resolution of a few decimeters was measured. While in the clear waters of Kolone penetration depth was up to 11 meters, turbid Lake Keutschach allowed only to document the upper 1.6 meters of its underwater topography. Our results demonstrate the potential of this technique to map submerged archaeological structures over large areas in high detail providing the possibility for systematic, large scale archaeological investigation of this environment.

scholarly journals Hyperspectral dimension reduction using global and local information based linear discriminant analysis

2014 ◽  
Vol II-7 ◽  
pp. 61-66 ◽  
Author(s):  
U. Sakarya
Keyword(s):  
Dimension Reduction ◽  
Image Classification ◽  
Hyperspectral Image ◽  
Hyperspectral Data ◽  
Hyperspectral Images ◽  
Hyperspectral Image Classification ◽  
Global Pattern ◽  
Linear Discriminant ◽  
Local Pattern ◽  
Pattern Information

Hyperspectral image classification has become an important research topic in remote sensing. Because of high dimensional data, a special attention is needed dealing with spectral data; and thus, one of the research topics in hyperspectral image classification is dimension reduction. In this paper, a dimension reduction approach is presented for classification on hyperspectral images. Advantages of the usage of not only global pattern information, but also local pattern information are examined in hyperspectral image processing. In addition, not only tuning the parameters, but also an experimental analysis of the distribution of the hyperspectral data is demonstrated. Therefore, how global or local pattern variations play an important role in classification is examined. According to the experimental outcomes, the promising results are obtained for classification on hyperspectral images.

Sign in / Sign up

Export Citation Format

Share Document