Principle Component Analysis for Crop Discrimination using Hyperspectral Remote Sensing Data

Crop discrimination is still very challenging issue for researcher because of spectral reflectance similarity captured in non-imaging data. The objective of this research work is to focus on crop discrimination challenge. We have used ASD FieldSpec4 Spectroradiometer for collection of leaf samples of four crops Wheat, Jowar, Bajara and Maize. We used vegetation indices and some spectral reflectance band for featuring our dataset. We applied Principle Component Analysis (PCA) for discrimination and it has been observed that when we use first and second principle component, it will give poor result but if third principle component is used then we get accurate and fine results.

Crop Classification Using MSCDN Classifier and Sparse Auto-Encoders with Non-Negativity Constraints for Multi-Temporal, Quad-Pol SAR Data

Accurate and reliable crop classification information is a significant data source for agricultural monitoring and food security evaluation research. It is well-known that polarimetric synthetic aperture radar (PolSAR) data provides ample information for crop classification. Moreover, multi-temporal PolSAR data can further increase classification accuracies since the crops show different external forms as they grow up. In this paper, we distinguish the crop types with multi-temporal PolSAR data. First, due to the “dimension disaster” of multi-temporal PolSAR data caused by excessive scattering parameters, a neural network of sparse auto-encoder with non-negativity constraint (NC-SAE) was employed to compress the data, yielding efficient features for accurate classification. Second, a novel crop discrimination network with multi-scale features (MSCDN) was constructed to improve the classification performance, which is proved to be superior to the popular classifiers of convolutional neural networks (CNN) and support vector machine (SVM). The performances of the proposed method were evaluated and compared with the traditional methods by using simulated Sentinel-1 data provided by European Space Agency (ESA). For the final classification results of the proposed method, its overall accuracy and kappa coefficient reaches 99.33% and 99.19%, respectively, which were almost 5% and 6% higher than the CNN method. The classification results indicate that the proposed methodology is promising for practical use in agricultural applications.

Crop Discrimination using Non-Imaging Hyperspectral Data

Crop type discrimination is still very challenging task for researchers using non-imaging hyperspectral data. It is because of spectral reflectance similarity between crops. In this research work we have discriminated between four crops wheat, jowar, bajara and maize. We have tried to overcome the problems which have been faced my researchers. Initially by visual analysis we have selected 22 reflectance band which shows the absorption property of particular molecules and classification technique is applied, but it has given us very poor result of classification. We observed only 24% classification accuracy. So we considered nine vegetation indices along with spectral bands and achieved better classification accuracy. ASD FieldSpec 4 Spectroradiometer device is used for capturing spectral reflectance data. We calculated nine different vegetation indices and some selective reflectance bands are used for crop classification. We have used Support Vector Machine (SVM) for classification.

EVALUATION OF THE PERFORMANCE OF SAR AND SAR-OPTICAL FUSED DATASET FOR CROP DISCRIMINATION

<p><strong>Abstract.</strong> Crop discrimination and acreage play a vital role in interpreting the cropping pattern, statistics of the produce and market value of each product. Sultan Battery is an area where a large amount of irrigated and rainfed paddy crops are grown along with Rubber, Arecanut and Coconut. In addition, the northern region of Sultan Battery is covered with evergreen and deciduous forest. In this study, the main objective is to evaluate the performance of optical and Synthetic Aperture Radar (SAR)-optical hybrid fusion imageries for crop discrimination in Sultan Bathery Taluk of Wayanad district in Kerala. Seven land use classes such as paddy, rubber, coconut, deciduous forest, evergreen forest, water bodies and others land use (e.g., built-up, barren etc.) were selected based on literature review and local land use classification policy. Both Sentinel-2A (optical) and sentinel-1A (SAR) satellite imageries of 2017 for Kharif season were used for classification using three machine learning classifiers such as Support Vector Machine (SVM), Random Forest (RF) and Classification and Regression Trees (CART). Further, the performance of these techniques was also compared in order to select the best classifier. In addition, spectral indices and textural matrices (NDVI, GLCM) were extracted from the image and best features were selected using the sequential feature selection approach. Thus, 10-fold cross-validation was employed for parameter tuning of such classifiers to select best hyperparameters to improve the classification accuracy. Finally, best features, best hyperparameters were used for final classification and accuracy assessment. The results show that SVM outperforms the RF and CART and similarly, Optical+SAR datasets outperforms the optical and SAR satellite imageries. This study is very supportive for the earth observation scientists to support promising guideline to the agricultural scientist, policy-makers and local government for sustainable agriculture practice.</p>

DISCRIMINATION OF SUGARCANE CROP AND CANE YIELD ESTIMATION USING LANDSAT AND IRS RESOURCESAT SATELLITE DATA

<p><strong>Abstract.</strong> The objective of this research work aims at crop acreage estimation at mill catchment level, derivation of sugarcane phenology and yield estimation at field level. The study was carried out in Kisan Sahkari Chini Mill catchment, Nanauta, Saharanpur, Uttar Pradesh. Extensive and systematic field sampling was carried out for ground-truth observations, biophysical measurements (LAI and above/below canopy PAR) and mill-able cane yield through crop cutting experiments. Major emphasis were laid on sugarcane crop discrimination, biophysical parameter estimation, generation of phenological metrics and yield model development for sugarcane crop at mill catchment level. Sugarcane crop discrimination and its acreage estimation was done using multi-sensor satellite data. The sugarcane classification accuracies were &amp;gt;&amp;thinsp;92% for LISS-IV, &amp;gt;&amp;thinsp;86% for Landsat-8 and &amp;gt;&amp;thinsp;83% for LISS-III classified image. The sugarcane phenological matrices at field level derived using time-series of NDVI for a period of 2015&amp;ndash;2016 through TIMESAT software. To retrieve the biophysical parameters particularly leaf area index, best predictive function developed with vegetation indices (EVI, NDVI, SAVI) through correlation and regression analysis along this cane yield estimation attempted with multi-date (eight-day) NDVI from Landsat OLI. Yield models developed for ratoon cane and planted cane explained variance in yield significantly with coefficient of determination (R²) values equal to 0.83 and 0.69, respectively. Similar predictive functions were also established with monthly composite dataset for village-level yield estimates with step wise regression (R²&amp;thinsp;=&amp;thinsp;0.83) (P&amp;thinsp;=&amp;thinsp;0.00001), Multi linear regression (MLR) (R²&amp;thinsp;=&amp;thinsp;0.792) (P&amp;thinsp;=&amp;thinsp;0.00081) and Random forest regression (R²&amp;thinsp;=&amp;thinsp;0.466) (P&amp;thinsp;=&amp;thinsp;0.038).</p>