
This will count as one of your downloads.
You will have access to both the presentation and article (if available).
The study was carried out in Cesis Municipality in Latvia using airborne flying laboratory ARSENAL – the constellation of hyperspectral imagers in the visible to mid-wave infrared (400-5000 nm) spectral range, topographic LiDAR and high-resolution RGB camera for simultaneous multi-sensor data acquisition. Hyperspectral data were used for both mapping of grasslands and assessment of grass biomass. Different spectral ranges and machine learning algorithms were tested in order to find the best one. The performance of Sentinel-2 like spectral bands also was tested for further possible further use of multispectral satellite data.
In this study, two simultaneous hyperspectral airborne data and in situ measurement campaigns were performed in the Gulf of Riga near the River Daugava mouth in summer 2015 to simulate Sentinel-3 data and test existing algorithms for retrieval of Level 2 Water products. Comparison of historical data showed poor overall correlation between in situ measurements and MERIS chlorophyll-a data products. Better correlation between spectral chl-a data products and in situ water sampling measurements was achieved during simultaneous airborne and field campaign resulting in R2 up to 0.94 for field spectral data, R2 of 0.78 for airborne data. Test of all two band ratio combinations showed that R2 could be improved from 0.63 to 0.94 for hyperspectral airborne data choosing 712 and 728 nm bands instead of 709 and 666 nm, and R2 could be improved from 0.61 to 0.83 for simulated Sentinel-3 OLCI data choosing Oa10 and Oa8 bands instead of Oa11 and Oa8.
Repeated campaigns are planned during spring and summer blooms 2016 in the Gulf of Riga to get larger data set for validation and evaluate repeatability. The main challenges remain to acquire as good data as possible within rapidly changing environment and often cloudy weather conditions.
In this study, land cover mapping scheme according to national end-user needs was developed and tested in two pilot territories (Cesis and Burtnieki). Hyperspectral airborne data covering spectral range 400-2500 nm was acquired in summer 2015 using Airborne Surveillance and Environmental Monitoring System (ARSENAL). The gathered data was tested for land cover classification of seven general classes (urban/artificial, bare, forest, shrubland, agricultural/grassland, wetlands, water) and sub-classes specific for Latvia as well as simulation of Sentinel-2 satellite data. Hyperspectral data sets consist of 122 spectral bands in visible to near infrared spectral range (356-950 nm) and 100 bands in short wave infrared (950-2500 nm). Classification of land cover was tested separately for each sensor data and fused cross-sensor data. The best overall classification accuracy 84.2% and satisfactory classification accuracy (more than 80%) for 9 of 13 classes was obtained using Support Vector Machine (SVM) classifier with 109 band hyperspectral data. Grassland and agriculture land demonstrated lowest classification accuracy in pixel based approach, but result significantly improved by looking at agriculture polygons registered in Rural Support Service data as objects. The test of simulated Sentinel-2 bands for land cover mapping using SVM classifier showed 82.8% overall accuracy and satisfactory separation of 7 classes. SVM provided highest overall accuracy 84.2% in comparison to 75.9% for k-Nearest Neighbor and 79.2% Linear Discriminant Analysis classifiers.

The multispectral imaging system Nuance based on liquid crystal tunable filters was adapted for diffuse reflectance and fluorescence spectral imaging of in vivo skin. Uniform illumination was achieved by LED ring light. Combination of four LEDs (warm white, 770 nm, 830 nm and 890 nm) was used to support diffuse reflectance mode in spectral range 450-950 nm. 405 nm LEDs were used for excitation of skin autofluorescence. Multispectral imaging system was adapted for spectral working range of 450-950 nm with scanning step of 10 nm and spectral resolution of 15 nm. An average field of view was 50x35 mm in size with spatial resolution 0,05 mm (the pixel size). Due to spectrally different illumination intensity and system sensitivity, various exposure times (from 7…500 ms) were used for each image acquisition.
The proposed approach was tested for different skin lesions: benign nevus, hemangioma, basalioma and halo nevus. Spectral image cubes of different skin lesions were acquired and analyzed to test its diagnostic potential.
In this work we demonstrate simple implementation of LASCA imaging technique as connection kit for mobile phone for primary low-cost assessment of skin blood flow. Stabilized 650 nm and 532 nm laser diode modules were used for LASCA illumination. Dual wavelength illumination could provide additional information about skin hemoglobin and oxygenation level.
The proposed approach was tested for arterial occlusion and heat test. Besides, blood flow maps of injured and provoked skin were demonstrated.


View contact details