No abstract

Multiple Kernel Learning (MKL) algorithms are among the most successful classification methods for hyperspectral data. Nevertheless, these algorithms suffer from two main drawbacks of computational complexity and debility to admit to the end-to-end learning paradigm. This paper proposed a Convolutional Kernel Classifier (CKC) for hyperspectral remote sensing images to address these issues. The CKC uses the Nyström approximation method to estimate a low-rank approximation of the basis kernels, thus solves the issues associated with the high dimensionality of the basis kernels. The CKC uses deep architecture to learn the optimal combination of the basis kernels and the classification task to enable end-to-end learning. The proposed CKC's architecture is based on a 1D-Convolutional Neural Network (CNN-1D), and it uses kernel dropout to prevent overfitting. It is the first instance of deep-kernel algorithms in the field of remote sensing. The proposed method was compared with several well-known hyperspectral image analysis MKL algorithms, including a multi-kernel variant of the deep kernel machine optimization (M-DKMO), MKL-average, Simple-MKL, and Generalize MKL (GMKL), and state-of-the-art deep learning models, including Vanilla Recurrent Neural Network (VanillaRNN) and CNN-1D in classifying four benchmark hyperspectral datasets. The experimental results show that the CKC consistently outperforms all the competitor methods, and its runtime is lower than its MKL algorithm counterparts on four benchmark hyperspectral datasets. Moreover, the Nyström approximation solves the high dimensionality of the basis kernels and boosts classification accuracy. The source codes of CKC are available from: https://github.com/MohsenAnsari1373/A-New-Convolutional-Kernel-Classifier-for-Hyperspectral-Image-Classification .

The present study was performed for the period of one year from January 2013 to December 2013 in order to understand the physico-chemical properties of Mahul Creek water. From the results of our study it was observed that the annual average COD level was 362.09 ppm which was far higher than the maximum tolerable level of 250 ppm set for inland surface water as well as for marine coastal water. The annual average conductivity was found to be 6122.81 µS cm -1 which was very much above the conductivity limit for inland surface water of 1000 µS cm -1 set by Central Pollution Control Board (CPCB) for propagation of fisheries. The annual average total alkalinity level was recorded as 166.25 ppm, which according to the UN Department of Technical Cooperation for Development (1985) was found to be strongly alkaline. The annual average hardness level of the creek water was found to be 2488.65 ppm which was above the limit of 300 ppm set by ISI. From the results of the present investigation it seems that the time has come to implement proper effluent water treatment techniques and enforcement of pollution control by the regulatory authority on the indiscriminate discharge of industrial wastewater into water bodies.

The present study was performed for the period of one year from January 2013 to December 2013 in order to understand the level of toxic heavy metals in the sediments of Mahul Creek near Mumbai. The annual average concentration of heavy metals like Cr, Zn, Cu, Ni, Pb, Cd, As and Hg was found to be 277. 5, 121.7, 100.3, 63.8, 21.5, 14.6, 10.4 and 4.9 ppm respectively. It is feared that this heavy metals accumulated in the creek sediments may enter the water thereby creating threat to the biological life of an aquatic ecosystem. The results of present study indicates that the existing situation if mishandled can cause irreparable ecological harm in the long term well masked by short term economic prosperity due to extensive industrial growth.

Abstract. Water salinity is a complex issue in coastal and estuarine areas. Currently, remote sensing techniques have been widely used to monitor water quality changes, ranging from river to oceans. The salinity of Karun River has been increasing due to some critical factors, therefore, This study aimed at building regression models to ascertain the water salinity through the relationship between the reflectance of the Landsat-8 OLI and In situ measurements. A total of 102 observed samples were divided into 70% training and 30% test from June 2013 to July 2018 along the Karun River. Spectral signature analysis showed that band 1 - Coastal/Aerosol (0.433–0.453 μm), band 2 - Blue (0.450–0.515 μm) and band 3 - Green (0.525–0.600 μm) are sensitive to salinity . Furthermore, to have a comprehensive investigation, the Support Vector Regression (SVR) method was applied. The outcomes related to the quality of the SVR depend on several factors e.g. proper setting of the SVR meta-parameters, therefore, to deal with this issue Genetic Algorithm (GA) was applied. The SVR model resulted in values of R2 and RMSE for test data which are respectively obtained to be 0.7 and 390 μs cm−1. Eventually, Karun water salinity maps were prepared by SVR method to demonstrate the Karun water salinity on 1 February 2015 and 5 September 2018.